pism

[fork] customized build of PISM, the parallel ice sheet model (tillflux branch)
git clone git://src.adamsgaard.dk/pism # fast
git clone https://src.adamsgaard.dk/pism.git # slow
Log | Files | Refs | README | LICENSE Back to index
DateCommit messageAuthorFiles+-
2021-08-05 00:01Fix antspin-coarse.shKyle Blum1+1-1
2021-06-14 00:39Add .deepsource.tomlDeepSource Bot1+17-0
2020-08-28 20:22Update PISM version to 1.2.2Constantine Khrulev2+3-3
2020-08-28 19:59Merge branch 'v1.2-bug-fixes'Constantine Khrulev132+349-350
2020-03-20 04:48Update PISM version to 1.2.1Constantine Khrulev2+3-3
2020-03-20 04:47Merge branch 'v1.2-bug-fixes'Constantine Khrulev45+220-1039
2020-02-13 22:18Fix funding acknowledgmentsConstantine Khrulev3+3-3
2020-02-13 19:04Update the CircleCI banner in the READMEConstantine Khrulev1+2-2
2020-02-12 05:24Update the list of configuration parametersConstantine Khrulev1+5-20
2020-02-12 05:23Clean up pism_config.cdlConstantine Khrulev1+269-287
2020-02-12 04:47Merge branch 'v1.2-pre-release'Constantine Khrulev591+32362-16938
2019-12-17 14:29Added missing build fileElizabeth Fischer1+19-0
2019-06-17 18:14Update the change logConstantine Khrulev1+9-0
2019-05-22 21:33Support PROJ.4 version 6.x (see #409).Constantine Khrulev1+2-1
2019-05-03 18:43Fix a bug in the Routing hydrology modelConstantine Khrulev1+1-1
2019-03-14 17:48Fix a bug in the diagnostic computation of the ice fluxConstantine Khrulev2+35-29
2018-12-06 18:06Fix Python 3 compatibility in conf.py and update the copyright yearConstantine Khrulev2+12-7
2019-03-08 20:32Create CODE_OF_CONDUCT.mdEd Bueler1+76-0
2019-02-27 00:38Update the change log in preparation for tagging v1.1.2Constantine Khrulev1+9-5
2019-02-01 00:02Update the change logConstantine Khrulev1+3-0
2019-01-31 23:57Use BedrockColumn in BTU_FullConstantine Khrulev6+27-100
2019-01-31 20:54Add a solver for the heat equation in a bedrock columnConstantine Khrulev5+178-5
2019-02-26 23:16Update the change logConstantine Khrulev1+5-0
2019-02-04 14:07Modified PIK temperature parameterizationTorsten Albrecht1+4-3
2019-02-26 21:54Fix units in ForceThickness::max_timestep_impl()Constantine Khrulev1+2-2
2019-02-20 23:05Fix "no model mask" treatment in "pismr -regional"Constantine Khrulev4+62-8
2019-02-20 23:09Update the change logConstantine Khrulev1+6-1
2019-01-17 20:08Use VERSION_LESS to support CMake < 3.7Constantine Khrulev1+1-1
2018-12-13 18:32Support CMake versions < 3.13.Constantine Khrulev1+5-1
2018-12-07 23:35Merge branch 'ckhroulev/v1.1rc'Constantine Khrulev657+24367-20847
2018-11-29 00:17PISM builds with GCC 8. Fixes #420.Constantine Khrulev3+8-7
2018-11-29 00:37Fix clang 7.0 compiler warningsConstantine Khrulev1+25-14
2017-12-19 10:01Fixes bug when using -subgl (-pik) and delta_SLTorsten Albrecht4+8-10
2018-03-01 05:25Use FFTW_ESTIMATE instead of FFTW_MEASURE. See #402.Constantine Khrulev1+6-9
2018-02-14 20:03Update the change log.Constantine Khrulev1+11-0
2018-02-13 19:27Fix #400.Constantine Khrulev1+5-1
2018-02-12 21:54IceRegionalModel: fix initialization of stored surface elevation.Constantine Khrulev1+23-15
2018-02-12 20:13Move the computation of omega to avoid errors with PISM_DEBUG=ON.Constantine Khrulev1+3-2
2018-02-12 20:12Avoid filling in temperature using heuristics when it is filled by regridding.Constantine Khrulev2+30-16
2018-02-12 20:11Add more sanity checks to bootstrap_ice_temperature.Constantine Khrulev1+28-3
2017-11-21 23:02Update diagnostics and configuration parameters in the manual.Constantine Khrulev2+18-13
2017-11-17 05:56Fix #222.Constantine Khrulev1+39-36
2017-11-16 22:24Don't need to check this: area_specific_volume is always availableConstantine Khrulev1+0-5
2017-11-16 20:43Add geometry.part_grid.max_iterations and increase it to 10.Constantine Khrulev2+15-8
2017-11-16 01:15Fix the tendency_of_ice_mass_due_to_flow scalar diagnostic.Constantine Khrulev1+6-5
2017-11-15 20:30Make it easier to compare scalar and 2D diagnostics.Constantine Khrulev1+22-22
2017-11-15 02:28Update .gitignore.Constantine Khrulev1+3-2
2017-11-15 02:24Improve the "balancing the books" section in the manual.Constantine Khrulev4+73-43
2017-11-14 23:07Make sure all "iceberg removing" contributes to computed ice discharge.Constantine Khrulev4+63-60
2017-11-14 22:17Cleanup.Constantine Khrulev4+10-11
2017-11-14 18:54Remove some outdated comments.Constantine Khrulev1+1-5
2017-11-09 04:46Fix a typo.Constantine Khrulev1+1-1
2017-11-09 04:45SWIG_GET_WRAPPER_DEPENDENCIES is broken. Time to remove it.Constantine Khrulev1+1-58
2017-11-01 20:31Add a work-around needed to use old-ish NetCDF (4.0 - 4.1) with OpenMPI.Constantine Khrulev2+11-1
2017-10-31 18:27Support older (< 1.7) OpenMPI versions.Constantine Khrulev1+7-0
2017-10-30 17:16Fix a regression test.Constantine Khrulev1+2-2
2017-10-26 23:08Trying to sort out function inlining issues.Constantine Khrulev8+205-251
2017-10-26 21:59Don't warn about unknown #pragma.Constantine Khrulev1+1-1
2017-10-26 21:51fixup! Tell Intel compiler that it's OK to inline these.Constantine Khrulev1+1-1
2017-10-26 21:39Tell Intel compiler that it's OK to inline these.Constantine Khrulev2+13-13
2017-10-26 21:39Tell Intel compiler to ignore assumed dependencies in these loops.Constantine Khrulev1+3-0
2017-10-26 18:13Create a link from build/pism->src instead of build/include/pism->src.Constantine Khrulev1+6-9
2017-10-26 07:24Fix a typo.Constantine Khrulev1+2-2
2017-10-20 18:37Add Git revision and date to HTML page footers and the PDF manual.Constantine Khrulev4+49-1
2017-10-20 17:43Fix a bunch of FIXMEs in the Manual.Constantine Khrulev8+76-33
2017-10-20 16:18Replace "pressure imbalance" with "pressure difference."Constantine Khrulev2+16-8
2017-10-20 04:14complement to Constantine: code got shorterEd Bueler1+3-3
2017-10-20 01:08Minor fixes.Constantine Khrulev2+3-3
2017-10-20 00:11Mark this as "stable" and "1.0".Constantine Khrulev3+6-6
2017-10-20 00:10Update the release checklist.Constantine Khrulev1+21-19
2017-10-20 00:09A whitespace change.Constantine Khrulev1+1-2
2017-10-20 00:07Add a Makefile and update a Makefile.Constantine Khrulev2+5-44
2017-10-19 22:57Improve the change log a little bit.Constantine Khrulev1+18-12
2017-10-19 22:29"Coding guidelines" is a better name.Constantine Khrulev3+407-395
2017-10-19 21:22Use funding.csv to generate lists of grants supporting PISM development.Constantine Khrulev6+192-40
2017-10-19 20:03Trying to fix the CSV table...Constantine Khrulev1+3-3
2017-10-19 20:00Add a CSV file listing grants that supported PISM development.Constantine Khrulev1+9-0
2017-10-19 18:59Replace a bunch of .md with .rst (pandoc makes this pretty fast).Constantine Khrulev11+821-702
2017-10-19 17:49Add a summary to CHANGES.rst.Constantine Khrulev1+13-4
2017-10-19 17:28Update the list of configuration parameters.Constantine Khrulev1+5-50
2017-10-19 17:26Remove -pdd_aschwanden.Constantine Khrulev5+1-141
2017-10-19 17:23Remove -ocean runoff_SMB,...Constantine Khrulev5+0-191
2017-10-19 16:06Add bed_deformation.lc.grid_size_factor. See #395.Constantine Khrulev2+6-1
2017-10-19 16:06Remove "PB" from names of bed deformation models.Constantine Khrulev9+37-37
2017-10-19 05:31Update a spinup script.Constantine Khrulev1+2-2
2017-10-19 05:23Fix the units of the basal_mass_flux_{grounded,floating} diagnostics.Constantine Khrulev1+8-2
2017-10-19 01:56Update the #74 test.Constantine Khrulev1+1-1
2017-10-19 00:58Mention that #390 is fixed.Constantine Khrulev1+17-16
2017-10-19 00:18Move cubature.c to src/external.Constantine Khrulev7+17-6
2017-10-18 21:30Remove some file name prefixes.Constantine Khrulev413+24498-24498
2017-10-18 20:55Fix #166.Constantine Khrulev329+1469-1440
2017-10-18 18:15Move code out of src/base into src/.Constantine Khrulev760+64079-64079
2017-10-18 17:45More directory structure improvements.Constantine Khrulev31+216-234
2017-10-18 17:37Little tweaks in the Manual.Constantine Khrulev1+12-10
2017-10-18 17:37Improve the directory structure.Constantine Khrulev110+10408-10399
2017-10-18 05:08Add links to GitHub issues plus minor edits.Constantine Khrulev1+59-28
2017-10-18 02:43Clean up the change log.Constantine Khrulev1+8-6
2017-10-18 02:35Mention "glacierized" time-series.Constantine Khrulev1+4-0
2017-10-18 02:35Add a couple of words about -topg_delta.Constantine Khrulev1+11-0
2017-10-18 02:05Document -extra_vars shortcuts "mass_fluxes" and "amount_fluxes."Constantine Khrulev1+14-0
2017-10-18 01:17Multiplicative factors are not offsets.Constantine Khrulev1+1-2
2017-10-18 01:17Add the updated change log.Constantine Khrulev6+978-618
2017-10-17 17:13Update the list of diagnostics in the manual.Constantine Khrulev1+89-94
2017-10-17 17:07Rename some rate of change time series.Constantine Khrulev1+20-20
2017-10-17 16:57Rename scalar diagnosticsConstantine Khrulev16+577-656
2017-10-17 00:05Don't use Jansson (for now).Constantine Khrulev2+6-3
2017-10-16 23:51Add two FIXMEs.Constantine Khrulev1+5-1
2017-10-16 23:45Add a link to the Installation Manual to README.rst.Constantine Khrulev1+2-6
2017-10-16 19:57Update std-greenland run costs.Constantine Khrulev2+2-2
2017-10-16 19:55Finish updating the section about re-building docs.Constantine Khrulev2+69-76
2017-10-16 19:44Make sure pismpython docs can still be built.Constantine Khrulev6+46-49
2017-10-16 19:16Whitespace changes.Constantine Khrulev1+4-4
2017-10-16 19:15Add an empty note to silence pybtex's error.Constantine Khrulev1+1-0
2017-10-16 18:07Improve CMake scripts building Sphinx docs.Constantine Khrulev1+28-9
2017-10-16 18:06Replace Pism_BUILD_PDFS with Pism_BUILD_DOCS.Constantine Khrulev2+4-12
2017-10-16 18:05pism_config.md is not a part of the browser any more.Constantine Khrulev1+1-1
2017-10-16 17:44Update a file name.Constantine Khrulev1+1-1
2017-10-13 18:51I think I'll still call it 0.8 for now.Constantine Khrulev1+2-2
2017-10-13 18:45Remove Markdown INSTALL and README.Constantine Khrulev4+8-971
2017-10-13 18:27Clean up the doc/ directoryConstantine Khrulev226+55-41498
2017-10-13 06:03Adjust column widths in a table.Constantine Khrulev1+1-0
2017-10-13 05:51More work on documentation.Constantine Khrulev8+78-91
2017-10-12 22:47Remove unneeded includes.Constantine Khrulev1+1-5
2017-10-12 22:47Remove some old crud.Constantine Khrulev14+2-1491
2017-10-12 22:10Better Makefile in docs/sphinx.Constantine Khrulev1+5-4
2017-10-12 21:43Document PISM's use of PROJ.4.Constantine Khrulev1+67-0
2017-10-12 21:43Don't report projection parameters redundantly.Constantine Khrulev1+2-4
2017-10-12 20:39Write PROJ.4 parameters to mapping:proj4_params (for CDO).Constantine Khrulev1+9-2
2017-10-12 18:57Remove the "quilt" -o_format.Constantine Khrulev17+12-1177
2017-10-12 18:34Now that we require PETSc 3.5 we always have TAO.Constantine Khrulev9+19-93
2017-10-12 18:22Stop trying to support PETSc < 3.5.0.Constantine Khrulev11+19-154
2017-10-12 18:15Fix a regression test.Constantine Khrulev1+6-14
2017-10-12 18:14Wrap long lines.Constantine Khrulev1+4-2
2017-10-12 18:14Support PETSc 3.8.Constantine Khrulev1+4-0
2017-10-12 05:37Improve documentation.Constantine Khrulev14+2676-29
2017-10-11 16:42Rename grid registration figures.Constantine Khrulev4+0-0
2017-10-11 15:30Grid registration is passed to IceGrid at initializationConstantine Khrulev35+96-114
2017-10-11 05:59Mention that we need CMake >= 3.1.Constantine Khrulev1+3-2
2017-10-11 05:47Actually use (inferred) grid registrationConstantine Khrulev4+21-14
2017-10-11 05:46Remove "NONE" from the enum Periodicity.Constantine Khrulev6+7-7
2017-10-11 05:21More work on grid registration issues.Constantine Khrulev5+45-22
2017-10-11 04:27Start work on cleaning up and documenting grid registration issues.Constantine Khrulev10+766-2
2017-10-11 03:31Put the minimum PETSc version in global.txt.Constantine Khrulev3+12-7
2017-10-10 22:07Improve an error message.Constantine Khrulev3+9-2
2017-10-10 21:44Fix a typo.Constantine Khrulev1+2-2
2017-10-10 21:42Fix a link.Constantine Khrulev1+1-1
2017-10-10 21:42Move an .. include:: up a bit.Constantine Khrulev1+2-2
2017-10-10 17:00Add a stub of a section about grid registration.Constantine Khrulev1+9-3
2017-10-10 16:59Add a stub of a section about mass conservation.Constantine Khrulev2+10-0
2017-10-10 16:58Require CMake 3.1; fix a deprecation warning (see #393).Constantine Khrulev2+7-15
2017-10-10 00:11Document e-age coupling.Constantine Khrulev1+14-2
2017-10-10 00:10Fix a Sphinx warning.Constantine Khrulev1+1-1
2017-10-09 23:50Fix a typo.Constantine Khrulev1+1-1
2017-10-09 23:48Rename two sections.Constantine Khrulev2+4-4
2017-10-09 23:45Mention that -gradient affects the SIA model (mostly).Constantine Khrulev1+25-18
2017-10-09 23:45Move the section about disabling sub-models.Constantine Khrulev4+27-27
2017-10-09 23:28More work on the manual.Constantine Khrulev7+100-96
2017-10-09 23:17Add some text to modeling-choices/index.rst.Constantine Khrulev1+22-3
2017-10-09 23:01Move the list of parameters.Constantine Khrulev8+1874-1893
2017-10-09 22:47Add intros to "subglacier" and "practical usage."Constantine Khrulev2+12-0
2017-10-09 22:47Move the list of diagnostics.Constantine Khrulev7+1052-1091
2017-10-09 22:13Add a comment about -Mx and -My.Constantine Khrulev1+11-1
2017-10-09 21:53Update performance stats.Constantine Khrulev1+1-1
2017-10-09 21:38Update the parameter list.Constantine Khrulev1+17-5
2017-10-09 21:32Update C++ class names in the climate forcing manual.Constantine Khrulev3+31-31
2017-10-09 20:19Fix a heading.Constantine Khrulev1+2-2
2017-10-09 20:17Add -regional.zero_gradient to spinup.shConstantine Khrulev1+1-0
2017-10-09 20:17Remove an old FIXME.Constantine Khrulev1+0-1
2017-10-09 20:17Overwrite attributes instead of appending.Constantine Khrulev1+2-2
2017-10-09 19:27Minor cleanup.Constantine Khrulev1+1-3
2017-10-09 18:47Fix diagnostics diffusivity_staggered, h_x, h_y.Constantine Khrulev1+16-3
2017-10-09 18:19Rename StressBalanceInputs and EnergyModelInputs.Constantine Khrulev39+215-160
2017-10-09 16:24Create the class YieldStressInputs.Constantine Khrulev9+46-24
2017-10-09 16:01SIAFD: provide "inputs" to surface gradient computation methods.Constantine Khrulev4+27-25
2017-10-07 00:39Fix the regional basal yield stress model.Constantine Khrulev5+35-24
2017-10-07 00:36Add regional inputs to StressBalanceInputs.Constantine Khrulev6+41-14
2017-10-07 00:11Minor cleanup: remove "virtual" and break a long line.Constantine Khrulev2+7-6
2017-10-07 00:10Tell RegionalGeometryEvolution about the no_model_mask.Constantine Khrulev2+10-0
2017-10-07 00:09Add set_no_model_mask() to the GeometryEvolution API.Constantine Khrulev2+18-3
2017-10-06 23:32Add stressbalance::create().Constantine Khrulev9+105-101
2017-10-06 20:27Fix preprocess.sh: the spun-up state we have uses an old variable name.Constantine Khrulev1+3-3
2017-10-06 20:27Use dHdt instead of dhdt.Constantine Khrulev1+1-1
2017-10-06 20:26PISMO is gone.Constantine Khrulev6+25-25
2017-10-06 20:25Fix a comment in IceRegionalModel.Constantine Khrulev1+1-1
2017-10-06 18:09Move bombproof_enth.Constantine Khrulev1+0-0
2017-10-06 18:03Add coding guidelines. (Unfinished and imperfect, but better than nothing.)Constantine Khrulev2+396-0
2017-10-06 17:23Add the how-to document. (Needs more work.)Constantine Khrulev2+378-0
2017-10-06 17:23Roles opt, var, and config: derive from literal instead of code.Constantine Khrulev1+4-3
2017-10-06 17:22Use literal text instead of :class:.Constantine Khrulev1+5-5
2017-10-06 16:27IceModelVec::write_in_glaciological_units is not needed any more.Constantine Khrulev38+69-110
2017-10-06 00:51Update labgum.Constantine Khrulev3+10-13
2017-10-06 00:45Allow overriding the value of the max. sensible SMB.Constantine Khrulev3+30-10
2017-10-06 00:45Mention -march=native.Constantine Khrulev1+12-0
2017-10-05 23:53Update the Ross validation example.Constantine Khrulev6+105-109
2017-10-05 19:15Add more FIXMEs.Constantine Khrulev2+5-2
2017-10-05 19:15Minor edits.Constantine Khrulev4+21-19
2017-10-05 19:13Enable ws-butler.Constantine Khrulev1+2-1
2017-10-05 19:12Remove FIXMEs.Constantine Khrulev5+12-16
2017-10-05 19:12Better table formatting.Constantine Khrulev1+3-1
2017-10-05 19:08White-space changesConstantine Khrulev1+3-2
2017-10-05 19:08Remove all mentions of -eo.Constantine Khrulev4+9-10
2017-10-05 19:05More work on the std-greenland example.Constantine Khrulev5+6-12
2017-10-05 19:05Add a list of CF standard names used by PISM.Constantine Khrulev3+177-0
2017-10-05 19:04Document -dry.Constantine Khrulev1+4-0
2017-10-05 02:09Allow -grid.Lz and -grid.Mz.Constantine Khrulev2+8-7
2017-10-05 02:08Use Diagnostic::wrap() to provide ocean_kill_mask.Constantine Khrulev2+1-33
2017-10-05 02:08Remove empty lines.Constantine Khrulev1+0-4
2017-10-04 23:29Update a footnote.Constantine Khrulev1+1-1
2017-10-04 04:17Fix some scripts.Constantine Khrulev2+9-7
2017-10-04 03:16Make scripts executable.Constantine Khrulev10+3-3
2017-10-04 02:12Extract more scripts.Constantine Khrulev5+43-37
2017-10-04 01:44Put code in separate files.Constantine Khrulev14+171-140
2017-10-04 01:00Remove lines saying "no short option."Constantine Khrulev2+5-168
2017-10-03 22:46Add IceModelVec2T notes to "Technical Notes."Constantine Khrulev3+293-261
2017-10-03 22:38Don't use math formulas in ice-bib.bib.Constantine Khrulev1+1-1
2017-10-03 22:38Get rid of the empty References section in the PDF.Constantine Khrulev3+49-7
2017-10-03 21:35Move two sections one level down.Constantine Khrulev1+2-2
2017-10-03 20:35Fix Tolly's name, add links to CalCalcs and VDT.Constantine Khrulev2+7-3
2017-10-03 20:23Move acknowledgements to the top, mention included third-party libraries.Constantine Khrulev3+15-3
2017-10-03 19:59Might as well call this 1.0.Constantine Khrulev1+2-2
2017-10-03 19:59Better document configuration.Constantine Khrulev1+13-6
2017-10-03 18:32Table column widths.Constantine Khrulev18+48-26
2017-10-03 18:32Fix indentation.Constantine Khrulev1+16-18
2017-10-03 04:39Work in progress. (Now we can produce both HTML and PFD.)Constantine Khrulev17+138-88
2017-10-02 22:56Split the Subglacier chapter.Constantine Khrulev5+609-594
2017-10-02 22:37\diff is for partial derivatives.Constantine Khrulev3+10-4
2017-10-02 22:14Add links to the section with the list of diagnostics.Constantine Khrulev5+12-16
2017-10-02 22:14Remove the link to page source from the footer.Constantine Khrulev1+2-0
2017-10-02 21:38Included files should be .txt (not .rst), add more Makefile targets.Constantine Khrulev94+439-210
2017-10-02 18:15A bit of cleanup (no real changes).Constantine Khrulev1+39-38
2017-10-02 18:14Clean up the list.Constantine Khrulev2+24-143
2017-09-30 06:56Remove the old list_diagnostics script.Constantine Khrulev1+0-42
2017-09-30 06:54Add an incomplete list of diagnostics and the script that created it.Constantine Khrulev3+1090-0
2017-09-29 17:13List parameter types, too.Constantine Khrulev2+316-310
2017-09-29 16:57Move parameter list to the top of the hierarchy in the Manual.Constantine Khrulev5+1850-1848
2017-09-29 16:49Wrap a line.Constantine Khrulev1+2-1
2017-09-29 16:49Add a script generating the parameter list and the list itself.Constantine Khrulev2+1984-1
2017-09-29 16:48Update doc string formatting in pism_config.cdl.Constantine Khrulev1+85-85
2017-09-28 23:54Add custom.cssConstantine Khrulev1+7-0
2017-09-28 23:24Add the Makefile.Constantine Khrulev1+35-0
2017-09-28 20:23GK does not need to be in a separate table.Constantine Khrulev1+15-13
2017-09-28 20:21More work...Constantine Khrulev24+604-441
2017-09-28 17:44Split the validation chapter.Constantine Khrulev3+362-352
2017-09-28 17:37Put intro first.Constantine Khrulev1+23-23
2017-09-28 17:37Remove extra tildes.Constantine Khrulev3+12-12
2017-09-28 17:24Improve some links.Constantine Khrulev15+48-43
2017-09-28 16:00Split the verification chapter.Constantine Khrulev12+233-225
2017-09-28 06:01Split the "Practical Usage" chapter.Constantine Khrulev16+1233-1177
2017-09-28 05:34Add "Previous, Up, Next" links to all pages.Constantine Khrulev1+35-0
2017-09-28 05:33Rename more entry points.Constantine Khrulev6+5-5
2017-09-28 04:54Rename entry points.Constantine Khrulev5+5-5
2017-09-28 04:52Split the climate forcing manual.Constantine Khrulev10+1401-1374
2017-09-27 05:57Split simplified experiments.Constantine Khrulev7+453-441
2017-09-26 23:04Add the bed roughness page.Constantine Khrulev2+305-0
2017-09-26 22:10More work on "modeling choices".Constantine Khrulev8+530-502
2017-09-26 21:57Re-organizing "modeling-choices"...Constantine Khrulev19+1994-1972
2017-09-26 21:17More work: shorten section names, other cleanup.Constantine Khrulev8+30-27
2017-09-26 18:31More re-organization.Constantine Khrulev10+542-525
2017-09-26 18:09Rename prologue.rst -> global.rst.Constantine Khrulev38+129-130
2017-09-26 18:04Better directory structure.Constantine Khrulev42+4915-4815
2017-09-26 17:36More proofreading.Constantine Khrulev1+32-27
2017-09-26 16:04More re-organization and proofreading.Constantine Khrulev42+1081-1057
2017-09-25 23:33Split the installation manual to make it easier to read.Constantine Khrulev16+655-667
2017-09-25 22:31More work: starting to re-organize stuff.Constantine Khrulev10+166-179
2017-09-25 20:23Put URLs in the "URL" field in ice-bib.bib.Constantine Khrulev1+30-27
2017-09-25 20:15Update a citation.Constantine Khrulev2+2-2
2017-09-25 20:14Changing the directory structure.Constantine Khrulev73+8395-7765
2017-09-22 20:37More progress.Constantine Khrulev9+59-57
2017-09-22 15:42Add more figures, wrap more text.Constantine Khrulev10+186-89
2017-09-22 06:03More work; add figures.Constantine Khrulev19+668-393
2017-09-22 01:22Fix ice-bib.bib (too many commas).Constantine Khrulev1+1-1
2017-09-22 01:21Use sphinxcontrib.bibtex for references.Constantine Khrulev17+277-396
2017-09-21 20:09Even more progress.Constantine Khrulev8+41-41
2017-09-21 19:52Lots of progress.Constantine Khrulev12+1066-530
2017-09-21 17:36Add more sections.Constantine Khrulev3+1190-0
2017-09-21 16:12More work.Constantine Khrulev10+1037-486
2017-09-21 02:38Add more fake references.Constantine Khrulev1+8-0
2017-09-21 00:00More progress.Constantine Khrulev20+1623-383
2017-09-20 19:05Add jako.rst.Constantine Khrulev2+392-0
2017-09-20 18:32More work.Constantine Khrulev7+101-123
2017-09-20 17:53More progress.Constantine Khrulev7+358-60
2017-09-20 06:05Minor fixes.Constantine Khrulev2+2-1
2017-09-20 06:05Add modeling-subglacier.rst.Constantine Khrulev1+384-0
2017-09-20 04:22Add getting-started to the whole thing.Constantine Khrulev1+1-0
2017-09-20 04:22Fix links by running "make linkcheck".Constantine Khrulev3+17-17
2017-09-20 04:12More progress.Constantine Khrulev3+21-29
2017-09-20 03:53Much better.Constantine Khrulev1+118-103
2017-09-19 23:55Good progress on practical-usage.Constantine Khrulev1+666-940
2017-09-19 21:41Add a bunch of files in doc/sphinx.Constantine Khrulev7+3449-0
2017-09-18 23:00Fix a heading.Constantine Khrulev1+1-1
2017-09-18 17:41Start converting bombproof_enth.md.Constantine Khrulev1+504-0
2017-09-18 17:18Commit vertchange.rst.Constantine Khrulev1+145-0
2017-09-17 01:21Commit latest changes in pism_forcing.rst.Constantine Khrulev1+18-15
2017-09-17 01:15Improve INSTALL.rst.Constantine Khrulev1+20-32
2017-09-17 01:03Add pism_forcing.rst.Constantine Khrulev1+1007-0
2017-09-17 00:50Add README.rstConstantine Khrulev1+60-0
2017-09-17 00:50Add a doxygen comment.Constantine Khrulev1+7-1
2017-09-13 23:59Fix a broken link.Constantine Khrulev1+1-1
2017-09-13 23:30Add icemodelvec2t.rst.Constantine Khrulev1+267-0
2017-09-13 22:16Add INSTALL.rst.Constantine Khrulev1+508-0
2017-09-13 01:10IceModelVec2T: handle the empty m_filename case.Constantine Khrulev1+5-0
2017-09-13 01:10Remove an extra initialization message.Constantine Khrulev1+0-1
2017-09-13 00:35Whitespace cleanup.Constantine Khrulev1+1-1
2017-09-13 00:34IceModelVec2T: make more methods protected.Constantine Khrulev1+3-3
2017-09-13 00:34IceModelVec2T::max_timestep() should be const.Constantine Khrulev2+2-2
2017-09-13 00:33Remove outdated doxygen comments.Constantine Khrulev1+2-54
2017-09-16 23:17Added support for EPSG:26710Andy Aschwanden1+5-25
2017-09-15 19:39Supporting EPSG:26710Andy Aschwanden1+31-0
2017-09-14 17:30Fixed ocean melt parameterizationAndy Aschwanden1+12-3
2017-09-11 21:34Fix a very stupid bug (I forgot about C++ switch fall-through).Constantine Khrulev1+4-0
2017-09-08 01:57Clean up pism::ocean::Runoff_SMB.Constantine Khrulev2+43-41
2017-09-07 21:38simplified runoff to ocean melt modifierAndy Aschwanden3+5-107
2017-09-06 22:55Use _floating instead of _shelf and mass_flux instead of ice_flux.Constantine Khrulev1+11-11
2017-09-06 19:54Better feedback on missing (or renamed) diagnosticsConstantine Khrulev1+51-3
2017-09-06 19:53Make tillwat available as a diagnosticConstantine Khrulev2+8-10
2017-09-06 19:53Always initialize IceModel::m_output_varsConstantine Khrulev1+2-1
2017-09-06 19:52Remove the IcebergMask variable from the list.Constantine Khrulev1+1-1
2017-09-06 18:27Finish implementing diagnostics needed to "balance the books."Constantine Khrulev2+247-156
2017-09-05 21:05Minor changes (white space and comments).Constantine Khrulev2+2-2
2017-09-05 23:46also added 'start' as an interval optionAndy Aschwanden2+8-4
2017-09-05 22:49timeline scripts:Andy Aschwanden2+24-8
2017-09-05 19:02Improve initialization of the PDD model.Constantine Khrulev3+96-61
2017-09-02 19:54these includes needed to compile on my laptop (Ubuntu 17.04 + petsc 3.7.6)Ed Bueler2+2-0
2017-09-02 00:33Revert "Remove Runoff_SMB (for now)."Constantine Khrulev5+281-0
2017-09-02 00:28Merge branch 'thk_calving' into devConstantine Khrulev13+1687-1399
2017-09-01 22:38Add a new command-line option: -uplift_file.Constantine Khrulev3+17-2
2017-09-01 20:23Fix a bug that resulted in running BedSmoother too often.Constantine Khrulev6+17-11
2017-09-01 19:44Clean up (replace get_theta() with theta()).Constantine Khrulev1+5-4
2017-09-01 17:25Update BedSmoother class name in the browser.Constantine Khrulev1+2-2
2017-07-31 21:51correct description of how effective pressure on till (Ntil) is computed from amount of water in till (Wtil); cite Bueler & van Pelt 2015; as suggested by Michael ImhofEd Bueler2+16-16
2017-07-31 20:53correction reported by Michael Imhof: till cohesion has input units of Pa not kPaEd Bueler1+1-1
2017-06-28 12:15Added factor 2 in deviatoric stress calculation according to eq.6 in winkelmann et al., 2011Torsten Albrecht1+3-3
2017-06-17 00:03Allow disabling PISM's doc targets.Constantine Khroulev1+7-4
2017-06-16 23:22Variables "lat" and "lon" should not have grid_mapping. See #384.Constantine Khroulev1+5-3
2017-06-16 22:35Fix the computation of the SMB in the PDD code.Constantine Khroulev1+9-4
2017-06-16 17:42Rename basal_{grounded,floating}_mass_flux.Constantine Khroulev1+6-6
2017-06-16 17:41Rename the basal melt rate field in EnergyModel.Constantine Khroulev16+44-42
2017-06-16 16:27Fix #387. (Use SWIG_ADD_LIBRARY if CMake version >= 3.8.)Constantine Khroulev1+15-3
2017-06-15 23:42Fix processing of "-no_foo" (e.g. -no_mass). See #385.Constantine Khroulev1+2-1
2017-06-15 23:31Fix #384.Constantine Khroulev1+4-3
2017-06-15 22:48Add -extra_vars shortcuts.Constantine Khroulev1+28-1
2017-06-15 22:47Fix variable names.Constantine Khroulev1+8-8
2017-06-13 23:18Implement 2D fluxes in terms of "amounts" and "masses".Constantine Khroulev3+576-226
2017-06-13 18:57Add a FIXME comment.Constantine Khroulev1+4-0
2017-06-12 22:24Add code needed to generate the list of supported diagnostics.Constantine Khroulev5+183-40
2017-06-12 22:22Fixed bug in slvol calculationTorsten Albrecht1+1-1
2017-05-26 08:54Added option -use_mean_hardness for calculation of deviatoric stressesTorsten Albrecht1+18-0
2017-05-05 21:08Floating-point addition is not associative.Constantine Khroulev3+54-40
2017-05-05 20:33A little better: now I can see both the old and the new H.Constantine Khroulev1+7-3
2017-05-05 20:28Add a bit of code for debugging.Constantine Khroulev1+6-0
2017-05-05 19:08Mark CMAKE_BUILD_TYPE as "not advanced".Constantine Khroulev1+1-0
2017-05-05 05:12More work on IceRegionalModel.Constantine Khroulev2+53-53
2017-05-05 04:41Some cleanup in IceModel.Constantine Khroulev3+47-55
2017-05-05 03:04Working on fixing the -regional mode.Constantine Khroulev3+53-82
2017-05-04 19:02Whitespace changes.Constantine Khroulev1+7-7
2017-05-04 18:44Fix #379.Constantine Khroulev1+14-12
2017-05-04 00:55Rename calving_mass_flux -> discharge_mass_flux.Constantine Khroulev1+5-5
2017-05-03 23:34Merge branch 'ckhroulev/geometry-evolution' into devConstantine Khroulev210+8765-8747
2017-05-03 19:59added script that allows overwriting the time axis of a netcdf file. Useful for postprocessing.Andy Aschwanden1+143-0
2017-05-01 16:10Minor cleanup after merging #381.Constantine Khroulev2+3-2
2017-05-01 16:07Merge pull request #381 from pism/efischer/170417-EverytraceConstantine Khroulev3+8-1
2017-04-24 17:32Remove the HDF5-based parallel I/O code. Again.Constantine Khroulev8+7-1364
2017-04-18 02:52SSAFD KSP solver: use the initial residual norm.Constantine Khroulev1+5-1
2017-04-19 23:41Fix to nc2cdo.py: always overwrite lat/lon and bounds, even if they are already presentAndy Aschwanden1+28-22
2017-04-13 13:08Fix -hydrology_const_bmelt units in manual.Julien Seguinot1+1-1
2017-04-12 17:51Fix #374 (both short and long command-line options should work).Constantine Khroulev1+17-3
2017-04-05 15:56Fix the LC bed deformation model regression test.Constantine Khroulev1+1-1
2017-03-27 17:48Make sure profiling data includes time spent updating the energy state.Constantine Khroulev3+8-3
2017-03-24 17:03Use the new IceModelVec2S constructor and remove redundant metadata.Constantine Khroulev1+2-6
2017-03-24 17:03Cleanup.Constantine Khroulev1+3-8
2017-03-24 17:00Add a constructor to IceModelVec2S.Constantine Khroulev2+10-0
2017-03-24 03:15IceRegionalModel: don't modify the no_model_mask if no_model_strip width is 0.Constantine Khroulev2+6-2
2017-03-24 03:14von Mises calving: calve cells right at the grounding line.Constantine Khroulev1+3-3
2017-03-24 03:13Stress-based calving methods use ghosts of strain rates.Constantine Khroulev1+2-1
2017-03-23 19:25Fix a bug in the previous commit.Constantine Khroulev1+5-4
2017-03-23 19:15Add -calving_wrap_around.Constantine Khroulev2+6-1
2017-03-23 18:57pismo: add -list_diagnostics and -profiling.Constantine Khroulev1+26-7
2017-03-23 18:56IceRegionalModel: add stress balance to the list of submodels.Constantine Khroulev1+3-1
2017-03-22 20:52pismo: Read the no model mask from the bootstrapping file.Constantine Khroulev1+8-0
2017-03-07 23:01Merge branch 'ckhroulev/issue-370-dev' into devConstantine Khroulev26+1073-579
2017-02-22 05:16Replace help@pism-docs.org with uaf-pism@alaska.edu.Constantine Khroulev5+10-10
2017-02-15 18:16Remove Pism_BUILD_TYPE...Constantine Khroulev1+2-6
2017-02-15 18:16Always install examples.Constantine Khroulev1+4-6
2017-02-14 23:32Improve calving code.Constantine Khroulev8+97-88
2017-02-14 00:17Add vonmises_calving and time-series to the "tongues" test case.Constantine Khroulev1+7-2
2017-02-13 20:02Remove bootstrapping.defaults.tillphi. See #372.Constantine Khroulev3+91-122
2017-02-09 17:26pism_strictly_static: check BUILD_SHARED_LIBS instead of setting it.Constantine Khroulev1+5-2
2017-02-07 19:57Fix examples/std-greenland/preprocess.shConstantine Khroulev1+7-7
2016-12-07 20:14Updates to commentsElizabeth Fischer2+16-0
2017-02-07 17:33Fix #366.Constantine Khroulev5+13-13
2017-02-07 18:48Merge pull request #369 from pism/efischer/170206-InstallExamplesConstantine Khroulev1+7-0
2017-02-07 18:46Merge pull request #367 from pism/efischer/170206-PISMPrefixConstantine Khroulev7+30-30
2017-02-06 17:41Document the recent change in examples/std-greenland/spinup.sh.Constantine Khroulev1+5-0
2017-02-06 16:56Update examples/std-greenland/spinup.shConstantine Khroulev1+4-2
2017-02-06 16:31Fix a typo.Constantine Khroulev1+3-3
2017-02-02 22:34Merge branch 'ckhroulev/extrapolation' into devConstantine Khroulev26+628-419
2017-02-01 22:27Enthalpy is managed by the energy model now.Constantine Khroulev1+1-2
2017-02-01 16:28Changed option name also in examples/ross/fractures/run_frac.shTorsten Albrecht1+2-2
2017-02-01 16:21Changed name of -fractures option in iMfractures.cc, to make examples/ross/run_frac.sh run again...Torsten Albrecht1+2-2
2017-01-31 21:39Remove a duplicate line from the list of diagnostics.Constantine Khroulev1+0-1
2017-01-31 17:18Fix #364.Constantine Khroulev1+1-0
2017-01-31 00:35land_ice_calving_rate is a CF standard name we can use.Constantine Khroulev1+3-3
2017-01-31 00:35Add ISMIP6 proposed standard names.Constantine Khroulev4+27-17
2017-01-26 19:24Terminus advance test: -part_grid is enabled whenever calving is on.Constantine Khroulev1+3-6
2017-01-26 19:23Fix CFL violation counting.Constantine Khroulev1+5-2
2017-01-26 07:34Fix a bug in the Logger code.Constantine Khroulev1+2-2
2017-01-26 05:21Fix options::deprecated().Constantine Khroulev1+2-1
2017-01-26 05:20Fix initialization of PrescribedSliding.Constantine Khroulev2+6-5
2017-01-26 01:14Improve command-line options for boolean flags.Constantine Khroulev1+58-19
2017-01-25 22:58Implement -float_kill_calve_near_grounding_line.Constantine Khroulev3+21-3
2017-01-24 20:36Fix compilation issues (PO_delta_MBP -> PO_frac_MBP).Constantine Khroulev2+2-2
2017-01-25 00:53updated 2017 funding sourcesAndy Aschwanden1+1-1
2017-01-24 01:39Cleanup: make it clear which run_stats attributes are time-independent.Constantine Khroulev1+13-9
2017-01-24 01:38Remove IceModel::write_metadata().Constantine Khroulev3+60-64
2017-01-19 23:33Make sure that petsc4py version gets recorded.Constantine Khroulev2+8-1
2017-01-19 20:02PO_delta_MBP.cc -> PO_frac_MBP.cc in src/coupler/CMakeLists.txtConstantine Khroulev1+1-1
2017-01-19 18:24Add the script making profiling pie charts.Constantine Khroulev1+139-0
2017-01-03 19:47Clean up: use initializer_list to build mask_values.Constantine Khroulev1+7-6
2016-12-30 02:27Clean-up: use the initializer_list AccessList constructor.Constantine Khroulev86+278-996
2016-12-30 02:22Clean up IceModel::compute_temperate_base_fraction(...).Constantine Khroulev1+9-11
2017-01-15 21:25renamed “delta_MBP” to “frac_MBP” since it is a multiplier not an offset.Andy Aschwanden5+80-80
2017-01-01 05:43minor cleanup in FE documentation (which is excellent)Ed Bueler1+24-19
2016-12-22 21:32Remove __attribute__((noreturn)) from virtual methods.Constantine Khroulev6+13-7
2016-12-22 18:53Fixing GCC's compiler warnings.Constantine Khroulev5+6-1
2016-12-21 21:44Clean up #includes.Constantine Khroulev31+15-32
2016-12-21 08:25Fix some clang static analyzer's warnings.Constantine Khroulev3+31-20
2016-12-21 01:37I think it's OK to use namespace pism here.Constantine Khroulev1+2-12
2016-12-21 01:29Add a constructor to InputOptions.Constantine Khroulev2+15-9
2016-12-21 01:29Fix file name case in #includes.Constantine Khroulev3+3-3
2016-12-20 20:55Rename some variables.Constantine Khroulev1+3-3
2016-12-20 20:55Add a default constructor to CFLData.Constantine Khroulev2+7-0
2016-12-20 20:03Fix a typo in the manual.Constantine Khroulev1+1-1
2016-12-20 19:36Turn off C++ extensions.Constantine Khroulev1+1-0
2016-12-20 17:58deleted src/base/iMbootstrap.ccConstantine Khroulev3+82-122
2016-11-15 22:05Move surface and basal MB into a separate method.Constantine Khroulev2+122-73
2016-11-15 18:58Separating source terms in the mass continuity equation.Constantine Khroulev1+86-58
2016-11-15 18:37Add a constructor to StarStencil.Constantine Khroulev1+9-0
2016-11-15 00:44More cleanup in iMgeometry.ccConstantine Khroulev4+56-46
2016-11-14 22:33Remove the do_redist variable.Constantine Khroulev1+8-14
2016-11-14 22:28Add flux and velocity arguments to massContExplicitStep().Constantine Khroulev7+27-16
2016-11-14 22:17Add the dt argument to IceModel::massContExplicitStep().Constantine Khroulev7+20-20
2016-11-14 21:32Clean up in the mass transport code.Constantine Khroulev1+4-4
2016-12-19 00:09Rename SurfaceModel methods.Constantine Khroulev36+135-135
2016-12-18 23:20Merge branch 'ckhroulev/dev' into devConstantine Khroulev267+6062-6612
2016-12-09 22:25Fix a typo.Constantine Khroulev1+2-3
2016-12-09 22:23Don't add -std=c++11 if it is set already.Constantine Khroulev1+4-3
2016-12-09 22:16Link to HDF5 libraries, too.Constantine Khroulev1+8-3
2016-12-09 22:02Use unique_ptr instead of auto_ptr to silence a warning.Constantine Khroulev1+1-1
2016-12-09 20:39Fix a compilation error.Constantine Khroulev1+2-1
2016-12-09 20:23Trying to support Intel compilers...Constantine Khroulev1+7-0
2016-12-05 06:07Use C++11 automatic type deduction to clean things up.Constantine Khroulev43+311-434
2016-12-04 05:35Add std::vector<int> IceGrid::compute_point_neighbors(...)Constantine Khroulev2+11-7
2016-12-03 02:54The EPSG code processing regression test requires PROJ.4.Constantine Khroulev1+3-1
2016-12-03 01:47Merge branch 'ckhroulev/cxx11' into devConstantine Khroulev35+128-241
2016-12-01 20:41Remove old code writing SSAFD systems to .m files.Constantine Khroulev2+0-111
2016-12-01 20:38Cleanup.Constantine Khroulev4+19-16
2016-12-01 20:28Fix ssa_test_cfbc.py, add it to regression tests.Constantine Khroulev2+15-13
2016-11-29 04:23Implement "glacierized" scalar diagnostics. See #284.Constantine Khroulev20+490-276
2016-11-28 20:34Remove IceModel::m_{strain_rates,deviatoric_stresses}.Constantine Khroulev4+52-58
2016-09-22 18:31Implement the 3D viscosity diagnostic. See #320.Constantine Khroulev4+222-8
2016-11-22 19:21Apply topg_delta *after* regridding.Constantine Khroulev1+4-4
2016-11-22 13:09updated velocity data url for Ross exampleTorsten Albrecht1+4-6
2016-11-21 21:11Add -topg_delta_file topg_delta.nc.Constantine Khroulev6+42-6
2016-11-21 20:06Sort parameters in pism_config.cdl.Constantine Khroulev1+1165-1186
2016-11-19 01:27Try to fix a compilation error.Constantine Khroulev1+3-3
2016-11-18 00:13PDD model: add {saccum,smelt,srunoff}_average.Constantine Khroulev2+139-0
2016-11-19 00:08Add a helper template class for computing time averagesConstantine Khroulev1+56-0
2016-11-16 20:39Remove all mentions of pism_installation.pdf.Constantine Khroulev6+5-9
2016-11-16 19:55Apply lapse rate corrections throughout the domain.Constantine Khroulev1+3-8
2016-11-16 19:54Use [kg m-2 s-1] in pism_forcing.tex.Constantine Khroulev1+13-11
2016-11-11 01:00Remove BedrockSystem.hh and close #329.Constantine Khroulev1+0-50
2016-11-10 17:15SSARun.write(): write more fields for debugging.Constantine Khroulev2+22-0
2016-11-10 17:15Add ocean_pressure_difference to SSA diagnostics.Constantine Khroulev3+78-4
2016-11-09 23:27Fix writing in SSARun.write().Constantine Khroulev1+4-4
2016-11-09 23:26Fix writing in ModelVecs.writeall().Constantine Khroulev1+11-1
2016-11-09 23:25Fix ssa_forward.py.Constantine Khroulev1+1-1
2016-11-08 00:24If a CMake option is set the corresponding package is *required*.Constantine Khroulev1+4-4
2016-11-04 00:14Cleanup: move implementations out of the header.Constantine Khroulev4+33-31
2016-11-04 00:05Cleanup: Remove "an unholy amount of escapes."Constantine Khroulev1+22-22
2016-11-03 20:22Added accessors to support IceBin/ModelE developmentElizabeth Fischer1+4-0
2016-11-02 22:52Fix #166: Install headers in include/pism, skip 3rd party headers and empty directories.Constantine Khroulev1+14-2
2016-11-02 22:30PIO::PIO(...) opens or creates the fileConstantine Khroulev43+247-392
2016-11-02 20:42Close #313.Constantine Khroulev1+27-11
2016-11-02 01:43Add SWIG and petsc4py to version info.Constantine Khroulev3+21-1
2016-11-02 00:33Add MPI library version; print more info when -version is set.Constantine Khroulev2+8-1
2016-11-02 00:05Add CMake and Jansson to version info.Constantine Khroulev4+17-4
2016-11-01 23:23Add pism::version()Constantine Khroulev5+49-1
2016-11-01 22:20Stop if PetscScalar is complex. See #237.Constantine Khroulev2+31-0
2016-11-01 00:49Convert pism_installation.tex to Markdown; replace INSTALL.mdConstantine Khroulev4+394-891
2016-10-11 11:49The newest TeXlive / Komascript bails when using \bf, \sc, etc. This is a temporary fix, ideally, one would replace all deprecated commands, but it appears that '\sc' is hard coded somewhere in the bibliography? style fileAndy Aschwanden3+33-3
2016-10-06 00:20PDD: add {define,write}_model_state().Constantine Khroulev2+11-0
2016-10-06 00:20SurfaceModel: add {define,write}_model_state.Constantine Khroulev2+16-0
2016-10-06 00:19PDD: improve API and add diagnostics.Constantine Khroulev2+201-0
2016-10-06 00:17Component: add {define,write}_model_state().Constantine Khroulev3+45-0
2016-10-05 23:16Remove a whole lot of repetitive code.Constantine Khroulev16+36-506
2016-10-05 23:00Add ocean model diagnostics.Constantine Khroulev3+144-0
2016-10-05 22:59Cleanup.Constantine Khroulev1+0-1
2016-10-05 22:15Take advantage of the recent change.Constantine Khroulev18+21-610
2016-10-05 20:06Create surface model diagnostics.Constantine Khroulev2+173-0
2016-10-05 16:55Fix a compilation error with old GSL (< 1.15).Constantine Khroulev1+2-2
2016-10-04 16:27Bring back -save_force_output_times.Constantine Khroulev1+2-0
2016-10-04 15:45Fix #354.Constantine Khroulev1+1-1
2016-10-04 00:57Remove the last "friend" of IceModel.Constantine Khroulev2+2-5
2016-10-04 00:34Move max CFL time step computation to timestepping.{cc,hh}.Constantine Khroulev8+275-180
2016-10-03 23:14Move max_timestep_cfl_[23]d out of IceModel.Constantine Khroulev2+115-45
2016-10-03 19:32Add error location info to RuntimeError. See #353.Constantine Khroulev104+659-610
2016-10-03 18:22Minor cleanup.Constantine Khroulev3+5-4
2016-10-01 05:04Fix compilation issues on beauregard.Constantine Khroulev4+10-7
2016-10-01 05:03Mark IceModel::initialize_[23]d as "noreturn".Constantine Khroulev1+2-2
2016-10-01 05:02Mark IceCompModel bootstrapping methods as "noreturn".Constantine Khroulev1+2-2
2016-10-01 04:06enthalpyAndDrainageStep -> enthalpyStepConstantine Khroulev7+19-21
2016-10-01 04:03Switch to etags from Emacs 25.1.Constantine Khroulev1+1-1
2016-10-01 03:29Move age time-stepping code to src/base/age.Constantine Khroulev5+199-148
2016-10-01 03:15SWIG bootstrapping heuristics (to make testing easier).Constantine Khroulev1+3-0
2016-10-01 02:27Put ice temperature heuristics into separate functions.Constantine Khroulev4+134-35
2016-10-01 01:59Remove IceModel::CFLmaxdt and CFLmaxdt2D.Constantine Khroulev3+5-20
2016-10-01 00:38Age, temperature, enthalpy step methods take the dt argument.Constantine Khroulev10+25-18
2016-10-01 00:33More cleanup in IceModel.Constantine Khroulev6+41-34
2016-09-30 23:49Start work on isolating the age code.Constantine Khroulev3+58-13
2016-09-30 22:31Cleanup in IceModel and elsewhere.Constantine Khroulev8+195-133
2016-09-30 21:07Clean up in the temperature-based code.Constantine Khroulev4+45-43
2016-09-30 19:51temperatureStep() uses EnergyModelInputs.Constantine Khroulev5+62-66
2016-09-30 19:14enthalpyAndDrainageStep() uses EnergyModelInputs.Constantine Khroulev5+146-61
2016-09-30 17:45Create EnergyModelStats.Constantine Khroulev8+53-57
2016-09-30 16:32Rename a variable: Rb -> basal_frictional_heating.Constantine Khroulev1+4-4
2016-09-29 19:58Fix targets "coverage_report" and "coverage_reset".Constantine Khroulev1+5-3
2016-09-29 19:57Timeseries: add {get,set}_use_bounds().Constantine Khroulev2+12-0
2016-09-29 18:11Improve PISM.Context().Constantine Khroulev1+22-27
2016-09-29 18:11Wrap TimeseriesMetadata.Constantine Khroulev2+5-4
2016-09-29 00:07Use new code computing lon. and lat. bounds.Constantine Khroulev3+15-48
2016-09-28 23:53Isolate code computing lon and lat bounds.Constantine Khroulev2+65-2
2016-09-28 23:40Better to have separate functions for lon and lat.Constantine Khroulev3+28-18
2016-09-28 23:31Clean up lon bounds and lat bounds diagnostics.Constantine Khroulev2+9-30
2016-09-28 23:29Isolate the code computing cell areas, lon, lat.Constantine Khroulev3+150-117
2016-09-28 22:28Add a simple wrapper for projPJ.Constantine Khroulev1+54-0
2016-09-28 21:32Remove IceModel::m_mapping.Constantine Khroulev4+8-10
2016-09-28 21:32Print commands to simplify debugging.Constantine Khroulev1+11-7
2016-09-28 21:23Hide private IceGrid methods.Constantine Khroulev2+43-43
2016-09-28 21:11Don't write the "grid_mapping" attribute when "mapping" is not set.Constantine Khroulev3+9-2
2016-09-28 20:57Use the new grid mapping code in IceModel.Constantine Khroulev2+31-66
2016-09-28 20:56Don't write the empty "mapping" to output files.Constantine Khroulev1+8-5
2016-09-28 20:55Improve handling of grid mappings.Constantine Khroulev4+120-29
2016-09-28 18:52Add VariableMetadata::has_attributes().Constantine Khroulev2+5-0
2016-09-28 18:51Improve docs of write_attributes().Constantine Khroulev1+6-5
2016-09-28 16:48Move check_mapping_equivalence() to projection.cc.Constantine Khroulev4+71-74
2016-09-28 16:47Add SWIG wrapping of std::map<...>.Constantine Khroulev1+3-0
2016-09-28 00:13Isolate the code checking the equivalence of an EPSG code to a CF mapping variable.Constantine Khroulev3+79-53
2016-09-28 00:13Use a const reference instead of a value.Constantine Khroulev2+4-4
2016-09-28 00:13Cosmetic changes.Constantine Khroulev1+2-2
2016-09-27 17:03IceModel: cosmetic changes.Constantine Khroulev1+6-6
2016-09-27 17:02SSAFD: m_ prefix, remove a friend class, add integrated_viscosity().Constantine Khroulev2+62-57
2016-09-27 16:59Clean up some command-line options.Constantine Khroulev15+50-100
2016-09-26 23:16Exclude vdt from doxygen docs.Constantine Khroulev1+2-1
2016-09-26 23:01Remove bedrockThermalUnit.cc.Constantine Khroulev1+0-467
2016-09-26 22:46Remove {get,set}VerbosityLevel(), verbPrintf().Constantine Khroulev40+23-210
2016-09-26 22:08Remove most uses of verbPrintf.Constantine Khroulev7+22-28
2016-09-26 21:26Cleanup in deformation.{cc,hh}.Constantine Khroulev2+28-28
2016-09-26 20:07Remove simple{H,I,J,K,L}. See #343.Constantine Khroulev18+312-728
2016-09-26 19:34Cleanup in ShallowStressBalance.Constantine Khroulev7+124-130
2016-09-26 18:48Add some fracture density options to the config.Constantine Khroulev2+52-9
2016-09-26 17:32Add a "regression test" using bed_deformation.pyConstantine Khroulev1+2-0
2016-09-26 17:08Fix bed_deformation.py, remove tryLCbd.cc. See #181.Constantine Khroulev3+78-363
2016-09-26 16:47Fix a bug in IceModel::list_diagnostics(); cleanup.Constantine Khroulev2+5-6
2016-09-26 16:44PBLingleClark: fix init_with_inputs_impl.Constantine Khroulev2+26-13
2016-09-24 00:29Remove IceModel::maxdt_temporary.Constantine Khroulev3+0-11
2016-09-24 00:16Move min and max thickness checks out of IceModel.Constantine Khroulev3+20-16
2016-09-24 00:11Remove -view_sys, -id, -jd.Constantine Khroulev7+0-44
2016-09-23 23:47Bed deformation without IceModel, in Python. See #181.Constantine Khroulev1+137-0
2016-09-23 23:26A bit of cleanup.Constantine Khroulev1+5-5
2016-09-23 18:00IceModel_max_diffusivity is not a "friend".Constantine Khroulev1+0-1
2016-09-23 17:40Cleanup in IceModel.Constantine Khroulev8+130-94
2016-09-22 18:31StressBalance: add const-qualifiers.Constantine Khroulev2+16-16
2016-09-22 17:27Diagnostics: add m_fill_value.Constantine Khroulev5+50-54
2016-09-22 05:45Add the 3D hardness diagnostic. See #320.Constantine Khroulev2+70-0
2016-09-22 05:36Cleanup.Constantine Khroulev2+0-2
2016-09-22 05:01Major cleanup in IceModel.Constantine Khroulev9+311-266
2016-09-21 20:57Option -part_grid implies -part_redist.Constantine Khroulev24+33-48
2016-09-21 18:34IceModel::updateSurfaceElevationAndMask -> enforce_consistency_of_geometry()Constantine Khroulev5+9-9
2016-09-21 16:32Cleanup in SIAFD.Constantine Khroulev3+6-8
2016-09-21 02:10Add a bed smoother regression test in Python.Constantine Khroulev3+139-0
2016-09-21 02:07Remove bedrough_test and the associated regression test.Constantine Khroulev4+0-191
2016-09-21 02:06Bed smoother: remove get_smoothing_domain().Constantine Khroulev2+9-19
2016-09-21 00:20Bed smoother: add the m_ prefix.Constantine Khroulev2+102-103
2016-09-20 22:47Remove flowlaw_test and associated regression tests.Constantine Khroulev9+78-364
2016-09-20 00:05Hide obscure HDF5-related variables.Constantine Khroulev1+1-0
2016-09-21 20:15CalvingFrontRetreat: avoid "wrapping around".Constantine Khroulev1+28-5
2016-09-21 18:34Forgot to enable frontal melt.Constantine Khroulev1+10-0
2016-09-16 23:51Implement calving front retreat due to frontal melting.Constantine Khroulev6+92-9
2016-09-16 00:06Fix a comment.Constantine Khroulev1+1-1
2016-09-15 22:05Fix a bug.Constantine Khroulev6+6-10
2016-09-14 02:23Remove an unused parameter (again).Constantine Khroulev1+0-5
2016-09-14 01:50Rename -cfl_eigen_calving to -calving_cfl.Constantine Khroulev3+6-5
2016-09-14 01:35Remove an old source code file.Constantine Khroulev1+0-490
2016-09-13 07:59Merge branch 'ckhroulev/json-config' into devConstantine Khroulev186+4552-2620
2016-09-13 07:02Fix compilation errors on beauregard.Constantine Khroulev3+5-0
2016-09-13 06:49Fix compilation errors and warnings.Constantine Khroulev4+8-8
2016-09-13 06:35Merge branch 'ckhroulev/issue-352' into devConstantine Khroulev69+972-1357
2016-09-13 00:49Implement topg_sl_adjusted.Constantine Khroulev1+4-3
2016-09-13 00:49Add IceModel::bed_model() const.Constantine Khroulev2+5-0
2016-09-13 00:41Add IceModel::ocean_model() plus cleanup in stress balance code.Constantine Khroulev17+81-44
2016-09-13 00:24Add const qualifiers to some OceanModel methods.Constantine Khroulev24+76-76
2016-09-13 00:11Cleanup in IceModel.Constantine Khroulev1+37-19
2016-09-09 23:55Try to fix inversion tools in dev.Constantine Khroulev1+2-2
2016-09-10 07:24Tweaking the comments.Constantine Khroulev1+2-2
2016-09-10 07:22Clean up in the UDUNITS2 wrapper.Constantine Khroulev1+7-8
2016-09-10 07:21Clean up in IceModel.Constantine Khroulev2+1-3
2016-09-07 19:14Update relax-topography/preprocess.sh.Constantine Khroulev1+4-4
2016-09-07 19:07Update the SeaRISE-Antarctica pre-processing script.Constantine Khroulev1+8-5
2016-09-07 18:55Update the std-greenland pre-processing script.Constantine Khroulev1+5-4
2016-09-07 18:22Merge branch 'ckhroulev/precipitation-units' into devConstantine Khroulev11+47-39
2016-09-07 15:35Remove "pdd_old". Again.Constantine Khroulev6+1-1130
2016-09-07 01:07Add an old code stub (to stop bit rot). See #329.Constantine Khroulev1+50-0
2016-09-07 00:14Commit some old work on #343 to prevent bit rot.Constantine Khroulev19+632-1011
2016-09-06 22:43Remove Component::find_pism_input()Constantine Khroulev31+257-361
2016-09-06 19:36IceModel::m_bedtoptemp is no longer needed.Constantine Khroulev3+3-11
2016-09-06 01:19More work on the bedrock thermal unit.Constantine Khroulev29+1194-803
2016-09-01 16:15Fix a typo in nccmp.py.Constantine Khroulev1+1-1
2016-08-31 17:13Set the flux at the ice base during initialization.Constantine Khroulev1+10-8
2016-08-30 21:47Fix processing of -Mbz and -Lbz.Constantine Khroulev1+15-18
2016-08-30 18:55Initialization overhaul. WIP; some tests fail.Constantine Khroulev15+280-313
2016-08-30 18:54Add BTU_Verification.{cc,hh}Constantine Khroulev3+140-0
2016-08-29 22:18pism::IceModel: the m_ prefix.Constantine Khroulev16+118-118
2016-08-26 20:33Initialization overhaul. Work in progress. Compiles.Constantine Khroulev16+490-640
2016-08-25 00:18Run clang-format on src/base/iMfractures.ccConstantine Khroulev1+113-130
2016-08-25 00:12IceModelVecs owned by IceModel: m_ prefix.Constantine Khroulev16+202-200
2016-08-24 23:42Remove IceModel::m_external_{surface,ocean}_model.Constantine Khroulev4+2-14
2016-08-24 17:02Remove mask_is_floating_thickness_standard. See #349.Constantine Khroulev3+4-12
2016-08-24 00:20Fix #351 (duplicate history in -extra_files).Constantine Khroulev3+59-27
2016-08-22 23:48Add an ocean model initialization helper.Constantine Khroulev5+282-2
2016-08-22 19:39pismr and pisms use surface::InitializationHelper.Constantine Khroulev3+17-9
2016-08-22 18:57More work on the initialization helper code.Constantine Khroulev2+95-26
2016-08-22 18:55Implement init_step()Constantine Khroulev3+31-23
2016-08-20 04:00Whitespace changes.Constantine Khroulev2+3-3
2016-08-20 03:04m_ prefix in IceModel.Constantine Khroulev5+38-37
2016-08-20 02:56pism::YieldStress models are diagnosticConstantine Khroulev14+73-79
2016-08-20 00:04Add new pism::surface::Initialization (still broken).Constantine Khroulev3+194-0
2016-08-20 00:03Very minor cleanup.Constantine Khroulev1+4-4
2016-08-19 21:10Modifier cleanup (m_ prefix stuff).Constantine Khroulev25+196-199
2016-08-19 20:26Store current scalar forcing in m_current_forcingConstantine Khroulev4+15-12
2016-08-19 20:05More cleanup.Constantine Khroulev4+0-9
2016-08-19 19:29Clean up in Timeseries.Constantine Khroulev2+17-14
2016-08-19 00:04Scalar forcing modifiers: add the m_ prefix.Constantine Khroulev11+97-97
2016-08-12 23:19Add the m_ prefix.Constantine Khroulev7+40-40
2016-08-12 22:07Fix sia_forward.py and add a regression test.Constantine Khroulev3+38-4
2016-08-12 22:01Delete tauc2tillphi.py (bit rot took over).Constantine Khroulev1+0-243
2016-08-12 22:01IceGrid::FromFile: use PIO instead of the file name.Constantine Khroulev5+25-14
2016-08-12 19:20Grid initialization: keep the input file open.Constantine Khroulev1+3-2
2016-08-11 19:01OK to use netcdf3 to initialize time.Constantine Khroulev1+1-1
2016-08-11 17:24Reduce the number of times the input file is openedConstantine Khroulev5+84-92
2016-08-10 18:42Whitespace changes.Constantine Khroulev6+19-18
2016-08-18 22:36Fix message order (-bed_def none initialization).Constantine Khroulev1+4-3
2016-08-18 22:14do_calving() calls Href_cleanup() already.Constantine Khroulev1+0-2
2016-08-18 22:15IceModel::do_calving(): update mask after Href_cleanup().Constantine Khroulev1+4-0
2016-08-18 22:15Always ensure consistency of geometry after initialization.Constantine Khroulev1+3-2
2016-07-21 18:09attempt to make links to netcdf4-python clear and up to dateEd Bueler1+4-4
2016-06-11 18:24fixed typo in factoryAndy Aschwanden1+1-1
2016-06-11 17:55fixing unitsAndy Aschwanden1+2-2
2016-06-11 17:40added new scalar fractions SMB to POFactoryAndy Aschwanden1+2-0
2016-06-11 17:08added ocean_frac_SMB fractional shelf-base mass flux offsetsAndy Aschwanden4+213-0
2016-06-09 16:55Merge branch 'dev' of https://github.com/pism/pism into devAndy Aschwanden1+1-1
2016-06-09 16:53Change of units for ocean sub shelf mass flux modifier since "-ocean_given file.nc" expects [kg m-2 s-1], the modifier should use the same units, not [m year-1]Andy Aschwanden2+4-6
2016-05-21 16:54Fix compiler warnings.Constantine Khroulev2+6-5
2016-05-21 16:42Merge branch 'dev' of https://github.com/pism/pism into devAndy Aschwanden1+1-10
2016-05-21 16:41writing extra split files "%s_%s.nc" instead of "%s-%s.ncAndy Aschwanden1+1-1
2016-05-20 23:02Merge branch 'ckhroulev/eigen-calving-cleanup' into devConstantine Khroulev51+2250-1259
2016-05-20 22:36-{surface,atmosphere} given: set valid_{min,max}Constantine Khroulev2+14-1
2016-05-19 22:32Install all Python scripts in util/. Fixes #346.Constantine Khroulev1+3-9
2016-05-19 22:27Re-organize the "util" sub-directory. See #346.Constantine Khroulev8+12-126
2016-05-19 21:45Rename cell_mass -> ice_mass, include Href.Constantine Khroulev4+35-29
2016-05-16 00:50added spatial diagnostic "cell_mass"Andy Aschwanden5+81-1
2016-05-11 01:54cosmetical change: report grid dx, dy in meters for dx, dy < 1000m during model initialization.Andy Aschwanden1+6-1
2016-05-03 19:47Remove all mentions of petsc_history.Constantine Khroulev1+0-9
2016-05-03 19:49Make PISM's inverse code compatible with PETSc 3.7.Constantine Khroulev1+6-1
2016-05-06 01:20Add wvelsurf to output_2dbig.Constantine Khroulev1+1-1
2016-04-28 17:06Clean up config. parameters output_{medium,2dbig,big}.Constantine Khroulev2+24-38
2016-04-12 23:30PIO: remove awkward "optimizations".Constantine Khroulev2+9-47
2016-04-12 22:05NetCDFConfig::all_strings_impl: skip booleansConstantine Khroulev1+18-2
2016-04-12 18:02Use clang-format to clean up code in src/icebin/Constantine Khroulev8+497-533
2016-04-12 15:32Add .clang-formatConstantine Khroulev1+68-0
2016-04-20 19:57removed relative_flotation (the meaning needs rethinking) and clarified height_above_flotationAndy Aschwanden3+1-75
2016-04-20 17:26fixed typo in relative_flotationAndy Aschwanden1+1-1
2016-04-20 05:06implemented 'height_above_flotation' and 'relative_flotation' (ice thickness divided by flotation thickness). Needs a better name.Andy Aschwanden4+87-49
2016-04-20 03:05Started implementing new diagnostic "height_above_flotation"Andy Aschwanden2+111-1
2016-04-07 20:40Add option -float_kill_margin_only. See #340.Constantine Khroulev3+16-2
2016-04-07 18:41It turns out M_PI is not standard.Constantine Khroulev17+25-10
2016-04-07 17:59Update PISM's CMake scriptsConstantine Khroulev2+2-3
2016-04-01 20:37lat_bnds and lon_bnds don't need the "coordinates" attribute.Constantine Khroulev1+1-0
2016-04-01 17:25Cleanup. Boring.Constantine Khroulev2+21-23
2016-04-01 17:15No need to #define pi (math.h defines M_PI for us).Constantine Khroulev6+20-26
2016-04-01 16:48Clean up exactTestO.cConstantine Khroulev1+10-15
2016-04-01 01:26Enable the "limnsw" diagnostic.Constantine Khroulev1+1-0
2016-03-31 22:02Update run_stats in -extra_file, -save_file, -ts_file. See #330.Constantine Khroulev2+18-5
2016-03-31 18:20Move Pism_BUILD_PDFS to top-level CMakeLists.txt. See #342.Constantine Khroulev2+1-5
2016-03-31 17:05Iceberg remover: use the right mask.Constantine Khroulev1+12-8
2016-03-31 16:49Rename max_iterations_ssafd -> ssafd_max_iterations.Constantine Khroulev2+6-6
2016-03-31 16:45Cleanup.Constantine Khroulev1+4-3
2016-03-31 16:44SSAFD: Replace zero diagonal matrix entries with large beta.Constantine Khroulev2+21-6
2016-03-31 01:46Fix a compilation error introduced by 74c109c4. (Or I think I did.)Constantine Khroulev1+1-1
2016-03-31 01:12Remove duplicate message ("Got projection parameters...")Constantine Khroulev1+1-9
2016-03-30 22:18New diagnostics: {surface,basal}_mass_balance_average.Constantine Khroulev3+145-0
2016-03-30 21:46Mass continuity time-stepping: transport first, then sources.Constantine Khroulev1+47-30
2016-03-30 19:27Always do flux accounting.Constantine Khroulev5+43-100
2016-03-30 18:58Restartability test: compare model_state variables only.Constantine Khroulev1+1-1
2016-03-30 18:57Improve nccmp.py.Constantine Khroulev1+2-1
2016-03-30 18:57Fix typos in code that should save cumulative fluxes.Constantine Khroulev1+3-3
2016-03-30 18:19Cleanup in IceModel: always allocate fields used for accounting.Constantine Khroulev1+35-45
2016-03-30 17:30SSAFEM uses the new thickness threshold.Constantine Khroulev1+1-1
2016-03-30 15:40The SSA class owns the mask.Constantine Khroulev3+89-69
2016-03-29 17:30Improve a doxygen comment.Constantine Khroulev1+9-9
2016-03-29 17:24Use GeometryCalculator methods for all mask/surface updates.Constantine Khroulev6+102-138
2016-03-29 16:02More cleanup (boring).Constantine Khroulev33+144-175
2016-03-29 00:08Cleanup.Constantine Khroulev8+80-80
2016-03-28 23:58Remove ShallowStressBalance::set_sea_level_elevation()Constantine Khroulev9+61-55
2016-03-28 22:56GeometryCalculator: support custom ice-free thickness thresholds.Constantine Khroulev4+31-6
2016-03-28 22:23Use shared_ptr to manage diagnostics.Constantine Khroulev34+236-251
2016-03-28 21:41Add ice thickness thresholds for reporting and stress balance.Constantine Khroulev1+10-0
2016-03-28 21:39Fix initialization order: get projection info before computing cell areas.Constantine Khroulev1+17-1
2016-03-28 21:30Report projection info obtained from an input file.Constantine Khroulev1+6-0
2016-03-28 21:29Add projection information to the std-greenland example.Constantine Khroulev1+3-1
2016-03-28 21:16Global attributes stored by IceModel are for *output* files.Constantine Khroulev9+28-23
2016-03-28 21:08Better standard out messages about cell areas.Constantine Khroulev1+7-2
2016-03-28 19:49Update a regression test.Constantine Khroulev1+1-2
2016-03-28 19:43Remove special "marginal" ice treatment from enthSystemCtx.Constantine Khroulev3+15-22
2016-03-28 19:38Update the test G regression test. See #292.Constantine Khroulev1+8-8
2016-03-28 19:38Don't ignore horizontal enthalpy advection and strain heating. See #292.Constantine Khroulev4+5-33
2016-03-28 18:52Data member names. Very boring. Move right along.Constantine Khroulev34+1102-1101
2016-03-28 16:31Clean up PISM's build scripts.Constantine Khroulev1+40-47
2016-03-28 16:07Whitespace changes.Constantine Khroulev2+6-7
2016-03-28 06:28FFTW really is not optional (it could be, though).Constantine Khroulev1+1-1
2016-03-28 06:26No need to support CMake 2.8.4: we require 2.8.8.Constantine Khroulev1+2-14
2016-03-28 06:25We don't need to support Subversion.Constantine Khroulev1+0-16
2016-03-27 21:55Merge branch 'efischer/dev' into devConstantine Khroulev4+10-58
2016-03-27 20:36Fixing compilation errors and warnings on beauregard.Constantine Khroulev11+52-52
2016-03-26 17:29Fix the regexp.Constantine Khroulev1+1-1
2016-03-26 17:26Old CMake versions don't know about C++ standards.Constantine Khroulev1+8-0
2016-03-26 17:09Done merging efischer/160313-BuildFixes into dev.Constantine Khroulev5+106-149
2016-03-26 15:41Merge branch 'efischer/160313-BuildFixes' into devConstantine Khroulev26+1878-426
2016-03-25 20:15Merge branch 'ckhroulev/initmip' into devConstantine Khroulev20+666-262
2016-03-25 19:13Turn test/enthalpy/column.py into a regression test.Constantine Khroulev2+33-14
2016-03-25 18:40Fix verification test "enthalpy diffusion in the column".Constantine Khroulev1+14-9
2016-03-24 22:28Fix compilation warnings on beauregard.Constantine Khroulev12+58-55
2016-03-24 19:33Add -fPIC to C compiler flags.Constantine Khroulev1+6-1
2016-03-24 16:40Move enthalpy solver tests into a sub-directory.Constantine Khroulev4+505-505
2016-03-24 16:12Fix a typo.Constantine Khroulev1+1-1
2016-03-24 16:09Always build inverse modeling code.Constantine Khroulev1+4-7
2016-03-24 06:15Remove Blatter code from the dev branch.Constantine Khroulev11+0-2598
2016-03-24 06:09Merge branch 'ckhroulev/single-library' into devConstantine Khroulev11+60-122
2016-03-24 05:35Fix a bug in horizontal enthalpy advection code.Constantine Khroulev1+3-3
2016-03-23 22:39Merge branch 'ckhroulev/blatter' into devConstantine Khroulev109+5269-2291
2016-03-23 20:16Re-implement lateral till water diffusion as in Bueler and Brown, 2009.Constantine Khroulev3+110-20
2016-03-22 21:15Try using first-order FD to compute u_x, v_yConstantine Khroulev2+67-10
2016-03-22 21:05Remove duplicate entries from pism_config.cdl.Constantine Khroulev1+1-6
2016-03-22 16:25Cleanup in hydrology code.Constantine Khroulev3+163-156
2016-03-10 19:59Add the pism_ file name prefix to PISM's PDFs.Constantine Khroulev106+3317-3317
2016-03-02 20:19Diagnostics have access to a unit system.Constantine Khroulev1+15-30
2016-03-02 20:10More consistent use of m year-1 and m second-1 in unitsConstantine Khroulev49+219-219
2016-03-02 01:29Fix a compilation error with Pism_USE_PNETCDF=ON.Constantine Khroulev1+3-3
2016-03-01 21:44Fix a sign error bug introduced by 97d231d2.Constantine Khroulev4+29-29
2016-03-01 21:30Add "const" to some enthSystemCtx methods.Constantine Khroulev2+5-5
2016-03-01 18:19Include PISM revision in libpismutilConstantine Khroulev2+11-12
2016-02-29 22:23Use set(...) instead of string(CONCAT ...).Constantine Khroulev1+2-2
2016-02-25 21:29Add committer's name and date to the version string.Constantine Khroulev1+5-0
2016-02-22 05:48Minor addition: README.md in src/ suggests how to count lines.Ed Bueler1+11-0
2016-02-11 18:36Consistent naming of strain rate diagnostics.Constantine Khroulev3+9-9
2016-02-01 17:45Merge branch 'ckhroulev/ssafem-cfbc' into devConstantine Khroulev32+1829-989
2016-01-29 03:40Support PETSc master (as of Jan 28, 2016).Constantine Khroulev3+19-6
2016-01-22 07:05Add forgotten files (pism_utilities.{cc,hh}).Constantine Khroulev2+178-0
2016-01-21 01:08Merge branch 'ckhroulev/netcdf-string' into devConstantine Khroulev81+366-254
2016-01-20 21:22Merge branch 'ckhroulev/new-gl-mask-code' into devConstantine Khroulev10+440-253
2016-01-16 01:14Remove unused variables.Constantine Khroulev2+5-8
2016-01-16 01:11Remove chunk_dimensions(...) and io_type_size(...).Constantine Khroulev1+3-59
2016-01-16 01:07Don't use C99 features in enthalpy_converter.cConstantine Khroulev1+7-7
2016-01-16 00:50Remove an unused variable.Constantine Khroulev1+1-2
2016-01-16 00:33EnthalpyConverter::c() should have no arguments.Constantine Khroulev5+10-32
2016-01-16 00:27Replace division with multiplication in enthSystemCtx.Constantine Khroulev1+8-8
2016-01-16 01:53Fix #321 in the dev branch.Constantine Khroulev1+3-2
2016-01-14 23:40Remove SIA_Sliding, EISMINT II tests G and H, verification test E.Constantine Khroulev20+278-1233
2016-01-14 18:23Fix SIA_Sliding::surface_gradient_eta().Constantine Khroulev1+17-18
2016-01-14 18:20SIA_Sliding::init_impl(): print a message.Constantine Khroulev1+3-1
2016-01-14 18:19Better error messages on stress balance failure.Constantine Khroulev1+3-2
2016-01-14 08:57Merge branch 'ckhroulev/top-surface-neumann' into devConstantine Khroulev21+1269-261
2016-01-12 21:06Fix #328.Constantine Khroulev2+22-7
2016-01-12 19:44Fix a copy-and-paste goof.Constantine Khroulev1+1-1
2016-01-12 18:18Use EnthalpyConverter::is_temperate_relaxed() in diagnostics.Constantine Khroulev2+14-14
2016-01-12 18:16Add EnthalpyConverter::is_temperate_relaxed().Constantine Khroulev3+17-8
2016-01-12 07:50Add a link to Thomas's paper.Constantine Khroulev1+6-1
2016-01-12 02:04Reduce the time step length in the "heated slab" setup.Constantine Khroulev1+1-1
2015-10-28 09:42Fix assemble_R() for cold ice condtionsThomas Kleiner1+13-2
2015-10-28 09:22Fix enthalpy boundary condition schemeThomas Kleiner1+19-9
2016-01-12 00:24Add scripts for the "heated slab" setup.Constantine Khroulev3+169-0
2016-01-11 22:03Fix a typo.Constantine Khroulev1+1-1
2016-01-08 19:01Remove all mentions of dt_force. See #335.Constantine Khroulev5+8-24
2016-01-05 18:37Use standard C++ in the initialization of gpbld_constants gpbld.Constantine Khroulev2+11-13
2015-12-17 05:31Merge branch 'ckhroulev/netcdf-chunking' into devConstantine Khroulev9+163-13
2015-12-17 05:13Remove linear-in-T specific heat capacity.Constantine Khroulev3+0-86
2015-12-16 23:43Minor: The running CHANGES.md is easier to interpret if it explicitly says the new version number. Suggest adding a sentence or two like this in future releases.Ed Bueler1+1-1
2015-12-16 23:04Remove varcEnthalpyConverter.Constantine Khroulev12+116-257
2015-12-16 20:11Move cold case E(T) and T(E) functions into separate methods.Constantine Khroulev2+15-2
2015-12-16 19:32Merge branch 'ckhroulev/enthalpyconverter' into devConstantine Khroulev4+24-122
2015-12-16 06:13Merge branch 'ckhroulev/fast-gpbld' into devConstantine Khroulev42+2971-197
2015-11-26 23:41Whitespace cleanup.Constantine Khroulev1+6-6
2015-11-26 23:30Precompute dz in SIAFD::compute_I().Constantine Khroulev1+6-2
2015-11-24 22:20Minor cleanup in the quadratic interpolation code.Constantine Khroulev1+14-21
2015-11-24 00:15Replace "-" with "_" in snapshots written with -save_split.Constantine Khroulev1+1-1
2015-12-15 23:14Please let me know if you think this is a bad idea: Add the content of the page http://www.pism-docs.org/wiki/doku.php?id=citing_pism as ACKNOWLEDGE.md. This is prompted by a Craig Lingle suggestion; see email.Ed Bueler1+58-0
2015-11-19 22:30Support both GSL >= 1.15 and GSL >= 2.0.Constantine Khroulev1+1-1
2015-11-18 20:30Merge branch 'ckhroulev/new-pismo' into devConstantine Khroulev21+1085-689
2015-11-18 19:49Fix compilation errors and compiler warnings.Constantine Khroulev2+5-4
2015-11-18 07:07Add an option and config. parameter "force_to_thickness_start_time".Constantine Khroulev3+12-0
2015-11-18 00:15Add new option: -save_force_output_times.Constantine Khroulev4+22-1
2015-11-17 23:30Use requested times in -save_split snapshot file names.Constantine Khroulev1+1-1
2015-11-17 23:30pism::Time::date(): don't pad result with zeros.Constantine Khroulev2+2-2
2015-11-17 22:38Merge branch 'ckhroulev/flowlaws' into devConstantine Khroulev55+1936-1171
2015-11-12 23:17Update README about Greenland hydrology runs.Ed Bueler1+2-4
2015-11-12 23:11Update README about hydrology verification test.Ed Bueler1+14-2
2015-11-12 22:11Hide StringLogger methods that are not a part of its API.Constantine Khroulev1+1-0
2015-11-12 21:15Use >= in EnthalpyConverter::pressure.Constantine Khroulev1+1-1
2015-11-12 17:32Add MPI_Finalize() to a code example in the browser.Constantine Khroulev1+2-0
2015-11-12 06:36Add the ability to add hooks to pism::RuntimeError.Constantine Khroulev3+43-1
2015-11-09 17:06Print error messages to stderr.Constantine Khroulev20+183-155
2015-10-28 19:51Fix the doc. comment of enthSystemCtx::solveThisColumn().Constantine Khroulev1+19-12
2015-10-19 20:51Improve the doxygen browserConstantine Khroulev5+10-7
2015-10-15 19:20compute_start_and_count checks that we have at least 1 dimension.Constantine Khroulev1+6-0
2015-10-15 19:18define_spatial_variable checks that we have more than 1 dimension.Constantine Khroulev1+2-0
2015-10-15 19:17Make dof > 1 IceModelVec2s report when they are written.Constantine Khroulev1+4-0
2015-10-14 22:42Improve -verbose 3 reporting.Constantine Khroulev19+221-136
2015-10-13 17:49Fix a minor bug in grid initialization.Constantine Khroulev1+2-1
2015-10-12 18:14Hide Pism_USE_TR1 (most users don't need to set this any more).Constantine Khroulev1+1-0
2015-10-12 17:45Cleanup in EnthalpyConverter.Constantine Khroulev1+1-6
2015-10-12 16:45Move a Config::get_double() call out of a loop.Constantine Khroulev1+2-2
2015-10-08 22:34Add option -profile <filename> to pismr.Constantine Khroulev3+49-0
2015-10-08 20:48Print time-stamps when PISM starts and stops writing a backup.Constantine Khroulev1+11-2
2015-10-08 20:39Add the "backup" profiling event.Constantine Khroulev1+5-0
2015-10-08 18:28Fix a bug: use nc_put_vara... with the "yxz" storage order, not "xyz".Constantine Khroulev1+2-2
2015-09-24 03:07Code formatting.Constantine Khroulev39+116-120
2015-09-24 06:00Remove unneeded "virtual" keywords.Constantine Khroulev1+7-7
2015-09-24 05:59Imbalance -> difference.Constantine Khroulev1+3-3
2015-09-24 03:08Delete a PDF file. Just run "make" to re-generate.Constantine Khroulev1+0-0
2015-09-23 21:02Remove some raw pointers.Constantine Khroulev3+14-25
2015-09-23 17:45PISM decides if we need to #include <tr1/memory> or <memory>.Constantine Khroulev2+33-8
2015-09-09 20:25Initialize m_rank and stat in NC3File.Constantine Khroulev1+18-18
2015-09-10 00:19pism::Vars::keys(): include shared variables.Constantine Khroulev1+5-0
2015-08-19 18:45Adding "2dbig" to "-o_size". I need an option that writes all 2D variables but only 3D variables required for restarting.Andy Aschwanden24+48-26
2015-07-18 15:58Add a thermo ref. The melting-CTS version might be worth trying some day.Ed Bueler1+12-0
2015-07-15 15:26bugfix in interpolation weightsFlorian1+1-1
2015-07-16 17:14Indicate that "make" should run again before "make test" if CMake flag changes.Ed Bueler1+1-0
2015-07-11 21:28allow create_timeline.py to cooperate with create_greenland_epsg3413_grid.py. You first create the grid, then add the time dimension. Now you can add whatever you like (e.g. ocean forcing).Andy Aschwanden1+7-1
2015-07-08 17:50Add a forgotten SWIG interface file.Constantine Khroulev1+9-0
2015-07-08 03:56Fix a bug in pism::ocean::Constant (-shelf_base_melt_rate was ignored).Constantine Khroulev2+31-36
2015-07-08 03:07Add a regression test exposing a bug in pism::ocean::Constant.Constantine Khroulev2+43-0
2015-07-08 04:16Fix a bug in a regression test.Constantine Khroulev1+12-8
2015-06-30 17:07Combine make commands needed to re-build docs.Constantine Khroulev1+4-6
2015-06-30 16:47Mark query functions as const.Florian2+4-4
2015-06-30 16:43Update discharge_flux_cumulative units in the manual.Constantine Khroulev1+1-1
2015-06-30 16:03Fix units of discharge_flux_cumulative.Constantine Khroulev1+2-2
2015-06-30 15:57Fix #325.Constantine Khroulev1+13-5
2015-06-30 15:00Add (and use) pism::wall_clock_hours().Constantine Khroulev5+35-52
2015-06-30 13:58Fix yesterday's hasty fix.Constantine Khroulev1+1-1
2015-06-30 03:39Don't clear m_filename in an implementation method.Constantine Khroulev6+1-9
2015-06-29 22:54Specify units for ocean forcing in forcing manual.Andy Aschwanden1+1-1
2015-06-29 22:19Fix a typo (reported over Skype by @florianziemen).Constantine Khroulev1+1-1
2015-06-29 18:10Print eigen-calving max. dt info with -verbose 3 (instead of 2).Constantine Khroulev1+1-1
2015-06-29 17:48Stop at configuration stage if PETSc 3.6.0 was found.Constantine Khroulev1+8-0
2015-06-26 17:29Document -no_subgl_basal_melt in the User's Manual.Constantine Khroulev1+2-1
2015-06-25 19:32Add an option to disable sub-grid basal melt interpolation.Constantine Khroulev2+9-3
2015-06-22 21:17Fix output order check with -o_format hdf5, netcdf4_parallel, quilt.Constantine Khroulev1+2-2
2015-06-22 17:19Fix (and improve) error messages.Constantine Khroulev2+6-7
2015-06-20 15:43Update desired misfit in an inverse regression test.Constantine Khroulev1+2-1
2015-06-20 01:34Closes #324.Constantine Khroulev1+2-2
2015-06-20 01:30Fix compiler warnings.Constantine Khroulev2+4-2
2015-06-20 01:17Merge branch 'ckhroulev/transpose' into devConstantine Khroulev38+283-272
2015-06-12 18:12Fix a regression test failure caused by updating pisms defaults (see #323).Constantine Khroulev1+2-1
2015-06-11 19:51Try to support PETSc 3.6. See #322.Constantine Khroulev2+24-10
2015-06-11 18:57Fix PETSc 3.6 directory layout in FindPETSc.cmake.Constantine Khroulev1+3-3
2015-06-11 18:42Update CMake scripts.Constantine Khroulev4+122-81
2015-06-11 18:15Update MacPorts Git port name in the Installation Manual.Constantine Khroulev1+1-1
2015-06-11 05:39Nothing is automatic in GridParameters.Constantine Khroulev9+105-86
2015-06-12 16:46This completes the fix, I think: Update run script. Improve by separating horizontal and vertical resolution, allowing control of duration, and adjusting Lz accordint to experiment.Ed Bueler2+50-29
2015-06-12 16:44Update text about running EISMINT II experiments. @ckhroulev : Where are options in Table 47 read? Are they valid?Ed Bueler1+13-15
2015-06-12 16:38Fixing EISMINT II issues reported by T Kleiner. See #323. This is the core fix: we now default to periodic, but interpreting EISMINT II as non periodic is easy and makes sense. Also, cold-ice mode is not enough; we should set the flow law to Paterson-Budd.Ed Bueler1+2-0
2015-06-09 18:51Fix chunking settings in the HDF5 I/O backend.Constantine Khroulev4+24-7
2015-06-10 18:16Remove the code that tries to enable HDF5 alignment to stripe size.Constantine Khroulev1+3-47
2015-06-09 18:51Prepend HDF5 headers to the list of includes.Constantine Khroulev1+1-1
2015-06-02 04:32Update vnreport.py.Constantine Khroulev1+1-1
2015-06-01 23:26Update command-line options in vfnow.py.Constantine Khroulev1+3-3
2015-05-29 15:49Correct a reference.Ed Bueler1+11-9
2015-05-28 22:04Add a regression test for the code processing EPSG codes.Constantine Khroulev2+75-0
2015-05-28 20:00PISM executables should set non-zero exit status on error.Constantine Khroulev16+16-0
2015-05-28 19:09Add a regression test of the EPSG-to-CF conversion.Constantine Khroulev1+34-0
2015-05-28 18:45Better print_config and report_to_stdout formatting.Constantine Khroulev2+86-45
2015-05-28 00:00Add code converting +init=epsg:XXXX to CF-style metadata.Constantine Khroulev5+194-12
2015-05-28 16:16Add pism::StringLogger, a logger that prints to a string.Constantine Khroulev3+48-0
2015-05-26 17:20SeaRISE-Antarctica: Remove time dimension during pre-processing.Constantine Khroulev1+6-2
2015-05-22 18:21Fix a bug: process mapping and proj4 when -bootstrap is set.Constantine Khroulev1+20-21
2015-05-21 18:45Try to support PETSc < 3.5.Constantine Khroulev1+1-1
2015-05-21 17:03vnreport.py: print convergence rates; don't show figures with --save_figures.Constantine Khroulev1+4-1
2015-05-19 19:56Delete an outdated doxygen comment.Constantine Khroulev1+3-49
2015-05-18 20:07Resurrect "test/test_shelf/exactV.py".Constantine Khroulev3+5-6
2015-05-18 20:07Cleanup in IceEISModel.Constantine Khroulev1+2-4
2015-05-18 19:34Update test/test_ssafd/test_ssafd.pyConstantine Khroulev1+2-2
2015-05-18 17:00Add a link to the NetCDF bug NCF-152Constantine Khroulev1+3-1
2015-05-15 05:13Use a bash "here document" instead of a zillion of "echo".Constantine Khroulev1+55-54
2015-05-14 20:16Fix a bug in varcEnthalpyConverter, improve test coverage.Constantine Khroulev5+130-55
2015-05-14 00:33Add "make coverage_report" and "make coverage_reset".Constantine Khroulev1+12-0
2015-05-13 22:09Add a CMake option adding coverage-related compiler options.Constantine Khroulev1+8-0
2015-05-13 20:31Minor: shorten a headingEd Bueler1+2-2
2015-05-13 19:48Fix markdown formatting.Constantine Khroulev1+1-1
2015-05-13 19:42Add the release checklist.Constantine Khroulev1+31-0
2015-05-13 03:16Clarify in table of contents.Ed Bueler1+1-1
2015-05-12 22:49Fix a bug in the grid initialization code.Constantine Khroulev2+10-7
2015-05-12 21:44Fix a stupid typo.Constantine Khroulev1+1-1
2015-05-12 21:42Better error messages.Constantine Khroulev1+6-5
2015-05-12 21:21Update tables of diagnostic quantities, fix some typos.Constantine Khroulev1+54-39
2015-05-12 21:19Fix a couple of typos.Constantine Khroulev1+2-2
2015-05-12 19:22Update tables listing scalar time-series.Constantine Khroulev1+33-14
2015-05-12 18:50Fix labels (so that one of them does not read like "sausage").Constantine Khroulev1+4-4
2015-05-12 17:23Use an STL deque to manage delayed writes.Constantine Khroulev1+5-4
2015-05-12 15:41Fix #319.Constantine Khroulev2+8-2
2015-05-12 07:08Fix compiler warnings on beauregard.Constantine Khroulev1+4-4
2015-05-12 00:31Fix pnetcdf wrapper code.Constantine Khroulev1+2-2
2015-05-11 18:01PIO::put_1d_var() delays writingConstantine Khroulev2+48-11
2015-05-11 17:22Use PIMPL in PIO.Constantine Khroulev2+80-75
2015-05-11 16:00enddef/redef cleanup.Constantine Khroulev9+28-75
2015-05-12 06:52Avoid assignment to *this in constructor.Constantine Khroulev2+29-13
2015-05-12 01:33Slightly better error messages in IceGrid.Constantine Khroulev1+8-7
2015-05-11 23:32Update manual: pism_overrides is no longer written to output files.Constantine Khroulev1+1-1
2015-05-11 22:59Add -time_file_continue_run.Constantine Khroulev3+32-6
2015-05-11 21:35Cleanup in pism::Time and pism::Time_Calendar.Constantine Khroulev4+144-92
2015-05-10 14:44Update and edit (and maybe improve) the laboratory Xanthan gum validation subsection.Ed Bueler3+18-13
2015-05-09 16:03Minor: Maintenance on section 12.1 of User Manual. Figure 29 will need to be regenerated. Everything from section 12.2 is already checked and just fine.Ed Bueler2+9-9
2015-05-09 15:07Minor: fiddles with verification section tables.Ed Bueler1+10-8
2015-05-09 14:46Minor fixes to Getting Started section. Everything in that section runs for me. Very small changes in results only; because the figures are so close there is no reason to update just for this kind of user doc.Ed Bueler2+2-2
2015-05-09 03:20Correct date of 0.6 release.Ed Bueler1+2-2
2015-05-09 03:15Add NEWS-07.md at top of CHANGES.md. Make more consistent style. Add brief table of contents at top.Ed Bueler2+507-509
2015-05-08 23:46Fix a bug in quickjakosetup.shConstantine Khroulev1+3-2
2015-05-08 23:45Fix a bug in pismo.Constantine Khroulev1+3-1
2015-05-08 23:44Fix a bug in pism::Vars::remove() that broke pismo.Constantine Khroulev1+1-1
2015-05-08 22:17Fixing Tolly's last name. Added little blurb on pism_{config,overrides}Andy Aschwanden3+9-3
2015-05-08 21:41Fix #318.Constantine Khroulev1+1-1
2015-05-08 21:31More manual updates.Constantine Khroulev1+27-11
2015-05-08 21:31Fix C++ class name.Constantine Khroulev1+1-1
2015-05-08 19:26Fix documentation of "diagnostic" runs now that -y0 does exactly nothing.Constantine Khroulev1+5-7
2015-05-08 19:26Fix some C++ class names in the manual.Constantine Khroulev1+2-2
2015-05-08 19:25Clean up a doxygen comment.Constantine Khroulev1+7-5
2015-05-08 19:25Fix a minor bug in initialize_time().Constantine Khroulev1+1-0
2015-05-08 19:24Better "Configuration flags and parameters" browser page.Constantine Khroulev1+10-4
2015-05-08 19:10new stable version needs new manual cover pageAndy Aschwanden3+2-2
2015-05-08 15:49A bunch of minor fiddles with text as I do through the examples in section 1. (I should be done with section 1 by tomorrow May 9, but there is no reason to delay the release, I think.).Ed Bueler1+20-19
2015-05-08 05:59More minor edits.Constantine Khroulev1+28-24
2015-05-08 03:19Mostly really minor edits, except for reducing length of comment about -hydrology distributed boundary condition. Again: WOW.Ed Bueler1+11-23
2015-05-08 02:58This change is needed to compile. I have PISM_USE_PROJ4=0. @ckhroulev: please review cause I am not sure if there are other grids or dx,dy around.Ed Bueler1+1-1
2015-05-08 01:33Add a paragraph about "Points" and "PointsWithGHosts".Constantine Khroulev1+21-0
2015-05-08 01:06Fix a typo.Constantine Khroulev1+1-1
2015-05-08 01:02Improve Markdown formatting.Constantine Khroulev1+14-15
2015-05-08 00:40Add news (a detailed change log since v0.6).Constantine Khroulev1+495-0
2015-05-08 00:40Minor cleanup.Constantine Khroulev3+7-6
2015-05-06 21:44Config::get_double(name, desired_units)Constantine Khroulev22+125-58
2015-05-06 21:04All config. parameters can be set using a command-line options.Constantine Khroulev6+302-65
2015-05-06 20:00Fix a stupid bug in pism::ends_with().Constantine Khroulev2+2-2
2015-05-06 17:25Fix a doxygen warning.Constantine Khroulev1+1-0
2015-05-06 17:25Update config_doc.py.Constantine Khroulev1+17-4
2015-05-06 17:18Improve the way configuration parameters are printed at startup.Constantine Khroulev1+13-3
2015-05-06 16:15Automatically process scalar, boolean, and string parameters.Constantine Khroulev2+174-187
2015-05-06 01:12Fix compiler warnings on beauregard.Constantine Khroulev2+10-10
2015-05-06 00:56Automatically process keyword parameters.Constantine Khroulev2+71-49
2015-05-06 00:02Add a trivial SSA test.Constantine Khroulev2+66-7
2015-05-05 22:00Clean up IceGrid crud.Constantine Khroulev10+203-299
2015-05-05 20:53Remove unneeded Python code.Constantine Khroulev10+20-48
2015-05-05 19:55NCFile: set m_filename only if and open or create call succeeded.Constantine Khroulev8+28-43
2015-05-04 22:53More IceGrid work.Constantine Khroulev27+418-565
2015-05-04 22:11Create Parameters::ownership_ranges_from_options().Constantine Khroulev2+10-1
2015-05-04 22:09Add Context::size() and Context::rank().Constantine Khroulev2+14-0
2015-05-04 21:41create_coordinates() is a static function.Constantine Khroulev2+9-14
2015-05-04 21:40Add IceGrid::Paramaters::validate().Constantine Khroulev2+52-0
2015-05-04 21:03Commit to back up.Constantine Khroulev2+214-62
2015-05-04 20:53Fix spelling.Constantine Khroulev2+2-2
2015-05-04 20:21PIO::~PIO closes open files.Constantine Khroulev1+10-0
2015-05-04 18:40Minor code cleanup.Constantine Khroulev2+30-48
2015-05-02 02:43Fix compiler warnings on beauregard.Constantine Khroulev6+18-18
2015-05-02 02:28Add grid_info_from_bootstraping_file(). Work in progress.Constantine Khroulev2+74-0
2015-05-02 02:05Use IceGrid::Create() in more places.Constantine Khroulev14+177-114
2015-05-02 00:02Create IceGrid::set_ownership_ranges and ownership_ranges_from_options().Constantine Khroulev7+89-70
2015-05-01 22:30Remove IceGrid::compute_ownership_ranges().Constantine Khroulev2+9-16
2015-05-01 19:58Clean up grid_info::report().Constantine Khroulev1+17-16
2015-05-01 19:52Remove some private methods.Constantine Khroulev2+3-21
2015-05-01 19:50compute_nprocs() and dm_key() are static functions now.Constantine Khroulev2+20-25
2015-05-01 19:49Use PetscInt to store procs_[xy] again.Constantine Khroulev1+3-15
2015-05-01 19:16compute_horizontal_spacing is a static fn, clean up compute_nprocs.Constantine Khroulev2+23-18
2015-05-01 19:01Reduce code duplication in the computation of horizontal spacing.Constantine Khroulev2+23-24
2015-05-01 18:34Remove IceGrid::Create(ctx).Constantine Khroulev2+0-23
2015-05-01 18:09Move grid size check to IceModel.Constantine Khroulev2+21-19
2015-05-01 17:58Move computation of vertical levels into a static IceGrid method.Constantine Khroulev7+37-24
2015-05-01 17:26Reduce code duplication in IceGrid.Constantine Khroulev2+34-29
2015-05-01 15:44Start cleaning up IceGrid.Constantine Khroulev5+31-20
2015-05-01 00:44Create initialize_time() to move this code out of IceModel.Constantine Khroulev3+42-41
2015-04-30 23:01Improve test coverage a tiny bit.Constantine Khroulev1+10-0
2015-04-30 22:48Fix Logger verbosity.Constantine Khroulev1+1-1
2015-04-30 21:57Fix a doxygen comment.Constantine Khroulev1+1-1
2015-04-30 21:15More cleanup.Constantine Khroulev8+22-54
2015-04-30 19:03IceModel::config -> m_config.Constantine Khroulev22+234-230
2015-04-30 18:47Cleanup in IceModel.Constantine Khroulev17+117-111
2015-04-30 18:22IceGrid is managed using shared pointers.Constantine Khroulev228+1749-1737
2015-04-30 06:37Context::m_impl does not really need to be a shared_ptr.Constantine Khroulev2+6-1
2015-04-29 20:34Use pism::Logger.Constantine Khroulev115+637-562
2015-04-29 20:29Add Logger to the Context class.Constantine Khroulev5+46-25
2015-04-29 20:28Add pism::Logger.Constantine Khroulev4+173-0
2015-04-29 20:20Don't initialize time more than we have to.Constantine Khroulev4+1-6
2015-04-29 20:20Whitespace changes.Constantine Khroulev6+37-15
2015-04-28 19:36Create config_from_options(), fix test failures.Constantine Khroulev25+172-131
2015-04-28 17:38Remove IceGrid::convert().Constantine Khroulev40+154-144
2015-04-28 16:02Add some const-qualifiers.Constantine Khroulev8+13-14
2015-04-28 16:01Add Diagnostic::m_sys (a short-cut) and use it.Constantine Khroulev11+98-90
2015-04-28 15:41Use Component::m_sys.Constantine Khroulev26+50-53
2015-04-28 15:34Make the unit system available to Components and m_sys.Constantine Khroulev2+6-1
2015-04-28 15:34PISMComponent.hh does not include MaxTimestep.hhConstantine Khroulev29+30-4
2015-04-28 06:31Remove IceGrid::config().Constantine Khroulev50+115-119
2015-04-28 06:06Remove IceGrid::time().Constantine Khroulev47+156-154
2015-04-28 05:52Whitespace changes.Constantine Khroulev1+7-3
2015-04-28 05:48Remove Config::unit_system().Constantine Khroulev60+219-221
2015-04-28 01:33IceGrid does not own Profiling.Constantine Khroulev8+74-55
2015-04-28 00:22IceGrid does not own Time. Compiles but does not quite work.Constantine Khroulev73+358-356
2015-04-28 00:21Add const qualifiers to Time methods.Constantine Khroulev4+86-84
2015-04-27 22:37IceModel does not own an EnthalpyConverter.Constantine Khroulev13+85-50
2015-04-27 22:17Add IceModel::ctx().Constantine Khroulev2+9-0
2015-04-27 22:12IceModel constructor takes a Context::Ptr argument.Constantine Khroulev10+20-15
2015-04-27 22:03Forward-declare MaxTimestep in iceModel.hh.Constantine Khroulev1+2-1
2015-04-27 21:58IceGrid::config is a method, not a public data member.Constantine Khroulev78+374-367
2015-04-27 21:17Allocate Context in pism(r|s|v|o).Constantine Khroulev8+20-37
2015-04-27 20:25Move setting config. parameters from options into functionsConstantine Khroulev36+278-280
2015-04-27 18:03{IceModel,Component}::get_grid() -> ...::grid().Constantine Khroulev5+6-6
2015-04-27 17:59IceModel::grid -> IceModel::m_grid.Constantine Khroulev25+796-796
2015-04-27 17:19Minor fixes in tryLCbd.ccConstantine Khroulev1+42-51
2015-04-27 16:43Allocate Context in some regression tests.Constantine Khroulev9+36-87
2015-04-27 16:31Add Config::parameter_from_options().Constantine Khroulev2+31-0
2015-04-26 16:55Use the new flag (DEFAULT means "don't override user-set values").Constantine Khroulev1+27-3
2015-04-26 07:19Add one more Config::set_...() flag value.Constantine Khroulev2+17-7
2015-04-23 21:42Update installation manual.Constantine Khroulev1+49-55
2015-04-23 17:53Make it easier to manually set locations of optional PISM dependencies. See #316.Constantine Khroulev4+45-39
2015-04-23 15:31Fix compilation on beauregard.Constantine Khroulev1+1-2
2015-04-23 06:40Add a Python re-implementation of flowlaw_test.Constantine Khroulev1+30-0
2015-04-23 06:26Add automatically-generated docstrings. Much better!Constantine Khroulev1+1-0
2015-04-23 06:25Simplify flowlaw_test.ccConstantine Khroulev1+5-10
2015-04-23 05:31Profiling::end(): throw instead of aborting.Constantine Khroulev1+2-1
2015-04-23 05:28Use shared_ptr to wrap pism::Context.Constantine Khroulev1+1-0
2015-04-23 02:17Implement a Context "factory".Constantine Khroulev2+28-0
2015-04-23 02:11Use shared_ptr to manage enthalpy converters.Constantine Khroulev49+170-178
2015-04-23 01:27Start on an EC "factory".Constantine Khroulev2+19-1
2015-04-23 01:08Add Context::Ptr and Context::ConstPtr.Constantine Khroulev1+3-0
2015-04-23 00:19pism::Time does not need a MPI_Comm.Constantine Khroulev5+10-11
2015-04-22 23:56pism::Time uses Config::ConstPtr.Constantine Khroulev5+26-25
2015-04-22 23:35IceGrid, Component, and IceModel own shared ptrs to Config.Constantine Khroulev124+1011-1006
2015-04-22 21:11Add macros PISM_SHARED_PTR(TYPE) and PISM_WEAK_PTR(TYPE).Constantine Khroulev17+48-45
2015-04-22 19:52Fix a typo.Constantine Khroulev1+1-1
2015-04-22 19:31Slightly better initialization messages.Constantine Khroulev1+43-13
2015-04-22 19:30Require FFTW >= 3.1 (used to be >= 3.0).Constantine Khroulev2+5-1
2015-04-22 18:28Fix name clashes.Constantine Khroulev3+8-8
2015-04-22 17:00FlowLawFactory should not process options.Constantine Khroulev11+57-111
2015-04-22 06:26Flow laws don't need MPI communicators.Constantine Khroulev13+140-143
2015-04-21 18:40Add a Python-based test for pism::Context.Constantine Khroulev1+22-0
2015-04-21 17:57LocalMassBalance: add the m_ prefix.Constantine Khroulev2+13-13
2015-04-21 16:25Call init_timeseries() before begin_pointwise_access(). See #315.Constantine Khroulev1+2-2
2015-04-21 16:25Add assert(m_surface != NULL) macro calls.Constantine Khroulev1+9-0
2015-04-21 16:18Fix a bug in the "-surface pdd" initialization code. See #315.Constantine Khroulev2+4-2
2015-04-21 00:30Improving UDUNITS-2 wrappers.Constantine Khroulev48+224-220
2015-04-20 21:52Fix a BibTeX warning.Constantine Khroulev1+13-5
2015-04-20 21:52BibTeX does not support comments.Constantine Khroulev1+6-4
2015-04-20 21:32Document ftt_mask.Constantine Khroulev1+6-7
2015-04-20 19:30Create time_from_options() and use it in IceGrid.Constantine Khroulev8+32-22
2015-04-20 18:56init_calendar() -> calendar_from_options().Constantine Khroulev3+44-40
2015-04-20 18:13Ensure that age >= 0.0.Constantine Khroulev1+11-0
2015-04-20 18:13Minor cleanup.Constantine Khroulev1+2-4
2015-04-20 15:32Add profiling and prefix to pism::Context.Constantine Khroulev3+30-9
2015-04-17 22:07Wrap pism::Context.Constantine Khroulev1+8-6
2015-04-17 21:51Forward-declare UnitSystem in PIO.hhConstantine Khroulev1+2-1
2015-04-17 21:47Move most PISM-specific I/O code out of PIO.Constantine Khroulev70+1837-1763
2015-04-17 16:47Making sure that PIO does not need to own a UnitSystem.Constantine Khroulev26+101-60
2015-04-17 06:16More VariableMetadata cleanup.Constantine Khroulev5+23-17
2015-04-17 01:58Remove last references to IceGrid from SpatialVariableMetadata.Constantine Khroulev43+196-193
2015-04-16 21:21Move regrid(...) out of SpatialVariableMetadata.Constantine Khroulev5+80-72
2015-04-16 20:52Move write(...) out of SpatialVariableMetadata.Constantine Khroulev4+19-17
2015-04-16 20:46Move define(...) out of SpatialVariableMetadata.Constantine Khroulev27+82-81
2015-04-16 20:13Re-arranging spatial variable I/O.Constantine Khroulev31+192-157
2015-04-16 15:19Move use_mapped_io() and compute_start_and_count() out of PIO.Constantine Khroulev2+106-133
2015-04-13 19:45Remove an unneeded #include.Constantine Khroulev1+0-1
2015-04-13 19:45Wrap more classes using shared_ptr.Constantine Khroulev2+23-11
2015-04-11 05:58Add typedefs.Constantine Khroulev1+5-0
2015-04-10 18:45Use shared_ptr to wrap Config.Constantine Khroulev1+8-2
2015-04-10 21:16Use PIMPL in IceGrid.Constantine Khroulev17+274-264
2015-04-10 18:44Use PIMPL in the interface class "Config".Constantine Khroulev2+35-27
2015-04-10 01:44Working on Context. Compiles, needs testing.Constantine Khroulev2+60-61
2015-04-16 15:30Fix a bug in MISMIP.py (dev branch).Constantine Khroulev1+3-2
2015-04-14 16:12Add sys import. Apparently running nc2cdo.py has never failed before ...Ed Bueler1+1-0
2015-04-07 02:24added option --no_bounds to avoid calculating grid corners while still adding the mapping attributesAndy Aschwanden1+26-14
2015-04-03 15:53Minor: add newline to history string in nc2cdo.py.Ed Bueler1+1-1
2015-04-02 22:24Bed deformation: improve API. See #181.Constantine Khroulev8+91-37
2015-04-02 15:19Minor: add link.Ed Bueler1+1-1
2015-04-02 15:18Update INSTALL to cover content at "Quick installation summary" dokuwiki page. Implementing issue #314.Ed Bueler1+24-7
2015-04-02 14:42Fix compilation issues on beauregard.Constantine Khroulev2+2-2
2015-04-02 01:25Headers: system #includes go first.Constantine Khroulev21+33-39
2015-04-02 01:07Fix two typos.Constantine Khroulev2+2-2
2015-04-02 01:03Clean up regression test names.Constantine Khroulev1+35-35
2015-04-02 00:42Use "-bootstrap -i file.nc" instead of "-boot_file file.nc". See #308.Constantine Khroulev22+127-184
2015-04-01 23:27Replace -boot_file file.nc with -i file.nc -bootstrap in docs and scripts.Constantine Khroulev57+103-104
2015-04-01 22:53Minor bed deformation model cleanup.Constantine Khroulev2+13-9
2015-04-01 20:51Finish #include cleanup.Constantine Khroulev225+825-835
2015-04-01 19:08Reducing the number of -I flags needed to build code using PISM.Constantine Khroulev165+423-455
2015-04-01 17:16SurfaceModel::atmosphere -> m_atmosphere.Constantine Khroulev5+39-39
2015-04-01 05:12pism::SurfaceModel has a non-virtual interface.Constantine Khroulev34+135-138
2015-03-31 14:55Add small precision in PETSc install doc.Julien Seguinot1+7-1
2015-03-28 07:23Clobber warnings.Constantine Khroulev2+15-15
2015-03-28 06:38Broadcast pism_timestamp() from rank 0. See #151.Constantine Khroulev3+6-4
2015-03-28 06:05Remove two upper_bound() calls. See #170.Constantine Khroulev1+26-29
2015-03-27 23:03Address #304.Constantine Khroulev2+44-1
2015-03-27 21:41Add coverage support to pismi.py. See #294.Constantine Khroulev3+28-5
2015-03-27 19:08Fix sia_forward.py and ssa_forward.py.Constantine Khroulev2+29-25
2015-03-27 16:47Cleanup: remove more unused typemaps.Constantine Khroulev3+2-53
2015-03-27 06:55Remove SWIG typemaps for IO_Mode and IO_Type.Constantine Khroulev2+6-35
2015-03-27 05:07TerminationReason is a return value instead of an output argument.Constantine Khroulev13+97-124
2015-03-26 22:58exactI() and exactJ() don't need to use output arguments.Constantine Khroulev11+81-88
2015-03-26 22:06Fix handling of used and unused parameters.Constantine Khroulev5+24-5
2015-03-26 19:41Rename NCVariable and derived classes.Constantine Khroulev81+1310-1310
2015-03-26 18:38PISMVars.hh does not need to include iceModelVec.hh now.Constantine Khroulev4+44-30
2015-03-26 18:25Create pism_memory.hh and move a bunch if #ifdefs there.Constantine Khroulev18+96-163
2015-03-26 17:09Move pism::Profiling to a separate file.Constantine Khroulev32+155-71
2015-03-26 16:41Minor cleanup (include petscsys.h instead of petscdmda.h).Constantine Khroulev1+2-5
2015-03-26 16:41Move function implementations out of pism_const.hhConstantine Khroulev16+69-46
2015-03-24 21:40De-couple config interface from the default implementation.Constantine Khroulev153+1572-1241
2015-03-24 21:34Config::com() is not used anywhere.Constantine Khroulev3+1-7
2015-03-23 19:45Test that omega == EC.water_fraction(EC.enthalpy(T, omega, P)). See #272.Constantine Khroulev1+13-4
2015-03-13 23:08Move units to separate attributes in the .cdl file.Constantine Khroulev2+480-278
2015-03-12 19:53Rename Config::get() to Config::get_double().Constantine Khroulev113+558-560
2015-03-12 19:38"make retest" spits out test output (for debugging).Constantine Khroulev1+1-1
2015-03-12 16:14Clean up #includes.Constantine Khroulev87+221-97
2015-03-11 23:21Include <cassert> instead of <assert.h>.Constantine Khroulev13+13-13
2015-03-11 23:18Add include dependency graph to the browser.Constantine Khroulev1+3-3
2015-03-11 23:07Add the -use_Kirchhoff_law option and the corresponding config. flag.Constantine Khroulev5+11-7
2015-03-11 23:06Automatically re-build the list of config. parameters.Constantine Khroulev1+1-0
2015-03-11 21:35Wrap pism::petsc::Vec.Constantine Khroulev1+13-0
2015-03-11 21:32Add more files to .gitignore.Constantine Khroulev1+4-1
2015-03-11 21:32Improve test coverage, fix a minor bug.Constantine Khroulev2+32-4
2015-03-11 19:06Fix a bug in listener.py.Constantine Khroulev1+1-1
2015-03-11 17:47Don't hide information (i.e. a better error message).Constantine Khroulev1+9-6
2015-03-11 17:44Remove commented out codeConstantine Khroulev1+0-18
2015-03-11 16:04Minor cleanup (using pylint).Constantine Khroulev6+11-6
2015-03-21 03:49How can a publisher so monumentally F-up the online description of the article, so that the list of authors is reversed from, and the year different from, the print version? Failure of OCR?Ed Bueler1+7-6
2015-03-11 03:00Squash compiler warnings.Constantine Khroulev2+16-16
2015-03-11 02:41Whitespace changes: run autopep8 on src, test, and util.Constantine Khroulev29+1119-995
2015-03-11 02:38Whitespace changes: run "autopep8 -i -r examples/"Constantine Khroulev47+3422-3344
2015-03-11 02:29Use LinearInterpolation in LocalInterpCtx.Constantine Khroulev1+46-126
2015-03-11 01:54Fix a subtle bug in BedSmoother.Constantine Khroulev1+7-0
2015-03-10 21:14Add the LinearInterpolation class and a test.Constantine Khroulev5+255-0
2015-03-10 17:27nccmp.py: treat denormalized numbers as zero.Constantine Khroulev1+13-4
2015-03-10 16:37IceGrid: we read x (y,z), so we can just use [xyz].size().Constantine Khroulev1+3-3
2015-03-05 19:29Close the file if regridding failed.Constantine Khroulev1+8-1
2015-03-05 19:27Improve error messages.Constantine Khroulev1+21-13
2015-03-05 16:38Tweak the Sphinx docs.Constantine Khroulev10+2-19
2015-03-04 20:05Python code cleanup.Constantine Khroulev20+139-141
2015-03-04 19:59Much better pylint support.Constantine Khroulev2+19-5
2015-03-04 02:21Add pylintrc and fix some errors and warnings.Constantine Khroulev8+416-32
2015-03-04 01:04Revert some changes made by autopep8.Constantine Khroulev2+4-4
2015-03-04 00:55Run autopep8 -a.Constantine Khroulev6+24-24
2015-03-04 00:49Run autopep8 on site-packages/PISM/.Constantine Khroulev15+3149-3009
2015-03-04 00:18Rename c_from_T(T) -> c(T), document recent changes.Constantine Khroulev4+101-11
2015-03-04 00:18Fix links in the browser.Constantine Khroulev1+12-14
2015-03-03 23:12Use EnthalpyConverter::L() in imEnthalpy.ccConstantine Khroulev1+18-8
2015-03-03 23:10Use cmbright in the browser.Constantine Khroulev1+8-0
2015-03-03 22:51Improve doxygen browser CSS.Constantine Khroulev2+63-9
2015-03-03 20:49Add EnthalpyConverter::L(T_melting).Constantine Khroulev4+103-146
2015-03-03 17:31Doxygen: return to EXTRACT_PRIVATE=YES.Constantine Khroulev1+1-1
2015-03-03 17:26Remove an unneeded #include.Constantine Khroulev1+0-2
2015-03-02 21:43..._flux diagnostics: state flux sign in a variable attributeConstantine Khroulev1+22-5
2015-02-28 23:48Host for netcdf4-python has moved.Ed Bueler1+1-1
2015-02-26 17:29Add a KirchhoffEnthalpyConverter regression test.Constantine Khroulev1+25-0
2015-02-19 17:08Minor: Maintenance of comments only.Ed Bueler1+14-16
2015-02-18 21:02Add a few more references, of greater or lesser prominence.Ed Bueler1+31-0
2015-02-18 17:09Address #137.Constantine Khroulev1+6-0
2015-02-18 15:37This ancient document should probably be removed completely. (I looked at it because I could not figure out how to build the browser.)Ed Bueler1+5-45
2015-02-18 15:36Add prominent recent PISM publications to .bib.Ed Bueler1+58-2
2015-02-18 00:26fixup! Add a workaround for comparing arrays of strings.Constantine Khroulev1+3-2
2015-02-18 00:10Add a workaround for comparing arrays of strings.Constantine Khroulev1+7-1
2015-02-17 20:14-surface ...,forcing: mention that floating ice is not affected.Constantine Khroulev1+11-7
2015-02-17 20:14-surface ...,forcing: stop if ftt_mask was not found.Constantine Khroulev1+3-0
2015-02-13 23:52NCVariable and NCSpatialVariable: API cleanup.Constantine Khroulev4+17-4
2015-02-12 23:23Make sure that we use PetscInt only if we have to (and add comments).Constantine Khroulev6+11-2
2015-02-12 23:13PISM compiles with PETSc 3.3 and 3.4 again.Constantine Khroulev2+6-3
2015-02-12 00:22Clobber compiler warnings.Constantine Khroulev2+33-33
2015-02-11 23:54Use GSL binary search functions in various places.Constantine Khroulev5+39-33
2015-02-11 22:57IceGrid::kBelowHeight() throws instead of aborting.Constantine Khroulev8+306-255
2015-02-11 22:21IceModelVec::AccessList: make sure that the destructor does not throw.Constantine Khroulev1+6-2
2015-02-11 22:11Fix a possible (but very unlikely) memory leak.Constantine Khroulev1+2-0
2015-02-11 21:46Add split() for splitting comma-separated lists (and some tests).Constantine Khroulev3+27-0
2015-02-11 21:46Don't pass std::string by value.Constantine Khroulev12+33-23
2015-02-11 19:59Add enthalpy converter regression (and sanity) tests.Constantine Khroulev3+164-2
2015-02-11 19:29Add enthalpy_liquid() to the interface.Constantine Khroulev3+10-6
2015-02-11 04:27Implement KirchhoffEnthalpyConverter (to be used in Glint2).Constantine Khroulev4+125-14
2015-02-11 03:47Replace enthalpy_interval() with enthalpy_liquid().Constantine Khroulev3+17-30
2015-02-10 23:59m_-ify EnthalpyConverter.Constantine Khroulev4+54-44
2015-02-10 22:48P.A. temperature does not need a virtual implementationConstantine Khroulev2+0-5
2015-02-10 22:41Move c_from_T_impl(), format a formula.Constantine Khroulev2+12-13
2015-02-10 22:32Remove ColdEnthalpyConverter method implementations from the header.Constantine Khroulev2+52-33
2015-02-10 21:32EnthalpyConverter has a non-virtual interface.Constantine Khroulev4+126-69
2015-02-10 16:25Clean up EnthalpyConverter API.Constantine Khroulev26+168-222
2015-02-10 16:22Add a getter for the EnthalpyConverter.Constantine Khroulev1+4-0
2015-02-10 07:29More work on PIO, NCVariable, and NCSpatialVariable.Constantine Khroulev4+53-48
2015-02-10 07:09Use metadata().set_string("glaciological_units", ...) instead of set_glaciological_units().Constantine Khroulev56+126-152
2015-02-10 06:20Use set_string("units", ...) instead of set_units(...).Constantine Khroulev40+202-202
2015-02-10 06:05Remove NCVariable::m_units and m_glaciological_units.Constantine Khroulev2+21-44
2015-02-10 05:44Remove PIO::inq_units() and NCVariable::get_units().Constantine Khroulev5+43-79
2015-02-10 04:05Remove an unneeded include.Constantine Khroulev1+0-1
2015-02-10 04:05Timeseries: add the m_ prefix.Constantine Khroulev2+127-127
2015-02-10 04:05Whitespace changes.Constantine Khroulev1+2-2
2015-02-10 04:04Better variable names.Constantine Khroulev1+21-18
2015-02-10 04:03Add and use NCVariable::get_unit_system().Constantine Khroulev2+17-14
2015-02-10 04:02Use UnitConverter instead of convert_doubles().Constantine Khroulev2+10-10
2015-02-10 04:01Add new UnitConverter constructor.Constantine Khroulev2+17-0
2015-02-10 03:57Change convert_vec() signature and remove convert_doubles().Constantine Khroulev5+20-40
2015-02-10 02:31Improve IceModelVec APIConstantine Khroulev18+99-147
2015-02-10 02:18Remove an unneeded line from pism_IceModelVec.i.Constantine Khroulev1+0-2
2015-02-09 20:31Fix the Tikhonov LMVM regression test (i.e. fix iteration listeners).Constantine Khroulev9+168-117
2015-02-08 04:49Move one line in pism_inverse.i.Constantine Khroulev1+1-1
2015-02-07 18:28Wrap a doxygen comment.Constantine Khroulev1+8-5
2015-02-07 07:53Rename a variable (for clarity).Constantine Khroulev1+12-11
2015-02-07 07:53Whitespace changes.Constantine Khroulev1+2-2
2015-02-07 01:46Improve error propagation between C++ and Python callbacks.Constantine Khroulev3+31-21
2015-02-07 00:36Wrap some lines.Constantine Khroulev1+6-2
2015-02-06 22:19IceModelVec::norm() can be implemented in terms of norm_all().Constantine Khroulev1+11-39
2015-02-06 21:53Remove IceModelVec::copy_to() and IceModelVec::nlevels()Constantine Khroulev37+97-108
2015-02-09 18:03Clean up funding sources in README.md. Closes #298.Ed Bueler1+3-2
2015-02-09 17:02Add a little script that shows hydrology time series.Ed Bueler3+77-4
2015-02-09 02:10Usermanual: Trying to have consistent spacing between numbers and units, work on the "Getting Started" section.Andy Aschwanden1+21-21
2015-02-08 16:06Minor: "saving spatial time-series to" stdout line lost because of conditional scope, probably.Ed Bueler1+6-4
2015-02-06 20:20These fixes to basemapfigs.py makes it produce figures like in chapter 2 of current User Manual.Ed Bueler1+9-8
2015-02-06 18:11Finally! Merge branch 'new-error-handling' into devConstantine Khroulev447+34591-33334
2015-01-27 21:11Merge branch 'add-hydrology-timeseries' into devEd Bueler6+282-53
2015-01-27 19:27PETSc website url update.Joseph H Kennedy2+7-7
2015-01-26 04:57This came from edrun/ in https://github.com/pism/green-hydro. I am committed to maintaining it here for stable0.7 and onward.Ed Bueler8+398-3
2015-01-25 16:34Pressure is set to overburden in grounded-ice areas with empty subglacial aquifer.Ed Bueler1+1-14
2015-01-22 21:37Add a wordy comment on a highly-technical detail: what pressure to assign to grounded icy areas where W=0 in the subglacial aquifer. No functional change; just a comment.Ed Bueler1+9-1
2015-01-24 03:36This script no longer uses correct options. See https://github.com/pism/green-hydro for (more) live material.Ed Bueler1+0-51
2015-01-23 00:27Fix an unreported bug in scalar time-series reporting.Constantine Khroulev1+9-9
2015-01-21 08:00Fix a memory leak in the dev branch. Port to 0.6.Constantine Khroulev1+16-13
2015-01-17 23:50Changes to comments only: add FIXMEs re issue #256. Minor clean up.Ed Bueler2+4-2
2015-01-17 23:12Comments only: clean up comments on Hydrology and derived classes.Ed Bueler2+24-19
2015-01-12 15:49Minor: update meaning of 7km grid (=7.5km now). Note searise-antarctica example works well.Ed Bueler1+1-1
2015-01-11 18:14Minor maintenance: refactoring scripts a bit to make them easier to maintain.Ed Bueler5+266-211
2015-01-11 17:22Minor maintenance on searise-antarctica: clean README.md.Ed Bueler1+3-4
2015-01-11 17:21Minor maintenance on eigencalving report-to-stdout code: remove non-parallel i0,j0 and clean up. (Note the Ross prognostic example works great.)Ed Bueler1+8-11
2015-01-09 03:30more maintenance on labgum example, esp redo of grid now that both dimensions are periodic ... but results are still not along dataEd Bueler4+45-41
2015-01-08 04:25maintenance on lab gum example; it is not very close to data right now, for unknown reasons ...Ed Bueler3+25-24
2015-01-08 02:28Revert "PISMS and PISMV use the 360_day calendar."Constantine Khroulev2+4-4
2015-01-07 23:07Rename tau3 -> age3.Constantine Khroulev6+29-29
2015-01-07 23:02PISMS and PISMV use the 360_day calendar.Constantine Khroulev2+4-4
2014-12-24 21:17minor: add performance comment clarity on grid sequencingEd Bueler1+1-1
2014-12-17 23:44Fix siafd_test in the dev branch.Constantine Khroulev1+1-1
2014-12-12 16:09minor: one character typoEd Bueler1+1-1
2014-12-11 23:00Add "make retest" to re-run tests that failed.Constantine Khroulev1+7-0
2014-12-12 07:18Replace test_00.sh. See #305.Constantine Khroulev3+22-56
2014-12-09 19:23Fix a minor bug in the vertical interpolation code (in dev).Constantine Khroulev1+7-2
2014-11-22 00:12Add "initialization" and "extra_file reporting" profiling events.Constantine Khroulev2+9-0
2014-11-21 22:55Add "flux" to variables saved with "-o_size medium".Constantine Khroulev1+1-1
2014-11-21 22:55Add "flux" to the list of 2D vector diagnostics in the manual.Constantine Khroulev1+1-0
2014-11-21 22:46Improve the "flux" diagnostic.Constantine Khroulev1+78-69
2014-11-21 21:02Add the "flux" diagnostic (uflux, vflux).Constantine Khroulev2+74-2
2014-11-19 21:34Reduce code duplication.Constantine Khroulev7+173-181
2014-11-18 17:16Move PS_EISMINTII from stc/eismint to src/coupler.Constantine Khroulev5+61-61
2014-11-17 23:05No need for temp. storage in IceModel::putTempAtDepth().Constantine Khroulev1+6-5
2014-11-15 04:30Clarify Lz in docsJoseph H Kennedy1+1-1
2014-11-14 23:31Remove automatic vertical grid extension.Constantine Khroulev16+9-334
2014-11-10 21:55Silence some pedantic compiler warnings.Constantine Khroulev4+4-3
2014-11-10 21:36Avoid a possible but harmless division by zero (in dev).Constantine Khroulev1+1-1
2014-11-10 17:52Fix Python wrapper docs (in dev).Constantine Khroulev1+6-1
2014-11-03 15:31Made case of file match others. (Someone has been using a case-insensitive OS again ...)Ed Bueler5+90-90
2014-10-29 18:54Fix a bug in -ocean ...,delta_SMB.Constantine Khroulev1+2-2
2014-10-31 22:48Add m_ prefix to BedThermalUnit data members.Constantine Khroulev4+34-34
2014-10-31 22:45Realized that we don't need IceModelVec3BTU.Constantine Khroulev4+27-106
2014-10-31 16:30beauregard has a really pedantic compiler:Ed Bueler1+1-1
2014-10-31 15:52these two refs have now appeared in printEd Bueler1+16-10
2014-10-30 19:45fixed units of climatic_mass_balance in util/flowlineslab.py (m/yr to kg m-2 s-1)Andy Aschwanden1+2-2
2014-10-19 21:35Silence a compiler warning.Constantine Khroulev1+5-2
2014-10-17 17:39another refEd Bueler1+8-0
2014-10-15 16:06Clean up the TR1 business a little bit.Constantine Khroulev8+6-46
2014-10-15 05:49Fix #299 by removing a bunch of old iffy code.Constantine Khroulev1+19-44
2014-10-14 17:35pism::NCFile: create a non-virtual interfaceConstantine Khroulev16+655-388
2014-10-14 16:05pism::NCFile: add the m_ prefix to data members.Constantine Khroulev10+332-332
2014-10-13 22:10Fix IceModelVec::check_array_indices() (now it works for 3D fields too).Constantine Khroulev1+6-1
2014-10-13 22:09Silence compiler warnings.Constantine Khroulev2+12-13
2014-10-13 20:16Replace -Winline with -Wno-inline to avoid bogus warnings.Constantine Khroulev1+1-1
2014-10-13 20:12Improve interpolation weight computation code.Constantine Khroulev2+45-22
2014-10-10 21:30Clean up the Mahaffy surface gradient code.Constantine Khroulev1+10-14
2014-10-10 19:52No need to check the -periodicity option by hand.Constantine Khroulev2+3-20
2014-10-10 16:44Tweak profiling calls.Constantine Khroulev2+14-8
2014-10-10 16:431 reduction instead of 8 in massContExplicitStep()Constantine Khroulev1+20-8
2014-10-10 16:24Add an optional "count" argument to Global{Sum,Min,Max}.Constantine Khroulev25+135-135
2014-10-10 06:11Shove profiling stuff into IceGrid for now.Constantine Khroulev6+49-35
2014-10-10 04:30Address Andy's compilation issue.Constantine Khroulev1+3-3
2014-10-10 03:39Create a DM with dof=1, width=0 and store in IceGrid.Constantine Khroulev3+14-5
2014-10-10 00:53Add some profiling events to IceModel.Constantine Khroulev6+121-16
2014-10-09 01:12Remove last DMLocalToGlobalBegin/End calls. Port to 0.6.Constantine Khroulev8+93-92
2014-10-09 00:35Generalize IceModelVec2::{get,set}_component().Constantine Khroulev2+30-27
2014-10-08 23:54Silence SWIG warnings.Constantine Khroulev1+3-0
2014-10-08 23:52Syntactic sugar (add PISMDM::operator DM()).Constantine Khroulev14+114-108
2014-10-08 23:03Improve IceModelVec2::{get,set}_component(...)Constantine Khroulev2+37-32
2014-10-08 20:11Rewrite put_on_proc0/get_from_proc0 IceModelVec2S methods.Constantine Khroulev8+133-130
2014-10-08 17:44Require PETSc 3.5 for TAO-based inversion tools.Constantine Khroulev13+92-419
2014-10-07 21:30Use IceModelVec2V to store solution and RHS in SSAFD.Constantine Khroulev9+61-77
2014-10-07 23:44Added missing characterAndy Aschwanden1+1-1
2014-10-07 23:38Updated funding sources in README. Port to 0.6.Andy Aschwanden1+1-1
2014-10-07 19:34Fix a typo introduced in 6f1fe04.Constantine Khroulev1+1-1
2014-10-07 18:36Commit ff45421 broke IcebergRemover. This should fix it.Constantine Khroulev2+3-6
2014-10-07 18:32Add IceModelVec::get_dm()Constantine Khroulev2+6-1
2014-10-07 18:03Cleanup.Constantine Khroulev1+0-3
2014-10-07 17:54thksmooth and schoofs_theta diagnostics don't need ghosts.Constantine Khroulev1+2-4
2014-10-07 18:45v.get_stencil_width() == 0 if v has no ghosts.Constantine Khroulev2+9-1
2014-10-07 00:24Basal heat flux does not need to be ghosted.Constantine Khroulev1+1-1
2014-10-07 00:13Add comments to benign uses of IceModelVec::copy_to_vec()Constantine Khroulev2+2-2
2014-10-06 23:52Simplify IceModelVec::update_ghosts(destination)Constantine Khroulev1+3-10
2014-10-06 23:15Use a "global" Vec in SIAFD_topgsmooth::compute().Constantine Khroulev1+1-2
2014-10-06 23:13Add an assertion (PISMRoutingHydrology::V has to have ghosts).Constantine Khroulev1+3-1
2014-10-06 23:08SSA::m_velocity_old -> SSAFD::m_velocity_old.Constantine Khroulev4+7-5
2014-10-06 23:03Remove IceModelVec2S::view_matlab().Constantine Khroulev3+0-48
2014-10-06 18:47Remove duplicate code in IcebergRemover.Constantine Khroulev3+48-85
2014-10-06 17:44Remove duplicate code in PBLingleClark.Constantine Khroulev2+6-62
2014-10-04 00:05Fix a typo, use a *fixed* large negative number.Constantine Khroulev2+35-15
2014-10-03 22:54The last PetscInt (I hope).Constantine Khroulev1+1-1
2014-10-03 22:51One more PetscInt.Constantine Khroulev1+1-1
2014-10-03 21:58Use "PetscInt" instead of "int" where needed.Constantine Khroulev12+26-21
2014-10-03 22:00Use IceModelVec::copy_to_vec() in one more place.Constantine Khroulev1+1-5
2014-10-03 03:54another refEd Bueler1+14-0
2014-10-02 16:29Use IceModelVec::copy_to_vec in more places.Constantine Khroulev2+4-10
2014-10-01 23:57IceModelVec::v -> IceModelVec::m_vConstantine Khroulev7+103-103
2014-10-01 21:39Add the -dump option to bedrough_test.Constantine Khroulev1+8-0
2014-10-01 19:49Add "shelfbtemp" and "shelfbmassflux" diagnostics to "-ocean th".Constantine Khroulev2+62-7
2014-09-26 21:09Work on supporting PETSc 3.5.Constantine Khroulev24+353-87
2014-09-25 00:53Don't look for TAO if PETSc is recent enough (>= 3.5.0).Constantine Khroulev1+14-2
2014-09-25 00:52Don't recommend trying to match compilers.Constantine Khroulev1+0-12
2014-09-24 23:14Add some code for single-column enthalpy solver tests.Constantine Khroulev1+126-0
2014-09-24 23:12Start cleaning up documentation of enthSystemCtx.Constantine Khroulev2+96-46
2014-09-24 23:11Change the IceModelVec3::setValColumnPL() signature.Constantine Khroulev5+6-7
2014-09-23 22:59Create a Python wrapper for enthSystemCtx.Constantine Khroulev2+7-1
2014-09-23 22:57Cosmetic changes in a regression test.Constantine Khroulev1+6-6
2014-09-23 22:56Update links for C++ classes in BOMBPROOF docs.Constantine Khroulev1+7-7
2014-09-23 22:55Define commands \dx, \dy, \dz, \dt in doxygen.Constantine Khroulev1+4-0
2014-09-23 22:55Create helper functions for creating more PISM fields.Constantine Khroulev1+35-0
2014-09-23 22:54Ensure that storage for a column solution is allocated.Constantine Khroulev1+2-0
2014-09-23 22:54Specify units in a doxygen comment.Constantine Khroulev1+2-0
2014-09-22 23:17Pass EnthalpyConverter to enthSystem by const reference.Constantine Khroulev3+55-46
2014-09-26 23:45Make "const IceModelVec" usable.Constantine Khroulev17+289-201
2014-09-29 15:26add a ref, and fiddle with themEd Bueler1+21-13
2014-09-26 12:52fix one refEd Bueler1+1-1
2014-09-25 20:38add one reference: Karniadakis and SherwinEd Bueler1+9-0
2014-09-23 16:05fixed a minor bug in ocean coupler, fix should be considered for stable0.5 tooTorsten Albrecht1+2-2
2014-09-21 06:16Try to support both old and new (>= 3.0) CMake versions.Constantine Khroulev1+7-1
2014-09-19 23:50Turn "Mask" in to a namespace (instead of a class).Constantine Khroulev8+109-130
2014-09-19 22:29Silence CMake warnings (for now).Constantine Khroulev2+3-0
2014-09-19 21:20Use the same "wide stencil" everywhere.Constantine Khroulev34+239-143
2014-09-18 19:48Fix a bug I introduced in the previous commit.Constantine Khroulev2+12-2
2014-09-17 22:51Remove unused IceModelVec3 methods.Constantine Khroulev2+4-44
2014-09-17 19:19Switch to IceModelVec::AccessList.Constantine Khroulev89+1630-2431
2014-09-17 19:18Disable the SWIG nested class warning.Constantine Khroulev1+2-0
2014-09-17 19:16Add IceModelVec::AccessList.Constantine Khroulev2+35-1
2014-09-16 19:56Switch to using Points and PointsWithGhosts.Constantine Khroulev86+4940-4958
2014-09-16 19:35Add pism::Points and pism::PointsWithGhosts.Constantine Khroulev1+66-5
2014-09-15 19:55Another fix for #297.Constantine Khroulev2+15-25
2014-09-15 19:43Fixes and closes #297. (Also fix a typo in flowlaws.cc)Ed Bueler2+3-1
2014-09-12 23:08Add more tests.Constantine Khroulev1+96-2
2014-09-12 22:25Fix a bug in PISM.model.ModelVecs.rename().Constantine Khroulev1+18-1
2014-09-12 20:01Add test/nosetests.py to "make test".Constantine Khroulev2+10-0
2014-09-12 16:59Add some superficial regression tests. See #294.Constantine Khroulev1+313-0
2014-09-12 16:57Update some Python scripts.Constantine Khroulev4+72-167
2014-09-12 16:55Improve the EnthalpyConverter Python wrapper.Constantine Khroulev1+3-3
2014-09-11 21:51Fix a bug in PISM.AlgorithmFailureException.Constantine Khroulev1+2-2
2014-09-11 14:18Indentation cleanup.Constantine Khroulev1+110-110
2014-09-11 06:31Replace "," -> ", ".Constantine Khroulev14+505-505
2014-09-11 05:58Remove trailing whitespace.Constantine Khroulev6+166-166
2014-09-11 05:55Remove trailing semicolons.Constantine Khroulev7+145-145
2014-09-10 18:26Add vertical grid spacing to run_stats.Constantine Khroulev3+23-0
2014-09-10 16:42Update a doxygen comment.Constantine Khroulev1+0-1
2014-09-10 16:41Add the ocean_pik_melt_factor config parameter. See #295.Constantine Khroulev4+32-18
2014-09-09 22:42Update some inversion-related Python scripts. Port to 0.6.Constantine Khroulev4+125-17
2014-09-08 18:58Add the "source" (PISM revision) attribute to run_stats.Constantine Khroulev1+1-0
2014-09-08 17:53Save 'grid_dx_meters' and 'grid_dy_meters' to run_stats.Constantine Khroulev1+2-0
2014-09-05 23:11Remove test/cases/haseloff/no_bed_smoother.ncConstantine Khroulev2+1-1
2014-09-05 17:53Fix ice discharge reporting. Port to 0.6.Constantine Khroulev1+7-1
2014-09-04 22:17Make Nq and Nk officially unsigned (because they are).Constantine Khroulev9+76-76
2014-09-04 21:37One more fix for #291.Constantine Khroulev2+11-9
2014-09-04 20:26Address #291.Constantine Khroulev8+104-56
2014-09-04 19:02Cleanup.Constantine Khroulev1+25-19
2014-09-04 19:00IceModelVec methods that modify the stored field call inc_state_counter().Constantine Khroulev5+48-22
2014-09-04 18:59Remove pism::Component::init().Constantine Khroulev4+6-2
2014-09-04 16:32Cleanup in PIO.ccConstantine Khroulev1+88-118
2014-09-04 16:17Add PIO::use_mapped_io().Constantine Khroulev2+79-7
2014-09-04 15:55Add "using std::vector" and "using std::string" to PIO.ccConstantine Khroulev1+133-130
2014-09-02 23:48Update the LocalInterpCtx documentation comment.Constantine Khroulev1+2-5
2014-09-02 23:45Add m_ prefix to protected data members of IceGrid.Constantine Khroulev2+7-7
2014-09-02 23:41Add the energy_advection_ice_thickness_threshold parameter.Constantine Khroulev3+5-2
2014-08-29 22:09Cleanup.Constantine Khroulev2+49-48
2014-08-29 20:52Cosmetic changes.Constantine Khroulev1+8-4
2014-08-29 18:15PETSC_NULL is deprecated. Replace with NULL.Constantine Khroulev33+74-74
2014-08-29 17:45Simplify adding more boundary modelsConstantine Khroulev5+130-232
2014-08-28 19:25Implement "-atmosphere one_station".Constantine Khroulev5+350-0
2014-08-28 18:15Use "std::vector<double> &" instead of "double *" in function arguments.Constantine Khroulev29+117-114
2014-08-28 18:12Use the "sia_" prefix in the ConstantInColumn class.Constantine Khroulev2+4-1
2014-08-27 19:07Use one-sided differences to compute w and strain heating at all margins.Constantine Khroulev2+58-22
2014-08-26 17:11Remove some uses of "const char *" and "const char []".Constantine Khroulev41+114-113
2014-08-26 00:43PIO::compute_start_and_count() supports t_count > 1.Constantine Khroulev2+11-7
2014-08-26 19:36Try not to use raw arrays allocated with "new".Constantine Khroulev15+188-181
2014-08-25 21:42Minor cleanup.Constantine Khroulev2+15-17
2014-08-12 07:22Addresses issue #289. Requires testing. No change in semantics of -topg_to_phi option. In particular, this mechanism cannot be turned on without the option (and all four parameters). So the defaults for the four parameters are never actually used.Ed Bueler3+28-23
2014-08-12 04:49Partially address issue #289. This version compiles and runs but does not actually set the config variable according to what the user option is. Thus the whole handling of option -topg_to_phi needs to move to src/base/util/pism_options.cc because that method has a non-constant copy of Config.Ed Bueler2+28-24
2014-08-09 16:47a ref got publishedEd Bueler1+5-6
2014-08-06 01:57one more useful referenceEd Bueler1+9-0
2014-08-05 19:23Update references related to recent acceptances etc.Ed Bueler1+17-12
2014-07-25 04:30Revert "Try to support PETSc 3.5.0 (in dev)."Constantine Khroulev14+21-134
2014-07-24 19:32Documenting options -sia_n and -ssa_n as in my previous commit. See #285.Ed Bueler1+63-31
2014-07-24 16:21Fix the SSA regression test I that uses Python bindings.Constantine Khroulev1+1-1
2014-07-24 06:46Addressing #285. Documentation not complete, but new options -sia_n and -ssa_n do work.Ed Bueler11+130-75
2014-07-17 15:16Fix a conflict resolution error.Constantine Khroulev1+1-6
2014-07-16 18:00Try to support PETSc 3.5.0 (in dev).Constantine Khroulev15+140-22
2014-07-12 21:42Indentation cleanup.Constantine Khroulev1+12-12
2014-07-15 02:03Change the rule for removing width-one ice tongues ("noses").Ed Bueler1+38-16
2014-07-13 08:03Minor: Fix typo in comment.Ed Bueler1+1-1
2014-07-12 20:37Changes to SSAFD convergence failure recovery code.Ed Bueler4+49-22
2014-07-11 18:45Clobber pedantic warning.Ed Bueler1+3-3
2014-07-11 01:05Fix #281. Port to 0.6.Constantine Khroulev1+13-1
2014-07-09 23:47All diagnostics pressure,tauxz,tauyz in #280 added. Note (1) they are shallow formulas at StressBalance level, and (2) there is code duplication between PSB_tauxz and PSB_tauyz.Ed Bueler3+160-5
2014-07-09 21:24This works. It is inelegant because it does the computation in PSB_pressure not in the ShallowStressBalance where the hydrostatic approximation already is assumed. My way of doing this may not generalize to tauxz,tauyz because I may not be able to get access to driving stress. See issue #280.Ed Bueler3+27-13
2014-07-09 20:03Start by implementing pressure diagnostic. Addressing issue #280.Ed Bueler4+52-4
2014-06-23 21:58It should read velsurf, not {u,v}velsurf.Andy Aschwanden1+1-1
2014-06-23 19:29in addition to wvelsurf, uvelsurf and vvelsurf should also be part of -o_size medium.Andy Aschwanden1+1-1
2014-06-20 03:34With this modification we have a precise relationship between the N(e) formula in Tulaczyk et al 2000 and the N_til(W_til) formula documented in Bueler and van Pelt (submitted soon). This changes make no meaningful difference to Greenland tau_c results, as far as I can tell.Ed Bueler2+37-26
2014-06-20 03:18Rename configuration parameter till_c_0 to till_cohesion, as clearer and avoiding naming by variable name.Ed Bueler3+8-5
2014-06-17 19:00Mark ability to Pism_TEST_USING_VALGRIND as advanced. Port to 0.6.Ed Bueler1+1-0
2014-06-17 18:59Clobber -Wshadow warning.Ed Bueler1+3-3
2014-06-17 06:47Yet another ref.Ed Bueler1+10-0
2014-06-17 06:17Another ref.Ed Bueler1+11-0
2014-06-15 01:45Added -subgl to -pik bundle. Documented it, and cleaned up marine and pik options in User Manual.Ed Bueler3+108-93
2014-06-15 00:54Change siam.bst->plain.bst to avoid koma-generated old font warnings; perhaps plain.bst has been maintained since 1988.Ed Bueler2+2-2
2014-06-14 23:34Adding -subgl makes it even better. Spectacularly steady 15km run.Ed Bueler1+2-2
2014-06-12 19:26another referenceEd Bueler1+10-0
2014-06-06 01:42Another reference.Ed Bueler1+11-0
2014-06-05 19:07Minor: Le Brocq as a last name.Ed Bueler1+1-1
2014-06-04 20:19Maintenance on references.Ed Bueler1+9-8
2014-06-03 08:04Several references relevant to green-hydro stuff.Ed Bueler1+69-11
2014-06-03 08:04Remove incorrect FIXME in script.Ed Bueler1+1-1
2014-06-01 22:40Add option -till_log_factor_transportable_water which sets parameter of same name.Ed Bueler1+2-0
2014-06-01 08:28Moved script showPvsW.py to pism/green-hydro where I am actually using it.Ed Bueler1+0-181
2014-06-01 08:09Fix bug in color scale in script.Ed Bueler1+9-8
2014-06-01 06:33Further improvement of scatter plot generator for P(W) evaluation.Ed Bueler1+59-13
2014-05-31 23:46Remove incorrect mention of Storglaciaren in intro material in manual.Ed Bueler1+1-1
2014-05-31 07:01Remove an incorrect FIXME.Ed Bueler1+0-2
2014-05-31 05:31Updating hydrology parameters to match paper and its choices. Removing overrides mechanism (as code duplication) from nbreen example.Ed Bueler4+8-60
2014-05-29 07:41Vastly more powerful scatterplot generator for testing whether points show emergent P(W). (And Python + Matplotlib rocks.)Ed Bueler1+66-5
2014-05-29 05:50Use CRITICAL_FILL_MISSING flag and fix committed by ckhroulev in 1f14abf. See 23dda01. Works for me.Ed Bueler1+2-4
2014-05-29 03:24Merge branch 'dev' into hydro-simplify-transferEd Bueler21+347-145
2014-05-28 06:54Clean up scripts.Ed Bueler2+2-2
2014-05-28 06:53Word "lost" should be "loss".Ed Bueler2+4-4
2014-05-27 19:20Being more careful about meaning of convergence of Test P.Ed Bueler2+30-12
2014-05-26 07:14tool to generate scatter plot of P versus W, to examine evidence for P(W) a la Flowers and ClarkeEd Bueler1+73-0
2014-05-26 06:19Minimal maintenance on Nbreen example.Ed Bueler2+17-1
2014-05-25 21:56Replace the bad choice for a W <--> Wtil transfer model with simpler (null) version.Ed Bueler7+101-110
2014-05-25 03:01Fix problem where read from file of diagnostic variable velbase_mag has _FillValue below valid_min.Ed Bueler1+3-1
2014-05-24 07:14Merge branch 'dev' into feed-hydrology-from-fileEd Bueler2+10-5
2014-05-24 07:05Add option -hydrology_velbase_mag_file and diagnostic hydrologyvelbase_mag.Ed Bueler4+57-5
2014-05-24 06:41Implement and document option -hydrology_bmelt_file and diagnostic hydrobmelt.Ed Bueler6+82-16
2014-05-24 00:28Correct option name to -hydrology_input_to_bed_file and document it and related options.Ed Bueler2+13-13
2014-05-23 18:42Use std::shared_ptr and std::weak_ptr to keep track of PETSc DMs.Constantine Khroulev25+251-197
2014-05-17 02:24Allow optional addition of -pik -subgl in std-greenland example.Ed Bueler1+5-0
2014-05-17 02:23Correct citation in manual regarding -subgl.Ed Bueler1+1-1
2014-05-14 19:44fill_missing_petsc.py: try to provide better error messages.Constantine Khroulev1+11-2
2014-05-14 19:13Create temporary file in the output directory.Constantine Khroulev1+9-6
2014-05-14 16:58Remove the basal_melt_rate.update_ghosts() call.Constantine Khroulev1+0-4
2014-05-14 07:20fixed two typos.Andy Aschwanden1+2-2
2014-05-13 18:56Add the PETSc-based parallel fill_missing script.Constantine Khroulev1+404-0
2014-05-09 21:12Update the float_kill command-line option.Constantine Khroulev1+2-2
2014-05-09 21:11Make sure that IceModel::combine_basal_melt_rate() gets called.Constantine Khroulev2+4-3
2014-05-09 19:01Remove Timeseries methods that we don't really need.Constantine Khroulev15+161-189
2014-05-13 07:43add a refEd Bueler1+12-0
2014-05-09 18:37Fix 2 minor errors in the source code browser.Constantine Khroulev2+3-3
2014-05-09 18:21Experimental: use force-to-thickness in grounded areas only.Constantine Khroulev2+10-1
2014-05-09 11:00Document new modifier PA_frac_P. Close #271.Julien Seguinot1+32-5
2014-05-08 23:07Make sure that ks is always valid.Constantine Khroulev9+123-109
2014-05-08 23:06Remove a redundant comment.Constantine Khroulev1+0-1
2014-05-08 23:52Fixed fill_missing.pyAndy Aschwanden1+23-12
2014-05-08 22:51allow fill_missing.py to handle files with a time dimension. Should this be considered a bug fix that needs to go into v0.61?Andy Aschwanden1+109-47
2014-05-08 19:49Minor cleanup in PA_frac_P. See #271.Constantine Khroulev1+13-14
2014-05-08 19:48Remove an unused variable.Constantine Khroulev1+0-2
2014-05-08 16:53Allow setting the output size without command-line options.Constantine Khroulev4+39-24
2014-05-08 14:26Enable scalar precipitation multiplier. See #271.Julien Seguinot5+245-0
2014-05-08 01:24Whitespace changes.Constantine Khroulev1+76-76
2014-05-08 00:57Close #99.Constantine Khroulev6+159-158
2014-05-07 22:31Add the "long_name" attribute to pism_config.Constantine Khroulev2+90-82
2014-05-07 22:30Remove the "PISM" prefix in the source code browser.Constantine Khroulev1+12-12
2014-05-07 22:05Add Egholm references.Ed Bueler1+22-0
2014-05-07 20:59Add "_flux" to land_ice_surface_specific_mass_balance.Constantine Khroulev25+28-28
2014-05-07 19:22Fix a doxygen comment.Constantine Khroulev1+2-2
2014-05-07 19:04Forward-declare "IceModelVec" in PISMVars.hh.Constantine Khroulev5+7-1
2014-05-07 19:04Cleanup.Constantine Khroulev2+25-23
2014-05-07 20:33Add _flux to standard name for climatic_mass_balance. See 995aac.Ed Bueler1+1-1
2014-05-07 19:39Fixing PISM parameter names (Paterson-Budd_*) to remove hyphen, to get CF compliance.Ed Bueler2+15-15
2014-05-07 18:01Improve code reuse in SSAFEM.Constantine Khroulev4+111-151
2014-05-07 02:15Changes in IceModelVel2T (needs code review).Julien Seguinot3+27-1
2014-05-06 21:38Fix a file name and small edits in section 1 of User Manual.Ed Bueler1+3-3
2014-05-06 21:08A better name for SSAFD CFBC notes.Constantine Khroulev2+7-7
2014-05-06 20:50Tweak SSA FEM notes.Constantine Khroulev1+9-7
2014-05-06 19:44Close #270.Constantine Khroulev2+9-13
2014-05-06 18:00Improve FEQuadrature::getWeightedJacobian().Constantine Khroulev9+51-64
2014-05-06 00:03Don't include .py files in etags.Constantine Khroulev1+1-1
2014-05-06 05:17Clean up indexing of Jacobian entries.Constantine Khroulev1+10-10
2014-05-05 22:43Avoid passing classes by value. Close #269.Constantine Khroulev156+1112-1037
2014-05-04 17:28Completion of revisions to first section of user manual, based on change in commit fc93b61ef.Ed Bueler18+16-7
2014-05-03 21:02Remove FEQuadrature::init().Constantine Khroulev6+43-39
2014-05-03 18:49Fixes to figures in User Manual resulting from change in previous commit; figures 3--9 fixed. Text updated.Ed Bueler14+29-31
2014-05-03 18:02Fix RACMO surface mass balance missing value issue. Port to 0.6.Ed Bueler2+11-4
2014-05-03 17:15SSAFEM: Move some computations out of inner loops.Constantine Khroulev1+39-38
2014-05-03 06:15Improve SSA FEM notes.Constantine Khroulev1+9-15
2014-05-03 06:12Re-write SSAFEM Jacobian and residual code.Constantine Khroulev1+53-41
2014-05-02 23:16Start switching to variables that actually mean something.Constantine Khroulev1+35-29
2014-05-02 23:11Add notes documenting SSAFEM residual and Jacobian computations.Constantine Khroulev1+405-0
2014-05-01 01:12Fix the -skip mechanism (again).Constantine Khroulev3+50-23
2014-04-30 19:45Remove the "seconds_per_year" config. parameter.Constantine Khroulev1+0-3
2014-04-30 19:01Allow using a different ftt alpha in "ice-free" areas.Constantine Khroulev5+57-35
2014-04-30 18:47Fix installation of the installation manual.Constantine Khroulev1+1-1
2014-04-30 17:56Add the m_ prefix to PSForceThickness data members.Constantine Khroulev2+62-62
2014-04-29 23:39My attempt to rename -force_to_thk. See #268.Constantine Khroulev3+3-14
2014-04-29 23:36Closes #268.Andy Aschwanden7+26-26
2014-04-29 16:11Whitespace cleanup.Constantine Khroulev134+750-760
2014-04-29 00:43Improve some doxygen comments.Constantine Khroulev2+22-10
2014-04-29 00:42Use typedef to clean up the signature of a FEQuadrature method.Constantine Khroulev2+9-5
2014-04-28 22:11Remove an outdated comment.Constantine Khroulev1+0-1
2014-04-25 23:39More cosmetic changes in SSAFEM.Constantine Khroulev5+144-108
2014-04-25 16:37Fix doxygen warnings.Constantine Khroulev1+2-2
2014-04-25 16:36Document IceFlowLaw::effective_viscosity().Constantine Khroulev1+17-3
2014-04-25 15:49More cleanup in SSAFEM.Constantine Khroulev3+52-46
2014-04-24 21:38Some more cleanup.Constantine Khroulev6+72-71
2014-04-24 19:56Minor cleanup in SSAFEM.Constantine Khroulev4+124-123
2014-04-24 19:26Add documenting comments. See #224.Constantine Khroulev1+4-4
2014-04-24 19:07Remove the "PISM" prefix from most names in PISM.Constantine Khroulev277+2845-2838
2014-04-23 19:57Read Href from an input file during bootstrapping.Constantine Khroulev1+9-5
2014-04-23 19:57Ice volume/area stdout reporting consistent with -ts_file reporting.Constantine Khroulev2+4-42
2014-04-23 13:48Address #267.Constantine Khroulev1+2-1
2014-04-23 12:29Warn for negative ice thickness in part_redist.Julien Seguinot1+10-0
2014-04-22 20:03Fix compilation errors on beauregard.Constantine Khroulev10+56-56
2014-04-18 19:26Move all public(*) symbols into the "pism" namespace. See #188.Constantine Khroulev310+3177-2051
2014-04-22 19:37Get rid of one more -Wshadow warning.Constantine Khroulev1+2-2
2014-04-22 19:33Get rid of a -Wshadow warningConstantine Khroulev5+84-85
2014-04-20 23:38I will submit this tiny thing on hydrology soon.Ed Bueler1+12-3
2014-04-18 02:55Document the -atmosphere_searise_greenland_file option.Constantine Khroulev1+1-1
2014-04-18 02:47Close #263.Constantine Khroulev3+46-13
2014-04-17 22:40Add taub_mag to output_medium and _big. See #266.Constantine Khroulev1+2-2
2014-04-17 22:02Split tables in the manual. Rename some variables. See #266.Constantine Khroulev45+197-172
2014-04-17 21:17Add "taud_mag" to the User's Manual.Constantine Khroulev1+1-0
2014-04-17 21:15Address the code part of #266.Constantine Khroulev7+209-138
2014-04-17 21:12Add "const" to some methods.Constantine Khroulev5+12-12
2014-04-17 21:44Trying yet again on the meaning associated to option -tauc_add_transportable_water.Ed Bueler2+8-6
2014-04-17 20:22Another round of edits to subglacial strength documentation in Users Manual. Port to 0.6 (without FIXME which should remain in dev).Ed Bueler1+16-24
2014-04-17 18:38Make flux_divergence appear in -o_sizee big files. See #165.Constantine Khroulev1+63-62
2014-04-17 17:42Close #210.Constantine Khroulev4+58-70
2014-04-17 17:10Dealt with some -Wshadow pedantic compiler warnings.Ed Bueler2+12-12
2014-04-17 17:04Remove unmaintained/unused PISMDistHydrologyALT model.Ed Bueler3+0-259
2014-04-17 16:58Added all state and diagnostic vars generated by PISMDistributedHydrology to output_big. "big" means BIG. Closes #264.Ed Bueler1+1-1
2014-04-17 16:06Remove tillwp, add bwat, bwatvel to output_medium. See #264.Constantine Khroulev1+2-2
2014-04-03 17:48Improve the PIO::open() API. Fix the non-issue #224.Constantine Khroulev58+249-119
2014-04-17 00:37Update README.md (in dev).Constantine Khroulev1+1-1
2014-04-15 22:37Trying again on the meaning associated to option -tauc_add_transportable_water. See 33acd210. Needs testing; perhaps this saves Greenland from catastrophic collapse.Ed Bueler1+5-1
2014-04-15 22:03Merge branch 'pdd-stdev' into devConstantine Khroulev8+203-26
2014-04-14 21:35Merge branch 'ocean' into devConstantine Khroulev19+669-609
2014-04-14 19:11Minor cleanup.Constantine Khroulev8+135-90
2014-04-14 17:14Looked at the code, as requested.Constantine Khroulev2+156-145
2014-04-14 17:01Document tauc_add_transportable_water.Constantine Khroulev1+9-1
2014-04-11 22:33Minor userman update.Constantine Khroulev1+2-2
2014-04-11 05:38white space fixEd Bueler1+1-2
2014-04-11 05:30Create new option and config flag -tauc_add_transportable_water.Ed Bueler4+55-7
2014-04-11 05:21add ability to start -hydrology distributed model if bwp is missing; uses bootstrapping_bwp_value_no_var valueEd Bueler1+20-4
2014-04-11 05:15fix unreported bug: nc file left open when PISMRoutingHydrology attempted to find bwat in input fileEd Bueler1+1-0
2014-04-10 21:52A slightly different version of the fix in ada8877. Port to 0.6.Constantine Khroulev3+7-3
2014-04-09 18:18Fiddling with the FEM code in PISM.Constantine Khroulev18+1406-1314
2014-04-10 15:25Prevent crash upon missing unit attr (port to 0.6)Julien Seguinot1+3-3
2014-04-09 05:58Whitespace changes.Constantine Khroulev3+194-199
2014-04-09 05:58Minor cleanup.Constantine Khroulev1+3-3
2014-04-09 01:03See #266. Added minimal doc of taub to manual. Port to 0.6. Note taub can be set equal to the driving stress when computed by the nonsliding SIA-only model.Ed Bueler1+3-2
2014-04-09 00:37See #266. Fix sign on tau_b. Make it bold using boldsymbol; note mathbf has no effect. Port to 0.6.Ed Bueler1+7-7
2014-04-07 22:08Replace 'mapping' with 'global_attributes' in one more place.Constantine Khroulev1+1-1
2014-04-04 21:16Fix the -..._times code.Constantine Khroulev1+3-1
2014-04-04 19:57"-surface pdd_old" should use new variable names.Constantine Khroulev2+110-88
2014-04-04 19:34Switch from gsl_sf_erfc() to the C library erfc().Constantine Khroulev1+2-3
2014-04-04 20:57fixed nc2cdo.py which is a bit trickier to handle correctly.Andy Aschwanden1+34-34
2014-04-04 20:01For correct calculations of cell areas, PISM now searches in the global attribute ‘proj4’ instead of the proj4 attribute in the mapping variable.Andy Aschwanden6+52-29
2014-04-03 17:59Fix a build issue on Mac OS X 10.9.2. Port to 0.6.Constantine Khroulev2+14-13
2014-04-02 15:30Document building docs in INSTALL.md.Constantine Khroulev1+9-0
2014-04-01 23:15a ref: Feldmann et al (2014) is publishedEd Bueler1+5-1
2014-04-01 18:56Address #251.Constantine Khroulev12+920-880
2014-04-01 18:19Build libpisminverse if Python bindings are being built too. (And not otherwise.)Constantine Khroulev1+7-2
2014-04-01 17:30Add more doxygen comments (PS_delta_T).Constantine Khroulev1+13-5
2014-04-01 17:36Name of x11 library with headers in debian, needed to get X functionality automatically when building petsc, has changed. Port to 0.6.Ed Bueler1+1-1
2014-04-01 15:33Add more doxygen comments.Constantine Khroulev5+56-3
2014-03-31 21:40Add the Pism_DOXYGEN_DOCUMENT_ALL CMake option.Constantine Khroulev2+11-2
2014-04-01 13:41Found a tiny error in random PDD init messageJulien Seguinot1+2-2
2014-03-28 21:21Convert SMB from m/s to (kg m-2)/s.Constantine Khroulev1+6-1
2014-03-28 17:25Resurrect old PDD code so that we can compare it to the new.Constantine Khroulev6+1171-0
2014-03-27 21:42Better error messages. Port to 0.6.Constantine Khroulev2+32-1
2014-03-27 20:36IceGrid::kBelowHeight() should call MPI_Abort().Constantine Khroulev1+6-8
2014-03-26 19:44Fix a typo in IceModel::step() (in dev).Constantine Khroulev1+1-1
2014-03-27 15:44Fix discharge reporting. Port to 0.6.Constantine Khroulev1+3-2
2014-03-26 19:07Update the gradient method test case.Constantine Khroulev2+3-3
2014-03-13 06:29Correct some doxygen comments.Constantine Khroulev2+8-9
2014-03-18 00:06Update the Ross run scripts.Constantine Khroulev2+4-4
2014-03-17 18:33Remove the "preliminary time-step duration" parameter.Constantine Khroulev1+0-3
2014-03-19 01:16IceModel::skip_counter() *returns* result.Constantine Khroulev2+30-39
2014-03-17 18:09Add -timestep_hit_multiples. Port to 0.6.Constantine Khroulev7+275-193
2014-03-14 17:08-surface_given_period X should take an integer X. Port to 0.6.Constantine Khroulev1+8-8
2014-03-13 07:11Initialize variables in PDDMassBalance::step(). Port to 0.6.Constantine Khroulev1+2-2
2014-03-13 06:30PDD sub-intervals consistent with interpolation scheme. Port to 0.6.Constantine Khroulev2+9-8
2014-03-13 06:29Fix snow depth reset code. Port to 0.6.Constantine Khroulev1+1-1
2014-03-12 18:55Very minor cleanup. Port to 0.6.Constantine Khroulev2+2-1
2014-03-12 18:53The "init. step" of couplers is a method now. Port to 0.6.Constantine Khroulev3+58-47
2014-03-24 22:57Remove undocumented storglaciaren example from users manual. Port to 0.6.Ed Bueler2+4-4
2014-03-24 17:11Save CTS when -o_size big is set. See #262. Port to 0.6.Constantine Khroulev2+2-2
2014-03-21 23:26yet more ref fiddles (YMRF)Ed Bueler1+15-3
2014-03-21 23:13more ref workEd Bueler1+12-2
2014-03-19 22:31a ref.Ed Bueler1+13-2
2014-03-19 21:14Allow setting units of the z dimension in IceModelVec3Custom.Constantine Khroulev1+7-1
2014-03-19 18:52Apply the patch provided by Leo van Kampenhout. (Port to 0.6.)Constantine Khroulev4+28-1
2014-03-14 19:21Added commentBob Fischer1+36-0
2014-03-14 19:17Merge branch 'dev' of github.com:pism/pism into devBob Fischer11+156-24
2014-03-05 14:08Merge branch 'dev' of github.com:pism/pism into devBob Fischer163+8059-6054
2014-02-07 04:15Merge branch 'dev' of github.com:pism/pism into devBob Fischer309+5208-6722
2014-02-07 04:14Working on moving Glint2 code into PISMBob Fischer5+451-0
2014-01-29 17:29Added comments. Changed init_run() to be not protected.Bob Fischer2+13-4
2014-01-29 02:35section 12.1 is done: PISM on a 2.5 mm grid and a nice match to observationsEd Bueler2+21-15
2014-01-29 00:02new section 12.1 on laboratory experiment validation mostly doneEd Bueler2+44-1
2014-01-28 21:32eliminate unmaintained EISMINT-Greenland directory; it is still available in stable0.4Ed Bueler8+0-493
2014-01-28 21:30Correct the README to reflect current.Ed Bueler1+3-2
2014-01-28 21:18Revised section headings for validation chapter, with spot for labgum material.Ed Bueler1+10-3
2014-01-28 21:12remove flux figures from User Manual description of MISMIP; not clear at this point that they mean anything other than that cflx is not the conserved fluxEd Bueler4+12-25
2014-01-28 20:18rename in prep for adding labgum to validation chapterEd Bueler3+3-3
2014-01-28 20:07A bit more clean up including consolidation into one preprocess script and one run script.Ed Bueler4+142-155
2014-01-28 19:07Clean up lab gum example in preparation for addition to Users Manual.Ed Bueler7+459-79
2014-01-28 02:06Add a paragraph about -subgl to the manual.Constantine Khroulev1+6-1
2014-01-28 01:13Remove 2 FIXMEs from the manual.Constantine Khroulev1+0-4
2014-01-27 23:33Clean up and document the new -subgl code.Constantine Khroulev4+232-165
2014-01-27 19:31Minor changes to the code picking the right (smallest) max_dt.Constantine Khroulev2+43-31
2014-01-27 18:17Minor cleanup (mostly whitespace).Constantine Khroulev3+128-142
2014-01-27 18:00Rename some IceModel data members.Constantine Khroulev25+523-524
2014-01-26 03:09More maintenance on Storglaciaren example: Move parameter choices to command-line options and revert to more typical sliding parameters; ran but generated crazy velocities under old values.Ed Bueler3+27-28
2014-01-26 02:25removing 3d example ... which I will not myself maintain ...Ed Bueler3+2-381
2014-01-25 20:30Attempt maintenance on storglacieren example: Shallow edits to manual. Changes to runscript; generate error about file missing tillphi. (Happens when switch from SIA-only run to SIA+SSA run; is this generic?)Ed Bueler2+14-16
2014-01-25 20:07Change pre-computed result used in examples/jako/ to g5km_gridseq.nc. Remove unmaintained Petermann stub script.Ed Bueler4+19-28
2014-01-25 08:26Updates to the subgrid grounding line interpolationJohannes Feldmann5+135-69
2014-01-25 04:09Remove redundant code.Constantine Khroulev1+2-13
2014-01-25 04:01Address #233.Constantine Khroulev1+3-3
2014-01-25 02:14A round of shallow edits to the Jako. regional model section of the Users Manual. In good shape but note #233.Ed Bueler1+15-20
2014-01-25 01:14Shallow edits to Verification and Simplified Experiments sections of manual; in good shape.Ed Bueler2+31-39
2014-01-24 21:37Fix minor manual build error caused by imperfect rebase.Ed Bueler1+1-2
2014-01-24 21:24Edits to Practical Usage section of User Manual, including move of flow-line example script into util/Ed Bueler3+123-155
2014-01-24 19:55Remove option -pismversion. As far as I can tell, no script or documented run even uses it. Option -version remains.Ed Bueler1+3-4
2014-01-24 19:09Change all executables to use default output name 'unnamed.nc'.Ed Bueler3+11-9
2014-01-24 19:53Zero-pad years in stdout and -extra_split file names.Constantine Khroulev2+3-3
2014-01-24 19:26Fix a typo in a doxygen comment.Constantine Khroulev1+1-1
2014-01-24 19:14Take care of FIXMEs in the init sequence graph.Constantine Khroulev1+17-25
2014-01-24 18:45Remove last mentions of -no_sia and -ssa_sliding.Constantine Khroulev2+2-2
2014-01-24 18:44Fix a typo in the manual.Constantine Khroulev1+1-1
2014-01-24 18:13Update stress balance options in the manual.Constantine Khroulev12+71-52
2014-01-24 03:52Update more scripts (-ssa_sliding, -no_sia).Constantine Khroulev7+9-15
2014-01-24 07:44Documentation of marine ice sheets and bed deformation in User Manual in pretty good shape.Ed Bueler3+56-59
2014-01-24 01:05I think this is satisfactory documentation of the two documented hydrology choices.Ed Bueler2+66-31
2014-01-24 00:50Update -ssa_sliding option in examples/*.Constantine Khroulev23+29-30
2014-01-24 00:41Update stress balance command-line options. See #144.Constantine Khroulev13+114-123
2014-01-23 20:35Move the basal sliding law into ShallowStressBalance.Constantine Khroulev45+328-334
2014-01-23 20:07Remove unused code.Constantine Khroulev2+0-8
2014-01-23 21:39Edits through end of section 7.1 on modeling basal strength. Much better.Ed Bueler1+44-54
2014-01-23 19:17More rational construction of basal strength documentation, but it is still not done.Ed Bueler2+91-55
2014-01-23 19:48Add "climatic_mass_balance_original". Closes #232.Constantine Khroulev2+26-1
2014-01-23 19:30Fix a bug in "Timeseries".Constantine Khroulev1+5-3
2014-01-22 20:11A bit of cleanup.Constantine Khroulev6+35-44
2014-01-22 18:39Remove Blatter stress balance code from dev.Constantine Khroulev12+6-2509
2014-01-22 18:16Remove the do_blatter flag (it is not used anyway).Constantine Khroulev5+29-39
2014-01-22 01:02Clean up energy command-line options. See #144.Constantine Khroulev60+204-1025
2014-01-21 22:28Added configuration flag summary_time_use_calendar. Defaults to "yes".Ed Bueler3+18-4
2014-01-21 20:26clobber pedantic warning about ierr set but not usedEd Bueler1+3-5
2014-01-21 18:16Require FFTW-3 when building PISM. See #230.Constantine Khroulev7+16-45
2014-01-21 06:13Edits on ice dynamics part of manual, including a brief new section on conservation of energy.Ed Bueler2+42-30
2014-01-21 04:11Significant reorganization: split out subglacial stuff from dynamics, to make new section of User Manual.Ed Bueler4+235-214
2014-01-21 04:11Shallow edits to section 5. In good shape.Ed Bueler1+46-68
2014-01-21 01:44More shallow edits through end of section 4 (page 36).Ed Bueler2+30-39
2014-01-21 01:05More shallow edits through end of section 3.Ed Bueler4+23-22
2014-01-20 21:44Many shallow edits through page 31 of User Manual.Ed Bueler4+71-62
2014-01-20 10:55deleted the plot script by accidentTorsten Albrecht3+103-9
2014-01-20 10:47simplified ross example structureTorsten Albrecht12+141-667
2014-01-19 20:26minor: fix bad ref in User ManualEd Bueler1+1-1
2014-01-19 06:28Update run script. See #175.Ed Bueler1+1-1
2014-01-19 06:17I am totally delighted with MISMIP3d. It works great with the dev branch using 2 km resolution, the subgrid method, and robust ksp options. (I did not do a thorough test of options.) This commit is a lot of shallow editing. Main question: is -gradient eta a key for performance? (Any other key options?)Ed Bueler9+232-187
2014-01-18 23:52editting of MISMIP3d scripts ... needs more experimentation now that I see all the options fairly clearlyEd Bueler3+68-49
2014-01-17 21:13Remove the -hold_tauc option. See #144.Constantine Khroulev10+48-38
2014-01-17 20:33Remove IceModelVec copy constructor and assignment operator.Constantine Khroulev21+73-163
2014-01-17 19:28Remove sounding viewers. Closes #223.Constantine Khroulev13+35-211
2014-01-17 08:32relatively small fixes in the simplified geometry part of the Users ManualEd Bueler1+5-5
2014-01-17 00:16Update calving command-line options. See #144.Constantine Khroulev39+153-122
2014-01-16 19:54Stop trying to "import netCDF3".Constantine Khroulev44+108-79
2014-01-16 19:40Remove pure-HDF5 I/O code.Constantine Khroulev9+8-1324
2014-01-16 19:33Improve the main page of the browser. Done for now.Constantine Khroulev2+44-163
2014-01-17 00:03working on script for MISMIP3d: clean up and track down why -subgl *does* workEd Bueler1+70-96
2014-01-16 20:21minor reorganization of User Manual source files: there is now a one-to-one correspondence between .tex files in doc/userman/ and the sections of the User ManualEd Bueler5+121-120
2014-01-16 04:37Address some easily noticeable flaws of the doxygen browser.Constantine Khroulev19+901-820
2014-01-15 02:59A minor fix.Constantine Khroulev1+2-2
2014-01-15 00:45Source code browser compiles again.Constantine Khroulev9+87-97
2014-01-14 23:39small edits of MISMIP3d documentation ... under constructionEd Bueler2+14-11
2014-01-14 23:38cut up MISMIP3d scripts into more easily chewed pieces; see new "usage" section of README.md; under construction and testingEd Bueler3+115-91
2014-01-14 20:39Minor cleanup.Constantine Khroulev7+43-33
2014-01-14 20:38Rename IceModelVec::stencil_width() back to get_stencil_width()Constantine Khroulev7+18-18
2014-01-14 20:36SIA mass conservation test prints the "pismv" command it runs.Constantine Khroulev1+5-0
2014-01-14 20:35Fix a build system heuristic.Constantine Khroulev1+0-1
2014-01-14 17:36Fix build issues on beauregard.Constantine Khroulev5+41-36
2014-01-14 18:50happy with MISMIP (2d) results for inclusion in stable0.6, though -subgl did not work for meEd Bueler3+16-9
2014-01-14 18:11fix minor build error (and allows me to play); see #228Ed Bueler2+2-2
2014-01-14 06:56Trying to update the doxygen browser.Constantine Khroulev18+2380-3964
2013-12-22 11:54Improving the IceModelVec, NCVariable, PIO API.Constantine Khroulev262+4804-5228
2013-12-22 09:01Initialize variables in PISMDistributedHydrology::update()Constantine Khroulev1+9-9
2013-12-20 23:21Minor manual fixes.Constantine Khroulev2+3-3
2014-01-14 06:18figures cleaned up for MISMIP 2dEd Bueler6+17-23
2014-01-14 05:39MISMIP 2D works fine in grid modes 3 and 2 *as long as you do not skip*, so it is turned offEd Bueler2+1-4
2014-01-14 05:08minor: clarifying section names in manualEd Bueler2+2-2
2014-01-14 02:05struggling a bit to understand where the MISMIP (old, not 3d) example standsEd Bueler2+33-26
2014-01-13 23:48fix doxygen comment for putTempAtDepth()Ed Bueler1+10-9
2014-01-13 23:28putTempAtDepth() did indeed have a bug in the default case where the SMB is used: if SMB<0 then ice was treated as advecting basal warmth upward from base, generating unreasonably warm ice. Fixed by treading SMB<0 columns as having vertical velocity (relative to the base) of zero.Ed Bueler1+22-26
2014-01-13 19:51Cleaning installation manual: add python-pyproj to recommended debian packages, group and alphabetize packages, and just go ahead and say FFTW is required.Ed Bueler1+29-31
2014-01-10 07:49Cleaned up and documented the prognostic and fracture_density extensions of the Ross ice shelf. They are ready to go in stable0.6 in my opinion.Ed Bueler7+34-119
2014-01-10 00:44Superficial changes for usability/clarity. (I did not even realize that these nice examples were here.)Ed Bueler5+24-12
2014-01-08 23:44more minor cleanup in antarctica exampleEd Bueler3+19-20
2013-12-17 20:35Address #74 (for now).Constantine Khroulev4+49-11
2013-12-16 23:13Add simple "narrow ice tongue removal" code.Constantine Khroulev3+98-5
2013-12-16 21:55A bit of cleanup.Constantine Khroulev2+16-22
2013-12-17 01:07adding comments to nice flood example ... @ck: please check that these are still truthsEd Bueler1+24-6
2013-12-16 21:31Try to avoid overflow in IceGrid::check_parameters().Constantine Khroulev1+8-2
2013-12-16 21:28Add the marine/tongues "example".Constantine Khroulev3+165-0
2013-12-15 03:24Changes to Antarctic run which update to new till strangth scheme (esp. -tauc_slippery_grounding_lines) and use better parameters (-sia_e 3, -meltfactor_pik 5e-3, -calving_at_thickness 200.0). Separate ability to use smoothed topg; does not make that much difference.Ed Bueler2+67-36
2013-12-14 20:23This should take care of #167.Constantine Khroulev2+17-6
2013-12-14 02:40Adds the "flood" example.Constantine Khroulev3+110-0
2013-12-13 21:40Code cleanup.Constantine Khroulev5+53-55
2013-12-14 05:37Clean up and simplify SeaRISE-Antarctica example.Ed Bueler6+6-520
2013-12-11 20:18Fix a typo in src/base/stressbalance/ssa/doc/MakefileConstantine Khroulev1+1-1
2013-12-12 05:55add the ability to not generate diagnotics (i.e. -ts_file ... -extra_file ...)Ed Bueler1+11-6
2013-12-11 20:09Fix a typo in examples/marine/circular/README.mdConstantine Khroulev1+1-1
2013-12-11 20:05Cleanup in examples/ and tests/.Constantine Khroulev16+148-99
2013-12-11 17:27Add a forgotten forcing file (delta_MBP.nc).Constantine Khroulev1+0-0
2013-12-11 02:40Add examples/marine/melange.Constantine Khroulev3+41-0
2013-12-11 02:40Rename examples/piktests -> examples/marine.Constantine Khroulev14+71-72
2013-12-10 21:02Delete all debian [meta]package stuff. Remove manpages.Constantine Khroulev18+1-333
2013-12-10 21:02Clobber a compiler warning.Constantine Khroulev1+0-1
2013-11-12 18:37Remove all uses of get_array() outside of FEM code.Constantine Khroulev15+459-424
2013-12-04 23:42Fix a typo in melange back pressure code.Constantine Khroulev1+1-1
2013-12-09 02:52I am going to call it good: a new Getting Started section for pism0.6Ed Bueler6+27-34
2013-12-09 02:16add csurf figures from parameter studyEd Bueler9+0-0
2013-12-09 01:12update script for generating time series, and the ivol .png files which go in User's ManualEd Bueler5+8-3
2013-12-08 22:01more text on parameter studyEd Bueler1+20-6
2013-12-08 21:41update and clean up advanced param study scripts ... only play small role in Getting Started sectionEd Bueler5+143-144
2013-12-08 05:58about half of section 2.9 showing a parameter study example done; need one more big figure and some text on what it all means ...Ed Bueler9+97-20
2013-12-08 01:43done with section 2.8 on grid refinementEd Bueler2+26-14
2013-12-08 01:15revise/add some scripts for sections 2.8 and 2.9Ed Bueler4+71-71
2013-12-07 04:37editing section 2.8Ed Bueler1+18-14
2013-12-05 01:37minor: fix usurf figs so they have a common scaleEd Bueler1+10-2
2013-12-04 18:59added escape character so that log file gets written on pacmanAndy Aschwanden1+2-2
2013-12-03 23:59most of text on grid sequencing doneEd Bueler1+18-8
2013-12-03 23:59scripts for parameter study running now; suitable for small machine (and supercomputer will be an exercise)Ed Bueler2+14-4
2013-12-03 20:08Document melange back pressure fraction forcing.Constantine Khroulev3+66-10
2013-12-03 18:54Implement -ocean ...,delta_MBPConstantine Khroulev7+212-1
2013-12-02 19:09Implement the melange back pressure parameterization.Constantine Khroulev22+173-25
2013-12-02 19:07Very minor code cleanup.Constantine Khroulev4+7-16
2013-12-02 00:59Working on new subsection 2.8 on grid sequencing. Clean up of other material in sections 2 and 3. Increase usability and documentation of scripts in examples/std-greenland/.Ed Bueler12+75-52
2013-12-01 18:49done with a sweep cleaning up section 3 ... I hope I have helped ...Ed Bueler1+70-95
2013-11-30 06:30key fix to stop jumps at regriddingEd Bueler1+1-0
2013-11-30 06:30more cleaning and editing in 'the PISM view' textEd Bueler3+23-25
2013-11-30 04:50remove preachy partEd Bueler2+2-15
2013-11-30 03:16continuing refinement of the gridseq.sh sample scriptEd Bueler2+5-5
2013-11-30 03:15edits to Constantine's new notes: minorEd Bueler2+49-49
2013-11-30 01:45fix gridseq.sh script and edit text on NetCDF toolsEd Bueler2+26-19
2013-11-30 01:18paleo subsection of Getting Started done; ivolboth.py --> tsboth.py is made more flexibleEd Bueler8+106-86
2013-11-29 22:10Continue updates to Getting Started section: section 2.8 on paleo spinupEd Bueler3+75-38
2013-11-29 20:37Add new SSA CFBC notes, including ideas re. melange back-pressure.Constantine Khroulev1+243-0
2013-11-26 20:21Clean up and minor optimizations.Constantine Khroulev16+91-81
2013-11-26 04:54Re-wrote IceModelVec3::getValColumnQUAD().Constantine Khroulev1+59-31
2013-11-26 04:52Fix a bug in IceModelVec::setValColumnPL().Constantine Khroulev1+3-1
2013-11-19 20:52New way of computing nuH.Constantine Khroulev5+220-24
2013-11-21 04:29Add the taub (taub_x, taub_y) diagnostic.Constantine Khroulev3+73-1
2013-11-19 21:19Inline IceFlowLaw::effective_viscosity().Constantine Khroulev2+23-25
2013-11-19 18:36ssa_test_cfbc improvementsConstantine Khroulev2+25-0
2013-11-19 18:37Bigger formulas in the source code browser.Constantine Khroulev1+2-0
2013-11-20 21:00Improve test and example scripts.Constantine Khroulev2+17-14
2013-11-29 08:15edits through end of section 2.6Ed Bueler1+31-26
2013-11-29 05:41edits through end of subsection 2.5, and add Dickens & Morey to refsEd Bueler2+37-30
2013-11-28 00:20apply slippery grounding line only at *marine* grounding lineEd Bueler1+6-2
2013-11-27 05:33undo change made in c6e0c522Ed Bueler1+1-6
2013-11-26 23:40minimal reorganization: put some advanced scripts, which do not get covered in Getting Started section, in subdirectoryEd Bueler5+11-3
2013-11-26 18:46method for helping with marginal stiple in basal fieldsEd Bueler3+18-6
2013-11-26 15:52Replace "full Stokes" -> "Stokes" throughout.Constantine Khroulev2+3-3
2013-11-26 08:13update text on spinup evaluation, and fix typo in scriptEd Bueler2+4-4
2013-11-25 03:35time series comparison between 20km and 10km, and other modifications to Getting Started sectionEd Bueler5+93-109
2013-11-24 22:59add Habermann et al (2013) and van Pelt et al (2013) to biblioEd Bueler1+22-6
2013-11-24 22:58showing runs with hybrid and const (climate) in getting started section, including comparison to observed and 10km runEd Bueler10+217-23
2013-11-24 08:24update comments on Mohr-CoulombEd Bueler1+6-8
2013-11-24 04:03figures from end of first runEd Bueler4+27-5
2013-11-24 02:14remove old SeaRISE-ish figures from Getting Started section ... will replace with quicker-to generate resultsEd Bueler7+7-7
2013-11-24 02:09ongoing revisions to Getting Started section: remove most vestiges of SeaRISE, plan four runs (const sia, const ssa, paleo, grid seq), add figures showing what ncview looks likeEd Bueler4+45-40
2013-11-22 02:27edit getting-started section ... needs more workEd Bueler2+26-13
2013-11-21 19:56maintenance/update lists of contributors and users in User ManualEd Bueler1+25-24
2013-11-15 20:21Revert "Added new interface to attach_atmosphere()..."Constantine Khroulev1+0-9
2013-11-14 20:20Un-comment method arguments in PISMComponent.hhConstantine Khroulev1+17-8
2013-11-15 16:35finite() is not part of ANSI C, and did not compile with G++ 4.7 -std=c++11. isfinite() is the standard replacement for it. See: http://bytes.com/topic/c/answers/217761-finite-ansi-cBob Fischer1+1-1
2013-11-15 16:02Added new interface to attach_atmosphere() that uses memory-safe C++11 pointers (std::unique_ptr<>). In general, the g++ flag --std=c++11 should be used with C++11 code; however, no changes needed to be made to the CMake stuff to get this to work on my setup. I was not able to verify the command lines make uses to compile. At some point, this change should be propagated elsewehre in the code, or ripped out based on a decision to NOT use this kind of C++ idiom.Bob Fischer1+9-0
2013-11-15 12:44fixed units for flux_divergence in extra variablesAndy Aschwanden1+1-1
2013-11-13 07:49That's about all I can say about update(). See #205.Constantine Khroulev1+49-7
2013-11-13 06:04Merge branch 'dev' of github.com:pism/pism into devBob Fischer88+1332-781
2013-11-13 06:03Added instructions on building Doxygen docs into places people might search for them.Bob Fischer2+16-0
2013-11-05 03:51Merge branch 'dev' of github.com:pism/pism into dev git told me i had to.Bob Fischer2+38-34
2013-11-05 03:51Added run_to() functionality. See Git Issue #196 https://github.com/pism/pism/issues/196Bob Fischer2+57-6
2013-11-05 01:12Fixes #199.Constantine Khroulev1+7-8
2013-11-05 00:26A little bit of cleanup.Constantine Khroulev3+7-9
2013-11-04 08:32reworking user manual example; new approach will do nicely ... will show: sia versus ssa performance (both computational and model) in 20km constant-climate context; difference between constant-climate and paleo-climate when compared to surfvelmag in data set ... i.e. the first section of the user's manual will be a 'toy' version of the Hindcasting paperEd Bueler3+40-33
2013-11-04 08:29poor nick ... I seem to lose the 'd' in his last name too easilyEd Bueler1+14-2
2013-11-04 08:28further improvments to spinup ... need revised approach to grid sequencing, which should probably use env vars REGRIDFILE and REGRIDVARSEd Bueler2+45-92
2013-11-03 22:03more improvement to spinup.shEd Bueler1+14-15
2013-11-03 21:48significant progress on highly-flexible spinup script ... but not done yetEd Bueler4+169-252
2013-11-03 18:56I think this is the preprocessing I wantEd Bueler2+95-65
2013-11-03 05:22Used search for 'unpublished' in ice_bib.bib, and then google scholar, to harvest updated bibtex entries for several papers.Ed Bueler1+33-29
2013-11-03 04:29clear out most searise-greenland stuff ... of course it is still in the stable0.5 branch, and can be brought back and updated by anyone who wants thatEd Bueler9+0-870
2013-11-02 19:27rename greenland example to remove association to seariseEd Bueler13+0-0
2013-11-01 22:17"Regrid" enthalpy using temp and liqfrac.Constantine Khroulev4+121-45
2013-11-01 16:52Trivial cleanup.Constantine Khroulev18+49-51
2013-11-01 16:51Get flow law names from IceFlowLaw.Constantine Khroulev6+6-10
2013-11-01 03:42Take the calendar from -time_file (#157).Constantine Khroulev8+118-41
2013-11-01 01:22Implement a "constant velocity" stress balance (see #171).Constantine Khroulev5+67-4
2013-10-31 19:21No, I don't want .tex files to be tagged after all.Constantine Khroulev1+1-1
2013-10-31 18:30Fix a documenting comment.Constantine Khroulev1+2-2
2013-11-01 08:28- fixed units (m year-1) of flux divergenceAndy Aschwanden2+3-3
2013-10-31 17:51Clean up standard out reporting.Constantine Khroulev5+42-13
2013-10-31 16:42Closes #155 (implement PISMComponent::regrid())Constantine Khroulev12+65-118
2013-10-31 01:23Remove unused IceModel members.Constantine Khroulev3+3-11
2013-10-31 00:12Re-implements "-float_kill". Needs testing.Constantine Khroulev8+181-15
2013-10-30 23:29Fix #115. Needs testing.Constantine Khroulev3+60-24
2013-10-30 23:14Update examples/piktests/flowline/.Constantine Khroulev2+4-8
2013-10-30 23:11PISMIcebergRemover should not remove SSA B.C. cells.Constantine Khroulev3+23-2
2013-10-30 08:10updates the url of surface velocity data downloadTorsten Albrecht2+4-4
2013-10-16 18:50Address #180.Constantine Khroulev6+146-132
2013-10-30 03:03Remove PismIntMask.Constantine Khroulev5+9-18
2013-10-29 23:57Remove unused code.Constantine Khroulev1+1-11
2013-10-29 22:03Fix a typo.Constantine Khroulev1+1-1
2013-10-29 21:43Fix a bug.Constantine Khroulev2+11-7
2013-10-29 19:59Remove "std::" that crept into a .tex file.Constantine Khroulev1+3-3
2013-10-29 18:08Remove unused variables in POGivenTH.Constantine Khroulev1+4-17
2013-10-29 17:31Fixes some scripts in examples/ross.Constantine Khroulev2+2-2
2013-10-29 17:31Clean up doxygen warnings.Constantine Khroulev8+16-16
2013-10-29 01:07Untangle the fracture-induced softening code.Constantine Khroulev8+123-66
2013-10-28 19:43Rename bwat -> tillwat in a regression test.Constantine Khroulev1+2-2
2013-10-28 19:32Some cleanup (mostly in PISMBedSmoother).Constantine Khroulev7+84-108
2013-10-27 03:33spinup not so belabored at endEd Bueler1+2-0
2013-10-27 03:30minor doc cleanupEd Bueler1+1-12
2013-10-27 03:23[I am forcing this commit message to appear in pism.pushes by a white-space commit. Sorry that it ain't elegant.]Ed Bueler1+1-0
2013-10-27 03:19Merge commit '2a4646d' into devEd Bueler0+0-0
2013-10-27 02:37Merge till-hydro-dynamics. The dev branch now has the new hydrology options -hydrology null [default] -hydrology routing -hydrology distributed The first two of these are documented in section 6.6 of the User's Manual. The third is unpublished and undocumented.Ed Bueler198+1875-1940
2013-10-27 02:17pre-emptively adding "std::" so that git gets less confused on merge of 4f7ddadb ... I hopeEd Bueler3+30-30
2013-10-27 01:35merge 878d17ee2ad1e772; fix trivial conflict in doc/ice_bib.bibEd Bueler7+490-10
2013-10-27 01:29merge c42b01a341e into till-hydro-dynamics; fix error in till-hydro-dynamics so that now it is till_water_thickness() and not basal_water_thickness() that is read in the enthalpy code for the water at the baseEd Bueler27+573-742
2013-10-23 00:14Add a forgotten if(...).Constantine Khroulev1+3-1
2013-10-27 00:48Merge commit 'fdfe1d4' into till-hydro-dynamicsEd Bueler20+970-41
2013-10-27 00:47Merge commit 'a2859a40' into till-hydro-dynamicsEd Bueler1+1-1
2013-10-27 00:45Merge commit 'dd95b6a14' into till-hydro-dynamicsEd Bueler40+1682-1207
2013-10-27 00:35Merge commit 'fe5a70d' into till-hydro-dynamicsEd Bueler6+51-5
2013-10-27 00:17merge bad7226 with conflict-resolve in doc/ice_bib.bibEd Bueler1+0-8
2013-10-03 21:47update routing test script to include thermo diagnosticsEd Bueler1+2-2
2013-10-03 18:38Fix basal B.C. for enthalpy (port to 0.5).Constantine Khroulev1+1-7
2013-10-03 18:32Tweaking PISMNC.Constantine Khroulev1+12-12
2013-10-03 18:32White-space changes.Constantine Khroulev4+91-89
2013-09-26 20:01Well-known reference for PETSc, out of date.Ed Bueler1+8-0
2013-09-25 01:38- added scalar time series 'timestamp' to merged pism filesAndy Aschwanden1+6-0
2013-09-28 19:32Add a run with till.Ed Bueler1+6-0
2013-09-28 19:31Edit metadata and stdout only.Ed Bueler2+6-6
2013-09-26 23:17This change is justified by thinking that water-free grounded ice locations should have a pressure (for the absent water ...) determined by the steady state value of the pressure. Thus there is a difference between having any sliding and not having any sliding. This change makes a huge difference in runs initialized with W=0 at the start. Experimentation is needed to determine if it affects W>0 steady states much at the boundary where W->0, in particular.Ed Bueler1+10-3
2013-09-26 20:01Well-known reference for PETSc, out of date.Ed Bueler1+8-0
2013-09-24 05:13Merge branch 'dev' into till-hydro-dynamicsEd Bueler9+83-43
2013-09-23 07:13Testing routing is interesting.Ed Bueler1+6-5
2013-09-23 02:11Add script to test -hydrology routing on searise-greenland.Ed Bueler2+51-1
2013-09-23 01:03Actually documented -hydrology null and -hydrology routing models. Note -hydrology distributed is not documented.Ed Bueler1+37-17
2013-09-23 00:09Simplifying unruly name for beta.Ed Bueler6+12-12
2013-09-22 19:37Minimally documenting the revised basal strength model, i.e. the compressible saturated till from Tulaczyk (2000b).Ed Bueler1+22-9
2013-09-22 18:52Removing this script after runnign it and looking at results. I believe it is no longer useful, if it ever was. To test -hydrology null, see runs of searise-greenland, searise-antarctica, etc. test/regression/test_29.py and test/test_hydrology/verifTestP.sh are serious verification of -hydrology distributed. Models -hydrology routing and -hydrology distributed can be compared in examples/nbreen.Ed Bueler1+0-43
2013-09-22 18:50Final state of this script.Ed Bueler1+3-3
2013-09-22 18:10Housekeeping: remove access of unused fields in the two versions of update().Ed Bueler2+3-11
2013-09-22 18:00Switching -hydrology distributed over to the superficially-hyperbolic way of updating pressure. This makes Test P (test_29.py etc) errors slightly smaller except for bwp_avg which is slightly larger. That is also the pattern with results from verifTestP.sh. Because I am not certain I want to toss it out, I've hidden old code in new derived class.Ed Bueler5+255-47
2013-09-19 22:55Set up for testing Ward's proposal to use same divergence of flux when updating pressure, as when updating W.Ed Bueler1+19-2
2013-09-19 21:36Move Wtil update to first action in time-stepping loop. Add dWtil/dt term to pressure equation (so as to match notes which already have it).Ed Bueler2+27-24
2013-09-19 20:30Use mask values to avoid averaging water thickness values from ice-free land or ocean (i.e. floating or ice-free ocean) cells, when computing the (staggered-grid) water thickness which enters into the Darcy flux calculation for subglacial hydrology. Seems to have very small affect on nbreen. Has small, and in every case error-reducing, affect on test P of -hydrology distributed.Ed Bueler2+31-6
2013-09-19 19:46Cuffey & Paterson does have good stuff on till so I should not be lazy and I should start using the latest edition ...Ed Bueler1+9-0
2013-09-17 22:30Merge dev. Fix trivial conflict in PISMMohrCoulombYieldStress.cc relating to assert.Ed Bueler27+426-175
2013-09-17 22:24In -hydrology null the floating or ice-free ocean areas have no water in till.Ed Bueler1+16-6
2013-09-17 15:48Adjustments based on running the searise-antarctica model with -hydrology null, the default. Note need for mask-awareness in updating tillwat in PISMNullTransportHydrology.Ed Bueler3+26-21
2013-09-16 19:28Tell the SSAFD KSP solver to use non-zero initial guesses. (Port to 0.5.)Constantine Khroulev2+12-3
2013-09-14 18:27Removing FIXMEs by formal changes only.Ed Bueler3+50-32
2013-09-14 17:59Removal of Y0 from PISMDistributedHydrology.cc so it runs again. Additional changes so that test_29.py passes again.Ed Bueler3+35-30
2013-09-12 23:02Added option -tauc_slippery_grounding_lines and config flag tauc_slippery_grounding_lines. By default it is 'no' PISM-wide but 'yes' in searise-greenland. Seems to have very desirable effect on the flow near grounding lines because the at-grounding-line thermodynamics that causes tillwat=0 and high tauc is bypassed. Reduces peak diffusivity if that occurs at these grounding lines, which seems common.Ed Bueler4+17-4
2013-09-12 19:07De-optimizing computation of tauc in preparation for implementing non-pointwise model: replace redundant computation on ghosted points with communication of ghosts.Ed Bueler1+4-3
2013-09-09 22:43Important point: Ntil is bounded above by overburden pressure.Ed Bueler1+1-0
2013-09-09 19:13Cleaning tauc computation without changing meaning. Fixing doxygen/LaTeX stuff. [I am seeing that grounded ice just at the grounding line of outlet glaciers currently is getting (e.g. in searise-greenland) very high tauc from MohrCoulomb because NullTransport is delivering tillwat=0 for thermodynamic reasons. Fixing the thermodynamics could be the issue but it is also possible that a special rule for such locations should apply back in MohrCoulomb (some kind of -slippery-grounding-lines rule).]Ed Bueler2+27-19
2013-09-09 07:07Cleaning up constants in examples/nbreen/ to agree with defaults.Ed Bueler1+1-6
2013-09-09 06:51Setting hydrology-related constants to match defaults listed in Table 1 of paper with Ward.Ed Bueler1+6-6
2013-09-09 00:02Minor convenience in script.Ed Bueler1+2-2
2013-09-08 06:05A bit more convenience on running shortssa.Ed Bueler1+3-1
2013-09-08 04:35For now I am guessing that till_effective_fraction_overburden should be 0.01 as default, but need to test to see.Ed Bueler2+8-6
2013-09-08 04:33Better README for future use. Indeed it should not be serious documentation but it should look o.k. at the github site, and be minimally helpful.Ed Bueler2+46-4
2013-09-08 01:47Allows control of till_effective_fraction_overburden (=delta in notes) for quick parameter study.Ed Bueler1+4-1
2013-09-08 01:47In the Mohr-Coulomb model, i.e. the Compressible-Coulomb-Plastic model of Tulaczyk 2000a, made the lower bound on effective pressure N_0 scale with overburden pressure through new configuration constant till_effective_fraction_overburden (=delta in notes). Removed previous configuration constant till_reference_effective_pressure (=N_0 itself in notes). New behavior for Greenland rather like Bueler & Brown 2009 result, i.e. in the NullTransport case. (Which is basically the desired behavior if we are working on better hydrology.)Ed Bueler4+41-21
2013-09-08 01:39Add initialization message for NullTransport default choice.Ed Bueler2+11-0
2013-09-05 06:23Need to not lose this idea.Ed Bueler1+1-0
2013-09-04 22:51I looked at the installation manual and found the section and subsection headings confusing. I changed them some. I am pretty sure this is an improvement, but further improvements are certainly possible.Ed Bueler1+66-79
2013-09-03 22:38I am confused by the two different ways the current time was turned into a string. One way is used in iMreport.cc (for stdout summary line) and one in iMutil.cc (for -USR1 file name). I chose the latter because it works correctly for current SeaRISE-Greenland script.Ed Bueler1+10-2
2013-08-24 06:03Removing unused options. Increase hydrology_tillwat_decay_rate_null from 1 mm/a to 1 cm/a; further increase to 5 cm/a worth considering.Ed Bueler2+8-16
2013-08-24 06:01Working on scripts for searise-greenland given new scheme using Tulaczyk relation to determine Ntil from Wtil. To consider: (i) bigger tillphi, (ii) bigger decay rate in -hydrology null, (iii) scale N0 with Po. Probably the last is the key.Ed Bueler3+7-6
2013-08-24 00:46A significant change to the model physics for till: Effective pressure in till is now computed by empirical (and exponential) formula from Tulaczyketal2000 relating water amount in till to effective pressure. Will generally affect all sliding models.Ed Bueler3+68-63
2013-08-24 00:43Fixing initialization and reporting for hydrology schemes.Ed Bueler2+16-11
2013-08-24 00:41Updating Test P related scripts.Ed Bueler2+15-6
2013-08-24 00:27Remove old regularization parameter Y_0 from exact solution.Ed Bueler2+9-10
2013-08-23 21:54Remove hydrology_englacial_porosity. Only phi_0 = hydrology_regularizing_porosity remains, and it is set by default to 0.01, noting 1% quoted in Bartholomaus.Ed Bueler4+9-25
2013-08-23 21:52Add tillwat to outputs.Ed Bueler1+3-3
2013-08-23 21:26Found and fixed serious bug in PISMRoutingHydrology::conductivity_staggered() which invalidates a lot of recent results. Made update of Wtil match notes. Made update of W match notes. Update of Wtil occurs before that of W (also matching notes). Removed unnecessary reporting in PISMRoutingHydrology::boundary_mass_changes(), and applied that routine to both W and Wtil.Ed Bueler3+145-137
2013-08-23 21:17Make omega=100, i.e. Wtil will tend to be 100 times if W is small, though Wtil is bounded and W is not and W+Wtil is conserved.Ed Bueler1+1-1
2013-08-23 21:15Made test script more flexible.Ed Bueler2+8-7
2013-08-23 20:35comment out fine-grid runsEd Bueler1+2-2
2013-08-23 01:04Single added FIXME in important location.Ed Bueler1+1-0
2013-08-23 00:46Fixed up test script so comparing apples to apples in -hydrology null and -hydrology routing. Runs are longish but easy to interpret. (To do list is: Fix FIXMEs in PISMRoutingHydrology.)Ed Bueler1+16-4
2013-08-23 00:42Made model in -hydrology null match notes (with additional constant decay rate only).Ed Bueler1+10-21
2013-08-22 23:40Made subglacial_water_pressure() pure virtual in base class PISMHydrology. It has the same implementation in PISMNullTransport hydrology for fundamentally different reasons.Ed Bueler4+24-13
2013-08-22 23:38Fixed hydrology test script so it has a minimal chance to work with current implementation.Ed Bueler1+7-5
2013-08-16 05:37PISMHydrology and derived are no longer in the business of modeling the till water pressure, but only the amount. This means PISMMohrCoulombYieldStress is broken (produces runtime error) but a fix is underway. (This isn't the dev branch anyway.)Ed Bueler7+34-118
2013-08-16 05:22Minor and superficial changes to references and scripts.Ed Bueler3+12-5
2013-08-15 02:55Schoof and Hewitt wrote a nice review article.Ed Bueler1+10-0
2013-08-12 08:03Another till/hydrology ref.Ed Bueler1+11-0
2013-08-12 03:40Fixing mis-use of bibtex.Ed Bueler1+9-10
2013-08-12 03:37Continuing to try to catch the literature of till and subglacial stuff.Ed Bueler1+40-5
2013-08-11 05:04Another hydrology ref. And, yes, ISO 4 says this is the right abbreviation for EPSL.Ed Bueler1+14-3
2013-08-06 02:13Simplified README.md to avoid maintenance. Remove mention of englacial water from run.sh.Ed Bueler2+9-17
2013-08-06 02:03Removed englacial storage, as conserved mass, from codes. But several new FIXMEs, especially related to: (i) move update of Wtil before W update; (ii) fix PISMNullTransport version of updating Wtil; (iii) what exactly *is* our model for pressure in the till pore-water?Ed Bueler8+29-196
2013-08-06 01:12Move testhydro.sh to examples/ directory because it is just play, not a test (where the answer is known).Ed Bueler2+7-2
2013-07-29 20:11Yet another hydrology ref.Ed Bueler1+12-0
2013-07-27 22:36merge dev into till-hydro-dynamicsEd Bueler321+50008-9545
2013-07-27 22:13Merge commit '40e07459' into till-hydro-dynamicsEd Bueler89+2009-2607
2013-07-27 22:09merge 6e5f47d into till-hydro-dynamicsEd Bueler18+630-526
2013-05-11 06:55in nbreen example we can (should?) turn off tillwat mechanismEd Bueler1+3-1
2013-05-11 05:12Update config list and a script to match revisions bwp-->tillwp and so on.Ed Bueler2+6-6
2013-05-11 04:55Mohr-Coulomb gets till_water_pressure().Ed Bueler3+47-57
2013-05-10 23:40Merge branch 'dev' into till-hydro-dynamicsEd Bueler88+1352-1217
2013-05-10 23:04Now it compiles. Results mostly make sense, but the major known issues are: (1) Wtil <--> W exchange in Routing can make W < 0; (2) yield stress calculation needs to read tillwp not bwp; (3) most runs like SeaRISE Greenland should switch to tillwp diagnostic not bwp, unless using -hydrology {routing,distributed}.Ed Bueler7+80-105
2013-05-10 20:20One remaining FIXME needs addressing in PISMDistributedHydrology.Ed Bueler1+10-19
2013-05-10 20:01Addressed some FIXMEs. Moved stuff into base class: Now PISMNullTransportHydrology and PISMRoutingHydrology share the exact same models for till and subglacial water pressure, and those are implemented in base class PISMHydrology.Ed Bueler5+65-131
2013-05-07 22:20through talking with Constantine, became aware of issues and added FIXMEs that break compilabilityEd Bueler4+24-31
2013-05-06 00:23this version compiles, but is untested; several parameters need setting in src/pism_config.cdlEd Bueler5+166-24
2013-05-05 22:45cleaning up init actions; still does not compile because not all needed methods have implementations, and because evolution of till is not in PISMRoutingHydrology::update()Ed Bueler3+61-50
2013-05-05 22:43PISMullTransportHydrology needs a bwat diagnostic method because it does not have bwat as a state variable (it has tillwat instead)Ed Bueler3+36-7
2013-05-05 22:02spliting routing from distributed ... these are long filesEd Bueler3+774-761
2013-05-05 19:53Refactoring. The base class PISMHydrology holds a Wtil variable because all models need somewhere to put water, and all models need to report what the Mohr-Coulomb relation needs. The new class PISMNullTransportHydrology updates Wtil by a new (more stable, but easy) implicit method. This new class can't conserve water and it does not attempt to report conservation. The "till can" aspect is de-emphasized but still present.Ed Bueler10+349-548
2013-05-03 00:20Merge branch 'dev' into till-hydro-dynamicsEd Bueler1+2-2
2013-05-02 23:16Merge branch 'dev' into till-hydro-dynamicsEd Bueler170+1199-1488
2013-05-02 23:14Merge branch 'dev' into till-hydro-dynamicsEd Bueler1+1-1
2013-05-02 18:18Move hydrology diagnostics into separate files. Implement tillwat and tillwp diagnostics. (But PISMHydrology code for these not implemented. Does not compile.)Ed Bueler5+342-216
2013-05-01 00:04Early work. Goal is to remove PISMTillCanHydrology and its derived. New PISMNullTransportHydrology to be implemented. Till dynamics will be a method available to all hydrologies. Thats the plan, anyway; this bit does not compile.Ed Bueler2+66-27
2013-04-30 22:32Cleaning up metadata in extra_files and ts_files.Constantine Khroulev3+14-9
2013-04-30 22:12Fixes a bug in the code handling units of time.Constantine Khroulev1+8-7
2013-04-29 23:56Replaces ncap with ncap2 in a preprocessing script.Constantine Khroulev1+8-8
2013-04-29 21:05Fixes a memory leak in IceModelVec2T::get_interp_context() and -verbose 5 I/O reporting.Constantine Khroulev1+7-3
2013-04-29 19:51Fixes a very minor bug in CalCalcs.Constantine Khroulev1+1-1
2013-04-29 18:46Closes #141 and #152.Constantine Khroulev16+75-695
2013-04-29 16:32Closes #147.Constantine Khroulev8+161-149
2013-04-28 07:45also enable 'wallmelt' diagnostic for distributed hydrologyEd Bueler1+1-0
2013-04-28 06:54remove feedback of wall melt; the result of wall_melt() is ONLY used diagnosticallyEd Bueler6+48-45
2013-04-27 18:26obviously, I should remove thisEd Bueler1+0-21
2013-04-27 18:17updating scripts so they seem to run; see issue #152Ed Bueler2+7-4
2013-04-27 17:49Merge branch 'dev' into try-new-putatdepthEd Bueler28+644-241
2013-04-27 06:57documenting new putTempAtDepth() methodEd Bueler1+14-11
2013-04-27 06:34cleaning up and continuing testing ... looks good and both options are availableEd Bueler3+27-25
2013-04-27 04:57This version apparently works. And now I know that the widespread appearance of the NE Greenland ice stream in other badly spun-up models ... is an artifact of their bootstrapping method.Ed Bueler3+104-82
2013-04-24 01:48thinking it through before writing code; without actually trying it I don't know if this is a good idea; obviously I'll remove the .m before merging into devEd Bueler2+68-1
2013-04-24 01:30Merge branch 'dev' of github.com:pism/pism into devEd Bueler1+23-10
2013-04-24 00:01minor: new manual text on time looks great (and this commit is tiny)Ed Bueler1+6-3
2013-04-23 22:18Merge branch 'udunits2' into devConstantine Khroulev157+4846-8216
2013-04-23 22:06minor: vbed not used so remove begin/end_access(); modernize commentsEd Bueler1+17-24
2013-04-22 05:27yet another hydrology refEd Bueler1+13-0
2013-04-22 02:52another hydrology refEd Bueler1+11-0
2013-04-21 18:55minor: uniformity for book reference styleEd Bueler1+11-13
2013-04-21 01:23a ref on tillEd Bueler1+12-0
2013-04-20 07:56Update python-side call to SIAFD::update (strain heating argument is now gone).David Maxwell1+1-5
2013-04-20 07:52Merge branch 'dev' of https://github.com/pism/pism into devDavid Maxwell380+1257-768
2013-04-20 07:52Fixed formula error in LogRatioFunctional.cc.David Maxwell1+2-3
2013-04-17 04:08Better diagnostic reporting for inverse misfits log_ratio and log_relativeDavid Maxwell2+8-98
2013-04-17 01:45added -inv_ssa_misfit log_ratio and -inv_ssa_misfit log_relativeDavid Maxwell7+432-7
2013-04-12 16:06Fixes the build system.Constantine Khroulev24+36-50
2013-04-11 23:15Updates test/vfnow.py and test/vnreport.py. See #147.Constantine Khroulev2+126-169
2013-04-11 20:34Re-organizes the src/stressbalance directory and updates the test G regression script.Constantine Khroulev47+97-69
2013-04-11 19:21Merge branch 'blatter' into devConstantine Khroulev42+3113-2564
2013-04-09 19:29simplify and remove broken linkEd Bueler1+4-8
2013-04-08 21:41Closes #91.Constantine Khroulev3+42-6
2013-04-05 22:26Cleaning up doxygen comments.Constantine Khroulev53+419-426
2013-04-05 22:24Addressing #91.Constantine Khroulev2+65-79
2013-04-04 19:31Fixes #134 (in dev; -eigen_calving *requires* -part_grid).Constantine Khroulev1+4-0
2013-04-04 18:25One more fix for the same bug.Constantine Khroulev1+3-0
2013-04-04 18:13This should fix a (recently introduced) bug in the lapse_rates modifier.Constantine Khroulev1+1-7
2013-04-04 16:54Updates the documentation for "-surface pdd" (in dev).Constantine Khroulev3+81-71
2013-04-03 22:03Cleaning up the doxygen browser build warnings.Constantine Khroulev14+80-80
2013-04-02 22:39Interpret climate fields as piecewise-constant in time.Constantine Khroulev15+203-258
2013-04-02 12:14PDD model: melted ice should not refreeze (port to 0.5)Julien Seguinot1+1-1
2013-04-01 21:13Fixes a typo at iceModelVec2T.cc:246Constantine Khroulev1+1-1
2013-04-01 16:57Classes derived from PGivenClimate use IceModelVec2T::average() in update().Constantine Khroulev4+8-8
2013-04-01 16:02Now periodicity parameters are arguments of IceModelVec2T::init().Constantine Khroulev28+96-98
2013-03-30 17:31clobber 2 of 3 pedantic warnings; see remaining in PA_paleo_precip.ccEd Bueler3+3-3
2013-03-29 23:45The C/C++ standard math library provides erfc().Constantine Khroulev1+2-2
2013-03-28 14:27Fixing more atmosphere models...Constantine Khroulev11+109-134
2013-03-27 20:35Removes -paleo_precip from "-atmosphere searise_greenland".Constantine Khroulev4+10-110
2013-03-26 22:35Improves the PDD model and adds "-atmosphere ...,paleo_precip" (in dev).Constantine Khroulev10+398-191
2013-03-26 17:01Improving atmosphere models... (in dev).Constantine Khroulev16+98-64
2013-03-27 00:04Inversion allows a design norm for tau_c that omits elements with floating ice. Use -inv_ssa_grounded_ice_taucDavid Maxwell6+279-2
2013-03-22 18:00Re-wrote a good chunk of the PDD code. (I think I'm happy with it now.)Constantine Khroulev9+400-403
2013-03-22 01:42Fixes a bug in lapse rate correction codes (port to 0.5).Constantine Khroulev1+29-27
2013-03-20 22:03Fixes a bug in IceModelVec2T::init_interpolation(). (Port to 0.5.)Constantine Khroulev1+2-2
2013-03-21 23:03something fun ... doesn't do anything yet but you can admire the pretty picturesEd Bueler2+117-0
2013-03-21 02:27- deleted deprecated option from scriptAndy Aschwanden1+1-1
2013-03-20 20:44Fixes a bug: "pismr -i i.nc -age" stopped if i.nc does not have the "age" variable. Port to 0.5.Constantine Khroulev1+4-0
2013-03-19 23:02Fixes units and metadata in PA_delta_P.Constantine Khroulev1+2-2
2013-03-19 17:46Closes #121.Constantine Khroulev11+218-162
2013-03-18 19:44More fixes (snow_depth==0 in the ocean; fixes memory allocation in PO_delta_SL).Constantine Khroulev4+22-7
2013-03-18 19:06More fixes (proper implementations of precip_time_series needed by the new PDD code).Constantine Khroulev4+33-15
2013-03-07 23:15Now we do the right thing when interpolating periodic data. See #139.Constantine Khroulev30+427-320
2013-03-13 22:01Moves memory allocation into constructors of (almost) all couplers. See #140.Constantine Khroulev69+1120-627
2013-03-19 18:55add ability to show advective flux q = V W and to shorten (crop) the largest arrows, noting color still reflects true magnitudeEd Bueler1+43-11
2013-03-19 18:54with closure of #138, remove commentEd Bueler1+1-2
2013-03-19 16:31Closes #138.Constantine Khroulev1+12-5
2013-03-18 17:59plot script in good shapeEd Bueler1+39-23
2013-03-15 16:27use argparse; play with transposeEd Bueler1+60-45
2013-03-15 05:49a code to make a colored quiver of water velocity from last frame of bwatvel[2] in .nc fileEd Bueler1+91-0
2013-03-11 06:30slight revision doubles Delta t_P, the pressure equation time step restriction in -hydrology distributedEd Bueler1+7-8
2013-03-10 20:25improved reporting on velocity and diffusivityEd Bueler2+27-35
2013-03-08 23:27Remove hard-wired Glen exponent from ThermoGlenIce.Constantine Khroulev1+1-2
2013-03-08 01:38-report_mass_accounting stdout reports time step in model hoursEd Bueler1+3-2
2013-03-07 04:42Merge branch 'dev' into hydro-fix-kEd Bueler8+185-68
2013-03-07 04:42Update of Test P to match current version of manuscript and current version of Darcy flux relation with 'k'. The issue that Test P uses an unrealistically low conductivity of ~10^-6 remains; compare k=0.01 in Schoof/Hewitt work. Fixes regression test for Test P.Ed Bueler3+24-22
2013-03-05 18:23enough of a start to break things, but not fix things; so it is a feature branchEd Bueler2+14-3
2013-03-04 20:17Disabled the PDD regression test.Constantine Khroulev1+0-5
2013-03-01 17:12Fixes the PDD code. See #133. Needs testing.Constantine Khroulev20+358-148
2013-03-01 17:11Adds Dukowicz, Price, and Lipscomb (Consistent approximations...)Constantine Khroulev1+11-0
2013-03-04 16:20Merge branch 'dev' of github.com:pism/pism into devJulien Seguinot0+0-0
2013-03-04 15:33Added a PDD model test (details below)Julien Seguinot2+39-0
2013-03-04 15:28Added a Python PDD modeling toolJulien Seguinot1+268-0
2013-03-02 04:01I had gotten confused over the units of, and the size of, the hydraulic conductivity constant. As a result I was working with really low-conductivity hydrology which implied unreasonably high W values (water thicknesses) and very fast run times. With this commit the code (but not all the test cases) fully match the Schoof et al [2012] and Hewitt et al [2012] conventions. The examples/nbreen/run.sh script and its override parameters in examples/nbreen/nbreen_config.cdl also match the Schoof/Hewitt work. The model results for Nbreen then seem very reasonable relative the Schoof/Hewitt work. (The results are more expensive computationally, but also more interesting physically! The documentation in the private hydrolakes/ repo has not been fully updated to match yet.) The Test P example cannot be used until it is inspected and updated; I think all we need to do is shorten the run.Ed Bueler2+47-47
2013-03-02 03:45minor: edit of comment to match notationEd Bueler1+2-2
2013-03-01 01:41minor clarification of choices relative to literatureEd Bueler1+3-2
2013-03-01 01:26Clarify units of k. Switch Nbreen example over to Hewitt et al (2012) Table 1 parameter values for alpha, beta, k, W_r = h_r. Looks good ... though its not clear what that means.Ed Bueler2+14-6
2013-03-01 00:17Merge branch 'dev' into hydro-betaEd Bueler16+151-151
2013-03-01 00:07Put computation of max(KW) in right spot. Avoid unnecessary computation in beta=2 case. Recovers full speed and exact regression on Test P.Ed Bueler2+65-44
2013-02-28 23:18This form builds and runs. Only tested with beta=2 so far. Results from Test P are slightly different in value and slower (perhaps a lot slower). I think the change that happened in the beta = 2 case is the change in the max(K W) code in PISMRoutingHydrology::adaptive_for_W_evolution().Ed Bueler4+87-40
2013-02-28 07:53refactoring so that K is out of flux expressions and in its own routine, and adding power beta; does not compileEd Bueler2+112-71
2013-02-22 01:30update optionsEd Bueler4+133-133
2013-02-19 19:12Adds "References" to the table of contents of the User's Manual.Constantine Khroulev1+2-0
2013-02-12 18:36minor: consistency and notes on issuesEd Bueler3+10-5
2013-02-11 05:20Renamed '-hydrology lakes' to '-hydrology routing' and PISMLakeHydrology->PISMRoutingHydrology. This rename reflects/suggests how the model works rather than the historical reference to what it has been used for. Completes and closes issue #135.Ed Bueler8+66-62
2013-02-10 22:26add a hydrology test case in the right placeEd Bueler2+111-21
2013-02-10 20:27minor: comments and doc strings and whitespaceEd Bueler2+8-5
2013-02-10 01:01changed code so that Wen=F(Wtot) is continuous; better Nbreen results; documented minimallyEd Bueler2+25-12
2013-02-10 01:00I can easily to 62 m runs overnightEd Bueler1+10-2
2013-02-09 03:53Merge branch 'dev' into jako-hydrologyEd Bueler1+24-0
2013-02-09 03:48this is minor, but there was no earlier reason to move itEd Bueler1+2-1
2013-02-09 03:43I think this rule captures the mass-conserving transfer to/from englacial storage.Ed Bueler2+27-16
2013-02-08 00:16this essentially is a set of notes (i.e. the diff is a set of notes) for the emergent mass conservation issue in the distributed hydrology model when there is 'actual' englacial storageEd Bueler1+7-12
2013-02-06 06:05insert a hydrology test case; this is hijacking the main purpose of century.shEd Bueler1+29-5
2013-02-06 05:58updating Test P results in regression; almost no changeEd Bueler1+4-4
2013-02-06 04:29Test P clarified; corresponds to zero englacial storage (thus hydrology_englacial_porosity=0) and 'normal' amount of regularizing_porosity; runs fast enough so we can go to 641x641 gridEd Bueler3+16-4
2013-02-06 04:27reimplemented porosity in this sense: now hydrology_englacial_porosity is the 'true' porosity associated to englacial storage and mass conservation, while the sum hydrology_englacial_porosity+hydrology_regularizing_porosity is the diffusive regularization of the pressure equation; merely runs slowly if hydrology_regularizing_porosity=0Ed Bueler3+27-17
2013-02-06 01:36clarifying that W_en=0 where W=0; and minor stuffEd Bueler3+12-8
2013-02-05 22:58found bug ... verification results now just fineEd Bueler1+3-2
2013-02-05 21:46added physics associated to phi and englacial connection; nontrivial FIXMEs remain in PISMDistributedHydrology.cc because mass transfer not implemented and because mass accounting not completed; verification results don't match and need examinationEd Bueler4+49-20
2013-02-05 20:14minor: naming and code for check W,Wen>=0 fixedEd Bueler2+11-11
2013-02-05 19:54Merge branch 'dev' into hydrology-englacialEd Bueler1+5-2
2013-02-05 19:54compiles; successfully writes diagnostic enwat in -hydrology lakes; note many FIXMEs in PISMDistributedHydrology::update(), because at this stage the actual model with W_en is still not implementedEd Bueler4+110-19
2013-02-05 18:31new englacial water thickness diagnostic called 'enwat'Ed Bueler4+34-5
2013-02-05 07:40does not compile in this condition ... needs PISMDistributedHydrology.cc to actually have W_en state variable and have the mass-conservation part of update work correctly and make the change E_0/H --> phiEd Bueler6+51-24
2013-01-30 01:02As per email with Ward: Replaced K(W) with k W^(alpha-1), so K(W) W is replaced by k W^alpha. This is part of the Manning-Weisbach form in Schoof, but only beta=2 case. Verification results identical because k=K_0,alpha=1. Nbreen results very similar with k=1.0e-3 and alpha=2.0. Observe that with alpha>1 we have a degenerate diffusion so that water doesn't spread out so much (e.g. far under non-melting base); I think this is good.Ed Bueler6+56-64
2013-01-29 23:14Documentation for the classes in TaoUtil.{hh,cc}David Maxwell2+271-135
2013-01-24 22:01Improved success criteria for inversion tests.David Maxwell4+36-20
2013-01-24 19:00I have found a helper script useful so let's give it to PISM users.Ed Bueler3+83-0
2013-01-23 22:58Merge branch 'dev' of https://github.com/pism/pism into devDavid Maxwell21+839-520
2013-01-23 05:32Fixed missing end_access.David Maxwell1+1-0
2013-01-23 05:29Allow python-based regression tests to pick up user's PYTHONPATH.David Maxwell7+7-7
2013-01-18 01:10Fix to tikhonov regression testDavid Maxwell1+1-9
2013-01-18 00:53Merge branch 'dev' of https://github.com/pism/pism into devDavid Maxwell3+108-92
2013-01-18 00:53Added a TAO-based inversion regression testDavid Maxwell5+98-43
2013-01-16 03:17Updating runs to use time-dependent input. Runs have uniform interface. Doc updated.Ed Bueler2+45-28
2013-01-16 02:59A straightforward simplification and removal of code duplication. Now there is only one method for getting the input rate.Ed Bueler3+9-51
2013-01-16 01:34Forgot to remove PISMDistributedHydrology::total_input so it does not shadow PISMHydrology::total_input. (This commit allows prev. commit to run and pass test_29.py.)Ed Bueler1+1-1
2013-01-16 01:12Merge branch 'dev' of https://github.com/pism/pism into devDavid Maxwell6+171-104
2013-01-16 01:12toproczero.py now uses IceGrid::get_dm as IceGrid::da2 is no longer exposed.David Maxwell1+2-16
2013-01-16 01:11Initial pass at adding a total variation-type norm for Tikhonov inversions.David Maxwell8+227-9
2013-01-14 22:50Adding documentation to 'inverse' subdirectory.David Maxwell3+61-3
2013-01-14 20:34Realized IceModel had better call PISMHydrology::max_timestep() or else it won't run. Wrote a helper macro to avoid code duplication when updating time step from submodel's restrictions.Ed Bueler2+17-24
2013-01-14 19:58Merge branch 'dev' of github.com:pism/pism into devEd Bueler17+64-39
2013-01-14 19:57built an -input_to_bed_file example that runs ... suspiciously easy ... does not have desired effect ... debugging ...Ed Bueler3+69-0
2013-01-14 18:23more neatening up in this directoryEd Bueler3+4-3
2013-01-14 07:52Various edits without change to functionality or major addition of documention. Except I am starting to revise subsection 5.14 to reflect recent additions of hydrology.Ed Bueler3+64-34
2013-01-13 21:15Renamed 'input' to 'total_input' to clarify that it is sum of bmelt and (possible) external (en/supraglacial) input to the bed. There are two versions of the method that fills this field, one which assumes that the only input is bmelt, and that it is independent of time during the time step. The other version get_input_rate_time_varying() reads the additional time-dependent input from IceModelVec2T *inputtobed. The code in various update() methods avoid calling get_input_rate_time_varying(), which is more expensive, unless inputtobed is actually present.Ed Bueler4+135-47
2013-01-13 08:02append to prev commit: neaterEd Bueler1+9-2
2013-01-13 08:00neatening upEd Bueler3+6-1
2013-01-13 07:37Starting to add code that allows time-dependent input to bed through option '-input_to_bed foo.nc' and member IceModelVec2T inputtobed. See FIXMEs for to do. Compiles. Function is unaltered.Ed Bueler3+84-12
2013-01-12 04:54added diagnostic 'effbwp' = P_o - PEd Bueler4+39-1
2013-01-12 04:32added diagnostic 'bwprel'=P/P_oEd Bueler4+57-9
2013-01-11 21:05trivial renameEd Bueler1+1-1
2013-01-11 20:49Merge branch 'bwp-in-hydro' into devEd Bueler41+2188-791
2013-01-10 20:43Fixes a bug in IceModelVec2T introduced while mergin mem_usage into dev (commit eaa485841abd077c3f70fcfc03c918cb7cc47499).Constantine Khroulev1+3-3
2013-01-09 22:47Moves exactV.py.Constantine Khroulev2+2-0
2013-01-09 18:22Widened the valid range for IceModelVec2Vs used in inversion scripts.Constantine Khroulev1+1-1
2013-01-08 19:47Fixed a bug in fill_missing.py (port to 0.5).Constantine Khroulev1+4-2
2013-01-08 19:43Closes #122.Constantine Khroulev41+223-430
2013-01-08 18:28Closes #124.Constantine Khroulev7+57-56
2013-01-07 00:02- cleaned-up script, added some comments, made variable names consistent with PISM.Andy Aschwanden1+19-13
2012-12-20 22:56- fixed a bug in calculating time axis. The bug, however, did not affect any simulations, as PISM reads the time_bounds instead, and time_bounds were calculated correctly.Andy Aschwanden1+23-15
2012-12-19 21:37Fixes a memory leak. Port to 0.5.Constantine Khroulev2+19-18
2012-12-18 00:51Fixes the high diffusivity issue (documented here).Constantine Khroulev1+23-4
2012-12-13 02:27Merge branch 'mem_usage' into dev. Closes #119.Constantine Khroulev29+236-198
2012-12-12 18:19Closes #97.Constantine Khroulev2+28-13
2012-12-12 16:24Do not allow omitting -Lz at bootstrapping. Closes #94.Constantine Khroulev3+4-26
2012-12-12 16:15Check if grid.Lz >= max(thk). Port to 0.5.Constantine Khroulev1+11-0
2012-12-12 16:06Use enthalpy of *ice* (not water) in IceFlowLawUsesGrainSize().Constantine Khroulev1+1-1
2012-12-11 23:00Allow global->local scatters using IceModelVec::beginGhostComm()/endGhostComm(). See #109.Constantine Khroulev1+38-14
2012-12-11 21:27Fixes #114.Constantine Khroulev5+44-54
2012-12-11 17:50Closes #113.Constantine Khroulev13+128-93
2012-12-10 19:45Fixes the recent "-options_left" code.Constantine Khroulev1+9-3
2012-12-10 19:45A little fix (bwat_max -> hydrology_bwat_max) and some cleanup.Constantine Khroulev2+9-15
2012-12-10 19:45NCConfigVariable keeps track of unused parameters. See #110 and port to 0.5.Constantine Khroulev4+230-93
2012-12-10 19:37Updated the signature of the IceBasalResistancePlasticLaw constructor.Constantine Khroulev16+47-92
2012-12-08 04:32Removed the "-f3d" option. (When was the last time someone used it?)Constantine Khroulev3+0-9
2012-12-09 20:34obvious minor rename to unclutter rootEd Bueler1+0-0
2012-12-09 19:48presumably final fiddles with looking at results from '-hydrology lakes' for AntarcticaEd Bueler1+49-25
2012-12-09 08:24more initialization, and the cleaning of the input rate (bmelt for now) can be done in more basic classesEd Bueler4+125-135
2012-12-09 07:14update to generate .png and show multiple resolutionsEd Bueler1+38-26
2012-12-09 06:28first draft of little script to clearly display result from showlakes.pyEd Bueler1+41-0
2012-12-09 01:39Made hydraulic conductivity (K) a command line option. lakestest.sh should recreate 'toy Antarctic' results from my IGS talk in June 2012.Ed Bueler3+57-0
2012-12-08 03:37Moves "bwp" to a PISMDiagnostic derived class.Constantine Khroulev5+70-76
2012-12-08 03:32Cleans up PISMTillCanHydrology::water_pressure().Constantine Khroulev1+23-10
2012-12-08 03:30PISMMohrCoulombYieldStress has no business writing "bwp" now that we have a hydrology model component.Constantine Khroulev2+0-77
2012-12-08 02:54Better naming scheme (I think) for configurable hydrology parameters. Added option to use a time- and space-independent constant as the input into the hydrology submodel; this is useful for testing.Ed Bueler4+56-34
2012-12-07 23:09Fixes a memory leak in IceModelVec2T (port to 0.5).Constantine Khroulev2+11-3
2012-12-07 22:37Restores the ability to save "taud" and "taud_mag" diagnostics in SIA-only runs.Constantine Khroulev10+230-87
2012-12-06 19:30defaults to thirty km for quick development runsEd Bueler1+5-3
2012-12-06 23:48First running implementation of 'lakes' hydrology. Removed PISMHydroloy's mask and replaced with standard PISM mask; much better that way.Ed Bueler2+133-65
2012-12-06 23:46update test script to new options and 20a runsEd Bueler1+7-7
2012-12-06 21:42Made K = 'subglacial_hydrologic_conductivity' configurable in distributed hydrology models. Improved config parameter doc strings.Ed Bueler4+32-19
2012-12-06 20:52Option '-hydrology' sets configuration parameter 'hydrology_model' as one sub-model from 'tillcan', 'diffuseonly', 'lakes', 'distributed'. (C: nice option setupgit st)Ed Bueler3+12-36
2012-12-06 20:48better name for classEd Bueler2+11-11
2012-12-06 20:47append to prev commit: rename scheme known to browserEd Bueler1+1-0
2012-12-06 20:15moving hydrology sources to unsurprising locationEd Bueler4+3-2
2012-12-06 05:43The 'subglacial lakes' hydrology model is a natural base for the more compute van Pelt & Bueler distributed model. In the lakes model the water pressure is a fixed fraction of overburden, but many concepts are the same, and the numerical is nearly a subset. Thus here: PISMDistributedHydrology is a derived class of PISMLakesHydrology with additional nontrivial pressure-related storage and computations. Here PISMLakesHydrology::update() is a stub. Compiles but neither run, presumably.Ed Bueler3+469-220
2012-12-06 04:03split source file into old and new models; a few clarifying comments; no functional changeEd Bueler3+613-585
2012-12-06 03:53descriptive commentsEd Bueler1+37-8
2012-12-05 23:04the constructor signature for PISMMohrCoulombYieldStress changed, as did its derived classEd Bueler5+9-7
2012-12-05 23:03now PISMMohrCoulombYieldStress has ghosted space for its own copy of bwat, and asks PISMHydrology to copy_to() it in the usual way by calling water_layer_thickness()Ed Bueler2+21-5
2012-12-05 22:35Merge branch 'dev' into new-hydroEd Bueler11+117-153
2012-12-05 22:31these small changes resolve the initialization of PISMHydrology so that it owns 'bwat' but other components (e.g. yield stress calculations) can get to itEd Bueler3+18-16
2012-12-05 21:29Sorry, I see this is the fix. Closes issue #108.Ed Bueler2+4-4
2012-12-05 07:13trying to change initialization order to fix issueEd Bueler1+8-8
2012-12-04 23:40This version has hydrology model fully moved into PISMHydrology objects PISMTillCanHydrology and PISMDiffusebwatHydrology. It passes existing regression tests, but needs further regression testing to see if (e.g.) effect on SeaRISE-Greenland is either small or reasonable.Ed Bueler12+52-185
2012-12-04 23:03need to use mask the same way as prevEd Bueler2+39-4
2012-12-04 22:30This time I understand the test results (i.e. from testhydro.sh). Differences at level of rounding error expected, and seen.Ed Bueler1+15-4
2012-12-04 22:10Previous test results were ... too good to be true. Rethinking.Ed Bueler1+1-1
2012-12-04 21:51Merge branch 'dev' into new-hydroEd Bueler20+343-352
2012-12-04 21:40The new PISMHydrology objects have been running redundantly with old (core) IceModel functionality. I can now assert, both for EISMINT II type runs and for SeaRISE-Greenland spinup runs, that the PISMTillCanHydrology and PISMDiffusebwatHydrology objects reproduce core (old) functionality.Ed Bueler1+9-3
2012-12-04 20:19forgot: arithmetic is a lot easier if 'bwat-tillcan' is not used as a variable name (i.e. no hyphens in var names)Ed Bueler1+29-30
2012-12-04 19:40the way I want to testEd Bueler1+12-3
2012-12-04 19:39dealing with Wnew to W ghost transfer; why not allow global to local here?Ed Bueler2+16-2
2012-12-04 18:56fixed my allocation-in-derived-class-with-different-stencil problem in a possibly ugly wayEd Bueler4+18-24
2012-12-04 03:44initial implementations of PISMTillCanHydrology and PISMDiffusebwatHydrology, which are the old technology; also a test script; does not compile because there is no clear way to revise an IceModelVec::create() call from a derived classEd Bueler7+378-34
2012-12-04 00:07initialize IceModel::subglacial_hydrology, which points to PISMHydrology instance, according to options -diffuse_bwat, -lakes_hydrology, -distributed_hydrology, or noneEd Bueler3+40-4
2012-12-03 23:53comments updatedEd Bueler1+7-0
2012-12-03 23:53Plan for the moment: PISMHydrology::update() updates internal hydrology model state, and we call water_layer_thickness() to update IceModel::vbwat. This design can change.Ed Bueler2+10-8
2012-12-03 23:43new branch for hooking-up stub hydrology. Here I create a virtual base class. Will have four derived classes of PISMHydrology: PISMTillCanHydrology = (Bueler&Brown(2009) without contrived diffusion), PISMDiffusebwatHydrology = (Bueler&Brown(2009) WITH contrived diffusion), PISMLakesHydrology = (most-minimal distributed hydrology as promised by Bueler at IGS2012), PISMDistributedHydrology = (under-development van Pelt & Bueler model).Ed Bueler2+56-36
2012-12-03 22:00Cleanup: removes ivoltempf, ivolcoldf, iareatempf, iareacoldf time-series.Constantine Khroulev3+0-168
2012-12-03 21:36Making sure that "-list_diagnostics" works correctly...Constantine Khroulev1+15-10
2012-12-03 20:33Documents new disgnostics. Closes #98. Closes #101.Constantine Khroulev1+25-6
2012-12-03 17:32Implements cumulative grounded, floating, and nonneg rule fluxes. See #101.Constantine Khroulev5+194-3
2012-12-01 01:46IceModel has a PISMHydrology* member but it is not allocated yet. Need to decide on how to initially hook up and test.Ed Bueler3+24-0
2012-12-01 01:42implementing obligatory PISMComponent methodsEd Bueler2+41-9
2012-12-01 01:39this may fix a small memory leakEd Bueler1+1-0
2012-12-01 00:58I have thought through the issues of parallel communication and how to get the basal sliding velocity in the right way. What do you think?Ed Bueler2+122-84
2012-11-30 07:15adaptive time stepping uses absmax() from IceModelVec2Ed Bueler1+16-10
2012-11-30 06:12I find these absmax() functions useful ... is that enough reason to implement them? did I do it right?Ed Bueler2+38-0
2012-11-29 23:45increasing amounts of the model equations, but this is still a stub; needs parallel communication; compilesEd Bueler2+197-28
2012-11-29 21:49work on the hydrology stub; compilesEd Bueler2+131-28
2012-11-29 18:35Closes #96 (plus some cleanup).Constantine Khroulev25+406-319
2012-11-29 01:43more initial implementingEd Bueler2+68-63
2012-11-29 01:13with this change I get graphs in the util browser; see 'diff doxyfile_base doxyfile_util' before and after this changeEd Bueler1+2-2
2012-11-28 23:43an actual start on the PISMHydrology class: handles allocation and initialization; starts to implement vanPelt-Bueler model which will become derived class; other derived class is old diffuse-bwat; not hooked-up and thus not runableEd Bueler4+247-49
2012-11-28 21:59this is the way the grid should be chosen in a scriptEd Bueler1+2-1
2012-11-28 03:15- added a new python script, create_timeline.py, to create a timeline for forcing with the -time_file command-line option. Doesn't contain actual forcing data, though.Andy Aschwanden1+118-0
2012-11-27 18:37Merge branch 'dev_glbug_solved' into dev. See #79.Constantine Khroulev5+45-57
2012-11-26 17:13Updates the long-forgotten ssa_test_cfbc. Adds a regression test.Constantine Khroulev5+82-31
2012-11-23 21:46Fixed the grid size check (large grids with -o_format quilt are OK).Constantine Khroulev1+4-3
2012-11-23 21:33Fixed a bug reported by Julien Seguinot (via e-mail; port to 0.5).Constantine Khroulev1+1-0
2012-11-23 00:31Merge branch 'improving-io' into devConstantine Khroulev117+3608-926
2012-11-20 01:31a hydrology refEd Bueler1+11-0
2012-11-17 07:32a general referenceEd Bueler1+8-0
2012-11-10 04:11Re-implements "-gradient haseloff" (and adds some fixes).Constantine Khroulev11+241-184
2012-11-14 23:43Fixed verification test L. (Port to 0.5.)Constantine Khroulev1+10-6
2012-11-13 00:29a ref which is important to no one but meEd Bueler1+9-0
2012-11-09 08:35nothing very important, but now my trivial experiment at least gives perfect conservation ...Ed Bueler2+21-15
2012-11-08 08:29drafting hydrology files; not functional; for discussionEd Bueler3+171-3
2012-11-08 03:03cleaning up; this one is still in the repo but not worth keeping in base/Ed Bueler11+0-635
2012-11-08 02:42consistent subdirectory names ... not a big thing I knowEd Bueler10+14-13
2012-10-31 23:36Minor fixes (port to 0.5).Constantine Khroulev2+24-1
2012-10-30 17:56Trying to fix Marijke's build issue.Constantine Khroulev1+2-1
2012-10-30 02:03A fix in PISMProf. (I think this should help some...)Constantine Khroulev1+4-3
2012-10-30 01:41Playing with pism_netcdf_test.Constantine Khroulev1+28-1
2012-10-29 21:03Renames a variable at NCVariable.cc:283 and later. (Port to 0.5.)Constantine Khroulev1+3-3
2012-10-29 20:53Merge branch 'ocean_kill_flux_2D' into dev. See #98.Constantine Khroulev13+306-130
2012-10-29 20:34new langen et al refEd Bueler1+11-0
2012-10-29 04:28referencesEd Bueler1+27-2
2012-10-17 01:02Documents "-extra_times monthly" and similar. Closes #90.Constantine Khroulev2+12-2
2012-10-17 01:01Updates MISMIP figures in the User's Manual.Constantine Khroulev3+0-0
2012-10-16 22:13Deals with #90. (Needs more testing and documentation, though.)Constantine Khroulev5+58-8
2012-10-16 18:54Merge branch 'grounding-line-driving-stress' into devConstantine Khroulev3+132-16
2012-10-16 18:44Making sure that forcing.pdf gets installed (port to 0.5).Constantine Khroulev1+1-0
2012-10-16 18:34Fix in the build system (port to 0.5).Constantine Khroulev1+2-0
2012-10-15 21:52Merge branch 'dev' of github.com:pism/pism into devAndy Aschwanden3+31-7
2012-10-15 21:52- updated examples, but needs testing before porting to stable.Andy Aschwanden4+26-27
2012-10-12 17:43Some code cleanup.Constantine Khroulev2+37-33
2012-10-09 21:29Updates the computation of principal strain rates.Constantine Khroulev1+22-28
2012-10-11 21:14Updates the plot.py command in the Ross section of the manual (port to 0.5).Constantine Khroulev1+1-1
2012-10-11 17:38Adds a sanity check (port to 0.5).Constantine Khroulev1+51-1
2012-10-10 20:15Minor fix (port to 0.5).Constantine Khroulev2+3-4
2012-10-11 17:47Very minor fixes (port to 0.5).Constantine Khroulev3+4-5
2012-10-10 23:31- updated ross plot script to argparse, please port to stable 0.5Andy Aschwanden1+16-17
2012-10-06 20:02some stuff got publishedEd Bueler1+6-4
2012-10-04 18:29Fixed the code in commit 7974bf56 (the hard-wired path to pism_config.nc was wrong).Constantine Khroulev1+5-4
2012-10-04 17:16Closes #80.Constantine Khroulev7+54-59
2012-10-04 16:02Merge branch 'flux-accounting' into devConstantine Khroulev36+1037-736
2012-10-04 15:14Adds a CMake macro protecting users from themselves (a little bit).Constantine Khroulev2+22-8
2012-10-02 21:42Restoring the ability to build PISM without FFTW-3. This should be ported to 0.5.Constantine Khroulev3+24-19
2012-10-01 18:35Adds "-extra_vars strain_rates".Constantine Khroulev3+48-2
2012-10-01 04:21correct LeBrocq refEd Bueler1+10-7
2012-09-25 21:25- nc2cdo.py now adds the global attribute 'projection' to a file.Andy Aschwanden1+2-0
2012-09-21 18:12Regression tests use unique file names.Constantine Khroulev35+153-153
2012-09-20 19:57Fixes PIO::inq_grid() and such. (Thanks, David!)Constantine Khroulev3+27-16
2012-09-19 21:08Removes $@ from vel2tauc.py arguments.Constantine Khroulev1+1-1
2012-09-19 19:47Adds a very simple inversion regression test.Constantine Khroulev3+173-0
2012-09-19 15:47One more PAGivenClimate fix.Constantine Khroulev1+3-1
2012-09-19 15:09A one-line change needed to make the dev version compatible with PETSc 3.2 again.Constantine Khroulev1+2-0
2012-09-18 20:40Closes #75 (PISM uses GSL's CBLAS now).Constantine Khroulev2+1-13
2012-09-18 17:58Fixes addressing #85.Constantine Khroulev6+69-37
2012-09-17 19:33Code cleanup and minor fixes to maintain PETSc 3.3 compatibility.Constantine Khroulev3+20-19
2012-09-17 19:09Closes #86.Constantine Khroulev2+13-7
2012-09-13 22:28Fixes and improvements to verify_ssa_inv.pyDavid Maxwell1+5-6
2012-09-13 22:08Added script for verifying the misfit and iteration count of an inversion match expectations.David Maxwell3+90-8
2012-09-13 20:53Fixes related to recent merge from dev.David Maxwell1+6-8
2012-09-13 20:27Merge branch 'dev' into tikhonovDavid Maxwell135+3121-1586
2012-08-17 19:23Fix for writing misfit history on multiple processors.David Maxwell1+14-12
2012-08-17 19:04Merge branch 'tikhonov' of https://github.com/pism/pism into tikhonovDavid Maxwell4+67-32
2012-07-30 04:29tikhonov_gn robustness during linesearchesDavid Maxwell1+6-7
2012-07-27 22:54vel2tauc supports explicitly provided driving stresses and explicitly specified periodicitiesDavid Maxwell9+55-29
2012-07-26 00:18Adding linesearch to TikhonovGN.David Maxwell4+66-11
2012-07-14 20:06proper convergence testing for TikhonovGN and adaptive parameterDavid Maxwell4+116-35
2012-07-12 06:46Adaptive tikhonov parameter uses logarithmic scale.David Maxwell2+77-60
2012-07-11 23:41Working on adaptivity in selecting Tikhonov parameter.David Maxwell6+287-46
2012-07-03 20:19Adding TerminationReason in an attempt to have a consistent approach to signaling algorithm success/failure.David Maxwell25+346-252
2012-07-03 00:12some testing codeDavid Maxwell10+587-339
2012-06-29 20:52Major bug fix in InvSSAForwardProblem (adjoint mismatch). First pass at InvSSATikhonovGN.David Maxwell23+920-214
2012-06-28 02:33invert_testi.py works again.David Maxwell3+2-7
2012-06-27 20:48Nearing end of Big Cleanup. Siple-based inversions use new code base. Old siple code based left for now for comparison.David Maxwell22+1965-1819
2012-06-27 06:22Rename of some classes.David Maxwell7+320-319
2012-06-27 05:17Big Cleanup continues. Tikhonov inversions officially moved to common code. Many classes deleted.David Maxwell13+50-1634
2012-06-27 04:55Updated Tikhonov LCL during Big Cleanup.David Maxwell10+517-563
2012-06-27 03:39Big cleanup begun. Removing stale files, consolidating and simplifying code.David Maxwell24+1056-2630
2012-06-26 21:16small fixesDavid Maxwell3+4-2
2012-06-26 19:33Merge branch 'tikhonov' into devDavid Maxwell66+8633-889
2012-06-24 20:52- fixed EISMINT Greenland preprocessing. Please port to stable 0.5Andy Aschwanden1+1-1
2012-06-23 21:29useful refEd Bueler1+12-0
2012-06-23 04:04fix in dev branch to issue #70 (bug)Ed Bueler1+1-1
2012-06-22 20:13and one more ...Ed Bueler1+9-0
2012-06-22 19:52references related to Stokes and Anne LeBrocqEd Bueler1+83-35
2012-06-21 20:49fiddle with capitalization in a refEd Bueler1+2-3
2012-06-09 23:52- a test commit for the summer schoolMcCarthy Summer School1+1-0
2012-06-09 22:11- updated Storglaciaren example to comply with new short names ice_surface_temp and climatic_mass_balance - should be ported to stable 0.5Andy Aschwanden3+10-10
2012-06-09 20:15fixed a typo and some stale text in description of bed roughness; stable0.5 ideally gets thisEd Bueler1+4-4
2012-06-09 19:23another ref on exact stokes solnsEd Bueler1+12-0
2012-06-08 00:40yet another refEd Bueler1+11-0
2012-06-07 22:13ref on marine ice in ice shelvesEd Bueler1+11-0
2012-06-02 00:01Fixes a typo in iceModelVec2T.cc. (This fix should be included in 0.5.)Constantine Khroulev1+1-1
2012-05-25 01:06Added cosine yearly cycle amplitude scaling.Constantine Khroulev3+101-11
2012-05-24 18:43Fixed broken references in PDF docs.Constantine Khroulev2+5-6
2012-05-24 03:01Implemented the 'generic cosine yearly cycle' atmosphere model.Constantine Khroulev5+167-0
2012-05-24 20:53one ref for proposalEd Bueler1+13-0
2012-05-18 20:04General cleanup.(cherry picked from commit 15bd84c9a1f100b43b2f5774f53f514e0ce97946)Constantine Khroulev42+16-4691
2012-05-18 19:32reference fixEd Bueler1+3-2
2012-05-17 22:32More MISMIP.Constantine Khroulev3+8-3
2012-05-17 22:03realized I wanted to fiddle again with subsection heads and text in Getting Started sectionEd Bueler1+21-9
2012-05-17 21:22Updates MISMIP figures in the Manual. Removes unused figures.Constantine Khroulev23+44-19
2012-05-17 19:57another refEd Bueler1+12-0
2012-05-17 19:38making parameter choices match searise-greenland ... it still was not correct :(Ed Bueler2+11-3
2012-05-17 04:15one more referenceEd Bueler1+11-0
2012-05-17 02:50Improves the MISMIP setup. Closes #61.Constantine Khroulev7+241-208
2012-05-16 23:25More MISMIP fiddles.Constantine Khroulev2+31-11
2012-05-16 22:59Closes #39.Constantine Khroulev37+899-88
2012-05-16 22:36Updates the data-flow figure in the Forcing manual.Constantine Khroulev4+70-6
2012-05-16 23:08one refEd Bueler1+9-0
2012-05-16 07:41one more refEd Bueler1+9-0
2012-05-16 06:41references in support of proposal writing; lots of PISM usesEd Bueler1+94-19
2012-05-16 04:42more minor fiddlesEd Bueler2+3-3
2012-05-16 01:25A one-line fix that seems to be needed to compile with NetCDF-3.Constantine Khroulev1+1-1
2012-05-15 23:09Code clean-up; fixing a bug reported by Marijke, MISMIP improvements.Constantine Khroulev11+49-24
2012-05-15 22:131 m/s is indeed a big number to a glaciologistEd Bueler1+5-5
2012-05-15 22:06fiddled with plotEd Bueler1+9-8
2012-05-15 20:19documenting pseudo plastic power law and related optionsEd Bueler1+46-22
2012-05-15 19:41More MISMIP fixes.Constantine Khroulev8+138-134
2012-05-15 17:32Updates the Ross section of the User's Manual.Constantine Khroulev4+42-105
2012-05-15 16:20Minor improvements of the MISMIP setup.Constantine Khroulev3+22-17
2012-05-15 18:40ongoing planning/deciding/moving in examples/Ed Bueler11+107-85
2012-05-15 06:52one refEd Bueler1+12-0
2012-05-14 23:11Replacing the "EISMINT-Ross" section with a "Ross" section.Constantine Khroulev5+162-161
2012-05-14 23:12edits on mismip README.md to clarify/expandEd Bueler1+48-28
2012-05-14 22:02mods to jako/ example doc based on marijke's commentsEd Bueler2+24-20
2012-05-14 21:54Update examples/mismip/README.mdConstantine Khroulev1+9-2
2012-05-14 21:40Finished updating MISMIP; some cleanup.Constantine Khroulev36+169-5115
2012-05-14 21:14-extra_times in script up to date with PISM semanticsEd Bueler1+5-5
2012-05-13 14:14The new MISMIP setup runs.Constantine Khroulev4+278-144
2012-05-13 21:03I think the first section is done, and adequate, for stable0.5 release. Now need g5km_0_ftt.nc from these same runs/settings for use with jako/ etc.Ed Bueler5+13-9
2012-05-13 19:17updating a figure in getting startedEd Bueler4+3-3
2012-05-13 06:18happy with getting started section through figure 4, which I will now replace (though no significant changes)Ed Bueler4+35-41
2012-05-13 04:53updated to ALBMAP/MEASURES result replacing EISMINT-Ross, in manualEd Bueler2+0-0
2012-05-11 22:37Almost ready to remove IceMISMIPModel.Constantine Khroulev3+370-9
2012-05-11 22:17Merge branch 'dev' of github.com:pism/pism into devEd Bueler7+509-46
2012-05-11 22:17ready to redo Figure 4 in manual when runs are doneEd Bueler3+80-55
2012-05-10 23:06Updates test G convergence rate plots in the User's Manual.Constantine Khroulev4+0-0
2012-05-10 20:02Additional references, esp. in support of proposal writing.Ed Bueler1+44-8
2012-05-09 18:44all preprocessing from one scriptEd Bueler1+23-0
2012-05-09 18:45trivial edits to headingsEd Bueler1+3-9
2012-05-09 18:31Closes #47. More work is needed, but *after* the release.Constantine Khroulev9+88-56
2012-05-09 16:21Closes #50 by removing the Cheat Sheet altogether.Constantine Khroulev13+9-431
2012-05-09 03:44Updates the test I convergence rate figure.Constantine Khroulev3+4-6
2012-05-09 00:18One more little tweak to make it easier to reproduce userman figures.Constantine Khroulev1+2-1
2012-05-08 23:37Updates the code producing convergence plots. Removes ssaplay.Constantine Khroulev16+146-1873
2012-05-08 20:54Merge branch 'dev' of github.com:pism/pism into devAndy Aschwanden16+279-241
2012-05-08 20:53- added options, and fiddled with some.Andy Aschwanden2+42-9
2012-05-07 20:59- added missing -pseudo_plastic option, we accidentally used plastic till. Thanks Constantine!Andy Aschwanden1+1-2
2012-05-07 18:57added short list with physical explanations of scalar diagnostics, separating that from usage issues; this picks up content from ancient email xchange w Maria (aug 2011)Ed Bueler1+16-10
2012-05-07 18:00Closes #62.Constantine Khroulev5+21-1
2012-05-07 17:44Addressing #62.Constantine Khroulev2+13-1
2012-05-07 17:13Fixes a typo in examples/jako/README.mdConstantine Khroulev1+1-1
2012-05-07 00:56renamed fields for -surface givenEd Bueler1+2-2
2012-05-07 02:02- added some env variables to make it easier to run script on various machines (PISM_DO, PISM_PREFIX, etc)Andy Aschwanden1+37-2
2012-05-06 19:16DUH. The big effect in incompatibility between whole-sheet and regional runs I saw was refreezing at base; bet this was explained by deleting and starting over on litho temperatures.Ed Bueler4+6-4
2012-05-06 01:02Jakobshavn: scripts in sufficient shape so that redoing whole ice sheet spinup, i.e. section 1 of user manual, is needed before returning and finalizing. the new text in user manual should mostly show off pictures, and leave much of the how-to to the README.md, which does the job fine.Ed Bueler3+99-23
2012-05-05 23:56spinup does not need any preceding SIA in this case I believe; max_diffusivity relaxes to same value if -skip is sufficiently small (e.g. 5)Ed Bueler1+19-29
2012-05-05 23:55rename fields to new nice namesEd Bueler1+5-5
2012-05-05 23:22jako work in progressEd Bueler4+206-95
2012-05-05 22:04ignore some more common files esp in examples subdirsEd Bueler1+12-0
2012-05-05 21:52avoiding specific relative prefix and making it user-adjustableEd Bueler1+5-2
2012-05-05 17:58Minor changes in the Installation Manual.Constantine Khroulev1+16-19
2012-05-04 23:20Working on #46. (Nothing to look at here...)Constantine Khroulev9+167-123
2012-05-04 20:59Minor fixes in the Installation Manual.Constantine Khroulev1+3-3
2012-05-04 20:01Merge branch 'issue-7' into devConstantine Khroulev82+1247-1275
2012-05-04 17:09Added some modified (preliminary) scripts to allow for a prognostic simulation of the Ross ice shelfTorsten Albrecht4+410-0
2012-05-04 06:34read through Installation Manual. Great shape. Made little edits.Ed Bueler1+98-93
2012-05-03 22:32Updates pism-toolkit (more library dependencies).Constantine Khroulev2+5-3
2012-05-03 22:24Closes #51.Constantine Khroulev2+368-183
2012-05-03 18:29added Dani's thesisEd Bueler1+11-0
2012-05-03 17:51Closes #56.Ed Bueler4+18-15
2012-05-03 04:13Addressing #56.Constantine Khroulev4+39-2
2012-05-03 00:22Working on #51 and #58.Constantine Khroulev4+92-161
2012-05-02 17:52looked at manuals' front matter; like itEd Bueler2+33-30
2012-05-02 16:31Closes #48.Constantine Khroulev1+33-0
2012-05-02 16:00Closes #8.Constantine Khroulev22+365-703
2012-05-02 08:08cleaning examples docsEd Bueler3+34-28
2012-05-02 07:53things work well enough to actually do modeling; some things are important: (1) a big band (e.g. 50 cells = 50km) around the ftt_mask drainage basin; (2) small skip value; (3) the first two values given to -topg_to_phi are sensitiveEd Bueler4+49-38
2012-05-02 03:36renamed to 'spinup.sh' and made more realistic hopefullyEd Bueler3+45-39
2012-05-02 00:31removed irritating subheadingsEd Bueler1+15-22
2012-05-02 00:09committing stuff that was already committedEd Bueler2+1-1
2012-05-01 23:00added smb,artm from 5km data set ... runs but needs checkingEd Bueler2+19-10
2012-05-01 22:34cleaned readme and factored scriptsEd Bueler4+183-117
2012-05-01 03:58Merge branch 'dev' of github.com:pism/pism into devEd Bueler18+101-46
2012-05-01 03:58more refsEd Bueler1+44-4
2012-04-29 21:46minor editsEd Bueler1+3-3
2012-04-29 21:29re change: don't know why this was needed, but it fixed it. re result: holy shit that's awesome.Ed Bueler1+2-2
2012-04-29 20:25Adds the plotting script.Constantine Khroulev2+82-1
2012-04-29 07:23Did a petermann example damn near instantly, though I don't have bandwidth to download whole ice sheet spun-up state, so I can't really test.Ed Bueler2+19-4
2012-04-29 05:13editingEd Bueler1+16-15
2012-04-29 03:17awesome. CK: since I take for granted you can build a minimial replacement for the plot scripts, clearly this should replace. note a brief tuning with -ssa_e is something I would be happy to do tooEd Bueler2+28-13
2012-04-28 19:47Closes #54. (Ross example works. Now we need to plot results.)Constantine Khroulev3+104-68
2012-04-27 22:49minor edits to run.sh as a way of keeping notesEd Bueler1+12-4
2012-04-27 21:28Fixes #53. Improves preprocess.py.Constantine Khroulev3+32-38
2012-04-27 20:46Fixes #52.Constantine Khroulev2+7-1
2012-04-27 19:58Works; looks pretty good, too.Constantine Khroulev2+66-49
2012-04-27 18:54new ross example runs for me with -surface given. produces reasonable cbase in output file. most serious issue for diagnositic computation is bad velocity boundary conditions at inflow through transAnt mtns and at siple coast?Ed Bueler2+12-8
2012-04-27 17:22changes to diagnostic behavior that I find helpful in evaluating runs: in -o_size medium, temp_pa more useful than temp, liqfrac needed for thickness/extent of temperate iceEd Bueler3+3-3
2012-04-27 16:34Adds preprocess.py and run.sh scripts for the Ross example. (Both need more testing.)Constantine Khroulev2+220-0
2012-04-27 00:58fixed param nameEd Bueler1+2-2
2012-04-26 21:47Clobbered a warning and removed an inactive line of code.Ed Bueler1+0-2
2012-04-26 19:25spinup script should use thickness from data file for -ocean_kill ... now it doesEd Bueler1+3-3
2012-04-26 04:49Dani's work involved no enhancement, yes? (Cool.) I see the need for Dirichlet b.c.s (i.e. in no_model strip) for enthalpy,SSA. Others?Ed Bueler2+20-6
2012-04-26 01:07Decided on a 3km spinup. Need to think about getting Dirichlet b.c.s from whole Greenland run; how to specifically? Also, what to do for 1km run?: 1 century?Ed Bueler2+32-16
2012-04-26 00:07Closes #43. Fixes #44.Constantine Khroulev8+102-82
2012-04-25 21:17Fixes #45.Constantine Khroulev2+29-3
2012-04-25 19:51Have a run that runs. Need more diagnostics to proceed ;)Ed Bueler2+63-31
2012-04-25 16:56Merge branch 'issue-41' into dev. Closes #41.Constantine Khroulev55+324-196
2012-04-25 08:29cutout stage seems to work; need to go back to preprocess.sh stage to fix surface fields, but ready to try pismo runsEd Bueler2+44-21
2012-04-25 07:32looked at the searise-greenland example, in part. and its doc in User Man. simplified spinup by removing PIK runs; these should appear in Antarctic context first. noted issue #44.Ed Bueler5+47-142
2012-04-25 06:26cleaning up stdout from pism executables (no gna ...) and pismo commentsEd Bueler2+16-16
2012-04-25 06:07Initial stages of a Jakoshavn (actually more generic) regional model using 1km data. Problem at the stage of using cutout command; see end of README.Ed Bueler2+106-0
2012-04-24 22:58documenting my changes in examples/Ed Bueler1+5-3
2012-04-24 22:56moved enth-temp/ to old to get it out of the wayEd Bueler3+0-0
2012-04-24 22:56edits in prep for mvEd Bueler1+11-8
2012-04-24 22:50eisgreen moved to old/Ed Bueler7+0-0
2012-04-24 22:48cleaning eisgreen/ in preparation for moveEd Bueler5+4-226
2012-04-24 22:29Removing 'PST' experiments which are shown in Bueler & Brown (2009). Checkout pism0.3 to reproduce those.Ed Bueler5+0-501
2012-04-24 08:16whitespace onlyEd Bueler1+0-1
2012-04-24 07:36In looking at massContExplicitStep() I desired to edit the comments.Ed Bueler1+7-6
2012-04-23 21:52Merge branch 'issue-22' into dev. Closes #22, closes #34.Constantine Khroulev75+1875-2103
2012-04-20 22:47update 1d ssa solver for bodvardsson example to petsc3.2Ed Bueler2+80-85
2012-04-19 19:50Makes lapse_rate modifiers report *corrected* fields.Constantine Khroulev4+153-4
2012-04-12 23:19Removes all mentions of scikits.delaunay.Constantine Khroulev7+11-37
2012-04-12 20:28a detail: the manual now says the front page is 1km ;-)Ed Bueler1+1-1
2012-04-12 19:47- new front page figure, now from a 1km run (instead of 2km). Doesn't look that different, but the devil's in the detail...Andy Aschwanden2+2-2
2012-04-10 16:21Flish time-series buffers when -extra_files are written.Constantine Khroulev1+3-0
2012-04-09 16:51tim refEd Bueler1+13-0
2012-04-06 18:04thinking about pism0.5 releaseEd Bueler2+48-17
2012-04-06 15:58Fixes PISMNC3File.cc (now PISM compiles on marmaduke again).Constantine Khroulev1+5-3
2012-04-06 07:25SSAFEM reverted to regularizing nuH by adding epsilion_ssa (rather than forming max(nuH,epsilon_ssa)).David Maxwell8+8-11
2012-04-06 03:49Merge branches 'dev' and 'dev' of https://github.com/pism/pism into devDavid Maxwell8+33-24
2012-04-06 02:43NC_BYTE->PISM_BYTE in vel2tauc.pyDavid Maxwell1+1-1
2012-04-06 01:48Quick and dirty commenting out of material in src/CMakeLists regarding pism_netcdf_test that break the build if PNETCDF_LIBRARIES is not defined.David Maxwell1+7-6
2012-04-06 01:32Merge branch 'dev' of https://github.com/pism/pism into devDavid Maxwell90+1493-349
2012-04-06 01:32Minor fixes from earlier commits today. I should learn how to redo my commits before pushing.David Maxwell3+2-108
2012-04-05 20:02Added sia_forward.py, the analogue of ssa_forward.py for doing a one-time solve of the SIA.David Maxwell1+115-0
2012-04-05 19:59Added support to vel2tauc.py for user-supplied python code to do final initialization (e.g. attach custom visualization code) to the inverse solver. Use -inv_prep_module mymodule.py The supplied module should have a function prep_solver(solver,method): where solver is the inverse solver and method is a string (e.g. 'ign', 'nlcg', etc) describing what the core algorithm being used is.David Maxwell1+10-1
2012-04-05 04:53Fixing some bizarre code I wrote in a hurry. (Thanks for reporting the associated compiler warning, Ed.)David Maxwell1+3-2
2012-04-04 05:49Fix of python binding of IceModelVec::regrid. In summary: vec.regrid(filename,True) was calling the regrid(filename,default_value) flavor, not the regrid(filename,critical,start=0) flavor. So there was no real checking for critical variables. This is now fixed, and the preferred calling sequence is vec.regrid(filename,critical=True).David Maxwell1+41-0
2012-04-03 19:55needed include so I can compileEd Bueler1+1-0
2012-04-02 21:55Merge branch 'issue-31' into dev. Closes #31.Constantine Khroulev47+778-409
2012-03-30 06:19updating refsEd Bueler1+25-14
2012-03-30 05:56minor changes to comments and std out messagesEd Bueler1+9-6
2012-03-25 20:10test script generates 20km greenland runs with known ksp badnessEd Bueler1+34-0
2012-03-28 18:50Moves config.get_flag(...) out of the i,j loop in SSAFD. (cherry picked from commit 1b6a71d8d09ca27e127e3687b8e19d715a5cf2c1)Constantine Khroulev2+11-11
2012-03-28 06:04vel2tauc now forces grids to be non-periodic when working in regional mode. This fixes a bug: the default SSA code decides to make non-periodic grids when working in regional mode and the ssa_method is set to 'fem' in the config file. For inversions, the value of ssa_method in the config file is irrelevant; we're always using FEM, so only the regional mode flag determines periodicity.David Maxwell1+10-1
2012-03-23 18:22note duration of preliminary step when reporting it to userEd Bueler1+1-1
2012-03-26 19:12Fixes the bug causing "KSP diverged" errors (the one reported by Andy 22 hours ago).Constantine Khroulev1+3-0
2012-03-25 17:13Adds a regression test that would have caught #30.Constantine Khroulev2+40-0
2012-03-25 16:54Fixes #30.Constantine Khroulev2+15-16
2012-03-24 02:00Closes #20.Constantine Khroulev4+94-1
2012-03-23 21:28Implements preciptation forcing using scalar offsets. Closes #17.Constantine Khroulev4+98-3
2012-03-23 20:58Renames command-line options. Closes #1.Constantine Khroulev3+5-5
2012-03-23 17:26Fixes an unreported bug in pclimate.Constantine Khroulev1+15-6
2012-03-23 08:07Added one more tauc parameterization type ("trunc", a truncated identity that is always non-negative but behaves like "indent" for reasonably large values of tauc).David Maxwell4+66-34
2012-03-23 03:01Fixes #29.Constantine Khroulev1+7-1
2012-03-23 01:54Fixed naming of files from cc0a4d so it compiles for me.Ed Bueler3+1-1
2012-03-23 00:08Implements scalar time-dependent forcing for sub-shelf temperatures and mass fluxes. See #1.Constantine Khroulev6+189-1
2012-03-22 23:37hoping this does not destroy everythingEd Bueler6+48-49
2012-03-22 23:29Merge commit '286d5eb' into devEd Bueler2+105-34
2012-03-22 23:28Merge commit '7ad9f87' into devEd Bueler15+176-217
2012-03-22 23:28Merge commit 'dcd3d3c' into devEd Bueler4+106-67
2012-03-22 23:23See previous message; I forgot to add these.Ed Bueler2+58-98
2012-03-22 23:23Edited scripts to make them less warning-filled (in preprocessing) and to make their user-behavior more like other examples/ stuff. Removed old bueler script; trying to avoid duplications. I am assuming these scripts are not in active use for PIK science, but rather for a (great\!) example of how PIK options work.Ed Bueler1+0-84
2012-03-21 19:31vel2tauc becomes clearer about various tauc's in its output file: 1. We now copy tauc_true to the output file if it exists in the inverse data file. 2. Kludge for now to ensure long names for tauc_true, tauc_prior are accurate in the output file.David Maxwell1+10-0
2012-03-21 04:03Fixed PISMIO->PIO change in vel2tauc.py. make_synth_ssa used to only use SSAFEM, now it doesn't.David Maxwell2+5-3
2012-03-21 03:05Closes #25. (PISM now writes the time since the beginning of the run to a variable in -extra_files.)Constantine Khroulev2+18-7
2012-03-21 02:29Implements the -tauc_to_phi option. Closes #9.Constantine Khroulev14+285-69
2012-03-20 00:12Minor changes in the -tauc_to_phi code. Seems to work.Constantine Khroulev1+5-2
2012-03-13 22:11Merge branch 'dev' into issue-9. See #9.Constantine Khroulev116+3070-1835
2012-01-29 00:35Work in progress.Constantine Khroulev2+2-2
2012-01-28 01:56Finishing touches.Constantine Khroulev3+22-5
2012-01-28 00:48Merge branch 'dev' into issue-9Constantine Khroulev212+11843-8237
2011-11-22 00:38A bit more work...Constantine Khroulev1+50-33
2011-11-21 22:45Renamed PISMDefaultYieldStress -> PISMMohrCoulombYieldStress.Constantine Khroulev14+732-739
2011-11-21 21:58WIP.Constantine Khroulev6+189-123
2011-11-11 18:16Re-arranges some PISMDefaultYieldStress code.Constantine Khroulev3+107-90
2011-11-10 17:54More WIP.Constantine Khroulev2+8-2
2011-11-10 06:56Work in progress.Constantine Khroulev2+57-3
2011-11-10 03:41- added missing (and needed for flux correction) acab_cumulative field to extra vars in SeaRISE Greenland spinupAndy Aschwanden1+2-2
2011-11-09 19:57Closes #12.Constantine Khroulev12+474-114
2011-11-09 02:44Silly whitespace changes to test the "push" notification.Constantine Khroulev3+11-11
2011-11-08 23:15Another try (debian/changelog format is fragile :-/ ).Constantine Khroulev1+2-2
2011-11-08 23:05Fixes btutest and pism-toolkit/debian/changelog.Constantine Khroulev2+4-2
2011-11-08 22:51Fixes the code handling -skip and -skip_max.Constantine Khroulev1+0-12
2011-11-08 21:41Merge branch 'dev' of github.com:pism/pism into devDavid Maxwell1+125-15
2011-11-08 21:39Fix of silly typo.David Maxwell1+1-1
2011-11-08 20:38Merge branch 'dev' of github.com:pism/pism into devDavid Maxwell137+1447-1389
2011-11-08 20:26IceModelVecs can be renamed.David Maxwell10+218-73
2011-11-07 02:37First pass at vel2tauc append mode. Not quite fully baked.David Maxwell3+56-11
2011-11-07 01:47model.ModelData.vars -> model.ModelData.vecs (it turns out that vecs is a builtin function in python; best not to shadow it)David Maxwell12+138-128
2011-11-07 01:08New variable/io model for python side. Common data (physics,variables) are held in a PISM.model.ModelData, which can be shared among other objects. PISM.ssa.SSASolver is gone, effectively replaced by PISM.model.ModelData; leftovers were migrated to PISM.ssa.SSARun.David Maxwell21+522-502
2011-11-06 01:29Working on task 10. Generation of synthetic ssa data removed from vel2tauc and placed in make_synth_ssa.py vel2tauc now reads inversion data from -inv_data fileDavid Maxwell8+239-140
2011-11-05 21:12clobber pedantic warningEd Bueler1+7-4
2011-11-05 19:15Updates config_doc.py and pism_config.cdl. fixes #11.Constantine Khroulev2+26-26
2011-11-05 10:44- added PIK + force-to-thickness spinupAndy Aschwanden1+29-8
2011-11-05 08:34Dorky first pass at algebraic tauc->tillphi via python added for the historical record.David Maxwell1+246-0
2011-11-05 08:01update INSTALL to reflect github hostEd Bueler1+14-22
2011-11-05 07:23Final edit.Ed Bueler1+6-6
2011-11-05 07:19Trying different extension.Ed Bueler1+0-0
2011-11-05 07:18Added a readme using git-flavored markdown.Ed Bueler1+34-0
2011-11-04 20:05Updated the code producing the revision string.Constantine Khroulev1+9-25
2011-11-03 22:34Fixed bug #18838.Constantine Khroulev7+13-0
2011-11-03 21:43Worked on verification docs. FIXMEs remain related to which ssa_testX things to use; we want to keep ssa_testX.py while C++ versions are deprecated.Ed Bueler3+56-74
2011-11-03 21:40Removed README which was out of date and did not say anything helpful. Renamed a directory to make it clear it was documentation not additional sources.Ed Bueler13+0-16
2011-11-03 19:49Added SIAFD to pythonbindings. Early pass at generating SIA velocities so as to generate "observed" SSA velociites. vel2tauc now allows noise to be added to its synthetic observed SSA velocities.David Maxwell5+178-11
2011-11-03 19:10removal of stuff we will never useEd Bueler4+8-1207
2011-11-03 06:55Committing my work on attaching THI to PISM.Constantine Khroulev7+723-891
2011-11-01 18:16Due care added for writing metadata to netcdf files in a parallel run. PISM package imports netCDF3/4 as netCDF minor bug fixes and tweaksDavid Maxwell7+45-50
2011-11-01 01:26Fixed the code handling the "-o_order foo" code.Constantine Khroulev2+20-17
2011-10-31 20:28Made LocalInterpCtx code more robust: it was sensitive to rounding errors in coordinates of domain corners.Constantine Khroulev5+84-38
2011-10-31 19:54Added the code needed to avoid reading tillphi from a file when it is set in "vars".Constantine Khroulev1+13-0
2011-10-31 04:41Added visualization that I found helpful.Ed Bueler1+20-6
2011-10-30 21:52minor tweaks siple log saved to output NC file.David Maxwell2+37-9
2011-10-30 07:15Adding verbose output and visualization to the linear steps of IGN inversions. NetCDF history updated during python SSA saves. Pedantic warning fixed in InvSSAForwardProblem.David Maxwell5+104-20
2011-10-30 06:30This piece of code adds a history string or updates it if already present.Ed Bueler1+15-1
2011-10-29 21:18My compiler does not seem to automatically know that 'sort' comes from this part of STL.Ed Bueler1+1-0
2011-10-29 01:16Moving code checking options and setting config flags and parameters out of IceModel.Constantine Khroulev62+1179-1081
2011-10-28 22:14* InvSSAForwardProblem saves a state for future autopsy when a nonlinear solve fails. * Better visualization in vel2tauc. * SSAFromBootFile can read in a 'tauc' if present. * Added a helper script for comparing the tauc_true and tauc for a synthetic inversion run.David Maxwell4+226-28
2011-10-27 20:06added examples/inverse/vel2ssa.py which is doing (or trying to do) our first steps at inverting for tauc from real-ish data. Right now it reads all SSA data from an input file, generates a corresponding forward solution of the SSA, and attempts to reconstruct tauc from an initial guess of 1/2 of tauc. E.g.:David Maxwell8+576-141
2011-10-27 19:12Fix typo in r2013 changes in directory stressbalance/. Allows build ...Ed Bueler1+1-1
2011-10-27 09:45after reading Bob's email, I realized I have some uncommitted changes: - fixed calculation of dhdt and acab during post processing (a consequence form switching to seconds)Andy Aschwanden2+8-2
2011-10-26 15:34pausing, plotting niceties in invert_testi.pyDavid Maxwell2+15-11
2011-10-26 07:15Found the one-line bug regarding weighted l2 norms in InvSSAForwardProblem. More visualization added to invert_testi. Added option to test adjoint to invert_testi. Adjoints look good.David Maxwell4+142-76
2011-10-25 22:30Re-arranging some code.Constantine Khroulev10+1781-1558
2011-10-25 22:30Minor fixes.Constantine Khroulev2+2-2
2011-10-25 21:25Added optional per-iteration visualization to pismssaforward, invert_testi. Added optional per-iteration pausing to invert_testi. Added support for copying vector-valued vectors to numpy vectors on proc0. Some of these changes require siple commit c4f27bb (Oct 25, 2011) or later.David Maxwell6+269-232
2011-10-25 07:11whoops. forgot to add a file.David Maxwell1+43-0
2011-10-25 05:54ssa_forward now uses PISMYieldStress GeometryCalculator gets a new method for updating a mask and surface vector all at once. Keeps python from doing the loop.David Maxwell4+38-12
2011-10-25 00:10Squashing pedantic warnings. Added one more InvTaucParameterization type for testing purposes.David Maxwell5+40-8
2011-10-24 05:10Added choice of inner products (various L2/H1 combos) for the domain in SSA inversions.David Maxwell2+50-22
2011-10-23 08:04invert_testi now has periodic boundary conditions in the downstream directionDavid Maxwell12+326-224
2011-10-22 19:40Adding support for direct specification of driving stress to the FEM SSA solver. This helps with certain test cases where the grid is periodic but the driving stress cannot be the gradient of a periodic function.David Maxwell5+47-8
2011-10-22 01:38Adding support for tauc parameterizations to python inverse code. This needs some testing.David Maxwell2+50-7
2011-10-21 23:44Moved inversion code out of src/base/stressbalance to newly created src/inverse Created attendant pisminverse dynamic library base/stressbalance/SSAFEM_Forward -> inverse/InvSSAForwardProblem Adding choice of parameterizations of tauc, e.g. tauc = zeta^2 or tauc = exp(zeta).David Maxwell14+1272-1034
2011-10-21 01:13Added weighted L2 range for SSA inversions.David Maxwell4+123-21
2011-10-20 22:39In progress: adding weighted L2 range for SSA inversions. Step 1: don't break things when a weight isn't specified....David Maxwell3+62-10
2011-10-18 22:23Performance improvements.Constantine Khroulev17+112-75
2011-10-17 22:55Two more fixes related to the same issue.Constantine Khroulev2+2-1
2011-10-17 22:47Mac OS X uses a file system that is case-insensitive, so my renaming attempts failed. Argh!Constantine Khroulev1+1-1
2011-10-17 22:16Some work on making the library part of PISM more library-like.Constantine Khroulev123+411-100
2011-10-17 07:17Upgrading to Ubuntu 11.10 meant I got an even more pedantic gcc. Apparently.Ed Bueler4+9-12
2011-10-16 16:39Fixed bug #18813.Constantine Khroulev1+2-0
2011-10-13 22:50Removed the compile-time variable order switch and added the "-o_order" option.Constantine Khroulev8+37-35
2011-10-13 21:39Minor: Added -pismversion line to btutest,pismtests,flowTable extra executables.Ed Bueler4+13-3
2011-10-13 19:45Added CMake variables and an option allowing one to set linker flags manually.Constantine Khroulev1+20-3
2011-10-13 18:39Fixed minor bugs in the re-gridding code.Constantine Khroulev4+38-63
2011-10-13 18:37Fixed a bug introduced in r1281.Constantine Khroulev1+4-3
2011-10-13 07:34Implemented task #7490.Ed Bueler10+65-70
2011-10-13 02:11Made the cold ice conductivity and specific heat capacity fully temperature-dependent if the temperature-dependent forms of EnthalpyConverter and enthSystemCtx are used. (These forms are used by IceModel when runtime options -varc and -vark are set, respectively.) That is, I made IceModel::enthalpyAndDrainageStep() call temperature dependent methods when it needs the physical coefficients k(T) and c(T) for the basal boundary condition. The bulk ice was already dealt with correctly in earlier commits. This should conclude development on task #6778, except for the variable density issue, which is a separate task.Ed Bueler6+76-36
2011-10-13 01:37Updated .dot file for initialization-sequence. Now 'make browser_base' does not generate some mysterious warnings, but still unable to make the image initialization-sequence.png appear in the browser.Ed Bueler1+13-8
2011-10-12 23:26Separated column-wise computations of E_s (the pressure-melting temperature enthalpy value) and the 'lambda' for BOMBPROOF. Documented some code related to variable (temp.-dependent) properties k(T) and c(T). No functional change.Ed Bueler6+76-43
2011-10-12 18:51Fixed a bug in scalar forcing introduced in r1958.Constantine Khroulev4+12-4
2011-10-12 05:36Removed a chunk of code from pismo.Constantine Khroulev1+0-26
2011-10-11 21:16Fixed a bug: IceModel::ageStep() was called every mass-continuity time-step but was using dt_TempAge.Constantine Khroulev1+1-1
2011-10-11 19:51Cleaning up pismo code.Constantine Khroulev12+92-120
2011-10-11 17:26Modified CMakeLists.txt to warn the user ifConstantine Khroulev1+15-1
2011-10-10 19:45Fixed a stupid off-by-one bug in the new regridding code I committed last week.Constantine Khroulev1+4-4
2011-10-10 17:18Fixed several bugs I introduced while switching from "years" to "seconds".Constantine Khroulev13+71-35
2011-10-09 04:51Minor stuff related to maintaining online browsers.Ed Bueler2+25-18
2011-10-08 22:13Addressed bug #18748: time steps of bwat diffusion process should only occur at the same time as (specifically, just after) energy steps. Also, made updates to yield stress occur only when updateAtDepth is true; this is morally right (because only the stress balance solve depends on tauc values) and it should be very slightly faster. Note that the frequency of calling the basal hydrology model (e.g. diffuse_bwat() is an extremely simple one) should be determined by the needs of the hydrology model. But tauc should be updated from subglacial state variables (e.g. bwat) only when tauc is actually needed.Ed Bueler2+12-19
2011-10-08 21:36Updated python-based SSA regression tests to new CustomGlenIce interface and to eliminate use of 'secpera'.David Maxwell5+30-18
2011-10-08 20:40Editorial changes on hydrology page. Added Schoof references. (By the way, has anyone reproduced bug #18785? I can't currently reproduce it. I will close that bug soon unless someone comments on it.)Ed Bueler2+53-28
2011-10-08 06:42A commit that does not touch anything in src/ (just to test if we're done with the task #7489).Constantine Khroulev1+3-3
2011-10-08 02:23Addressed task #7489.Constantine Khroulev2+14-7
2011-10-08 00:16Cleaning up enthalpy-related code.Constantine Khroulev35+422-394
2011-10-07 04:26Fixed a bug reported by Andy.Constantine Khroulev5+11-7
2011-10-07 00:27- updated efgrun.sh to reflect removal of varkpismrAndy Aschwanden1+5-5
2011-10-06 22:47Fixed a bug in the time-series code.Constantine Khroulev2+3-1
2011-10-06 21:33Corrected constants so we are making no unnecessary error in implementing equation (4.39) in Greve and Blatter.Ed Bueler2+6-6
2011-10-06 21:23Objects of type IceFlowLaw should not know about thermal conductivity, specific heat capacity, or the latent heat of fusion. Now they do not. As a result, the corresponding configuration constants from pism_config.nc will more directly and more transparently configure various code that uses ice flow laws; before this was less direct and went via the IceFlowLaw constructor. (By the way, this commit and r1962, r1961, r1955, r1954, r1952, r1951, r1945 are all implementing an old task. Namely, task #6778.)Ed Bueler10+73-54
2011-10-06 01:59Removed pgrn and all the mentions of it.Constantine Khroulev20+77-334
2011-10-05 21:36Removed varkpismr; added the -vark option instead.Constantine Khroulev12+157-687
2011-10-04 22:01Added a page on adding hydrology to PISM, to the related pages page. This evolved from an old email to Ward. We aren't helping him much, but at least I should reply to his email now ...Ed Bueler4+101-6
2011-10-04 19:17Nothing important: clobbered pedantic warnings.Ed Bueler2+5-2
2011-10-04 17:24Major changes related to task #7450.Constantine Khroulev66+522-514
2011-10-04 16:55Re-wrote some of the re-gridding code (see task #7487).Constantine Khroulev6+159-181
2011-10-02 23:14- added varc, vark and varck runs to efgis test suiteAndy Aschwanden2+46-10
2011-10-02 20:46O.k. That was easy. (Constantine's design really is extensible.) Now PISM has option -varc which allows the enthalpy method to use the linear-in-temperature specific heat capacity function C(T) defined by equation (4.39) in Greve & Blatter (2009). Formulas need to be checked before the results should be trusted; of course it makes some difference and we need to know if that difference is 'real'. See examples/enth-temp/comparevariabletemp.sh for how to use new option -varc and new executable varkpismr. FIXME: the latter might as well become pismr -vark.Ed Bueler7+61-18
2011-10-02 19:19Code for linear-in-temperature specific heat. Compiles but not tested yet. (I need to recall how an alternate EnthalpyConverter could be attached to IceModel.) Andy and Constantine: please 'make browser_base' and check my formulas for enthalpy-temperature conversions, based on (4.39) in Greve and Blatter (2009).Ed Bueler4+81-12
2011-10-02 19:15Merely updating PISM-PIK references in docs; no functional change.Ed Bueler3+5-5
2011-10-02 08:44The new implementation of varcEnthalpyConverter will be pretty easy. I had earlier committed these codes with no difference from base EnthalpyConverter; at least now I have identified the parts that actually differ, and those are slight.Ed Bueler2+24-372
2011-10-02 04:49Removed FIXMEs from r1945 code. Now there is a more complete, less inefficient implementation of variable conductivity, including use of the right conductivity in the Neumann b.c. But it still seems to make very, very little difference. And the code duplication remains. (Has slightly increased, actually.) Now I am going to look at implementing a variant EnthalpyConverter that uses linear-in-temperature specific heat capacity.Ed Bueler2+91-20
2011-09-30 01:26Dealt with pedantic compiler warnings from two recent commits of pism_const.cc and SSAFD.cc. Minor formatting, use of 'const' in SSAFD.cc.Ed Bueler2+7-8
2011-09-29 22:03Modified PISMOptionsInt and PISMOptionsReal: now they print more helpful error messages when an invalid option argument is given.Constantine Khroulev1+32-8
2011-09-29 18:11Added installation text because: We forgot to tell users everything is done and they can go use PISM.Ed Bueler1+22-3
2011-09-29 12:38Correcting the Calving Front Boundary Condition (CFBC) for varying sea level.Matthias Mengel1+4-5
2011-09-29 06:51Implemented temperature-dependent cold-ice conductivity k(T) in enthalpy scheme. Implementation is by code duplication of enthSystemCtx and IceModel::enthalpyAndDrainageStep(), and new executable varkpismr.Ed Bueler8+1164-4
2011-09-28 21:25Re-organized src/coupler.Constantine Khroulev33+51-43
2011-09-28 19:58Working on task #7450, which turns out to be more important than I thought.Constantine Khroulev63+599-536
2011-09-28 19:33DrainageCalculator needs to be visible outside of iMethalpy.cc for extensibility. enthSystemCtx does not need EnthalpyConverter.Ed Bueler4+60-36
2011-09-28 17:15Some changes related to Dirichlet SSA B.C.Constantine Khroulev2+5-10
2011-09-27 22:52- fixed SeaRISE-Greenland experiment script: turned bed deformation model on. - removed PIK demo, not needed anymoreAndy Aschwanden2+3-35
2011-09-22 22:13Revising the previous commit... (Clearly I need a cup of coffee.)Constantine Khroulev1+8-9
2011-09-22 21:58Avoid reading in no_model_mask if -no_model_strip is set.Constantine Khroulev1+10-0
2011-09-22 20:22Renamed {u,v}bar_ssa to {u,v}_ssa. Renamed {u,v}bar to {u,v}_ssa_bc in SSA.cc and iceModel.ccConstantine Khroulev3+16-32
2011-09-22 19:36Fixed refs to PISM-PIK papers.Ed Bueler3+15-19
2011-09-22 18:00Fixing a result of some late-night coding.Constantine Khroulev1+1-1
2011-09-22 17:49Added the -ssa_dirichlet_bc option to pismo.Constantine Khroulev4+128-72
2011-09-20 22:50One more fix. I should slow down.Constantine Khroulev1+1-1
2011-09-20 22:48Fixed a stupid bug in my previous commit.Constantine Khroulev1+16-0
2011-09-20 22:32Added pismo code to keep the energy state in the no_model_strip from evolving.Constantine Khroulev1+85-14
2011-09-20 01:39Fix to bug reported by Andy, for SeaRISE-Greenland. Runs were using (not totally unreasonable) linear till. Issue does not afflict Storglaciaren nor SeaRISE-Antarctica; fixed in former and inactive in latter. See task #7484.Ed Bueler1+3-0
2011-09-20 01:05A commit I forgot to make earlier, but no big deal: Slightly better parameters for modeling using fine (esp. vertical) grids on Storglaciaren. And removal of code to deal with bug #18593, which is fixed.Ed Bueler1+11-7
2011-09-16 19:00Fixed a bug in the "daily" time-series and extra file reporting.Constantine Khroulev1+8-8
2011-09-15 18:03* One fix necessary to make PISM work with proj.4 version 4.5 * One minor fix: -{surface,atmosphere} given now save mass fluxes in m/year and not m/s.Constantine Khroulev2+15-2
2011-09-15 16:49More work on the task #7482. Now it compiles (r1924 didn't; sorry about that).Constantine Khroulev4+106-81
2011-09-14 22:41Working on task #7482. (I'm about half-way done and will be done soon.)Constantine Khroulev4+119-0
2011-09-14 18:45Added surface and ocean update() calls to IceModel::model_state_setup() to ensure that a BTU get the correct top-of-the-bedrock B.C. at the beginning of the run.Constantine Khroulev1+2-0
2011-09-14 17:05Added a check in the code so that 'pismv -test K -Mbz 1' does not crash.Constantine Khroulev1+12-1
2011-09-13 19:20Fixed 'pismv -test K -eo' (was not respecting PISMBedrockThermalUnit's time-step restrictions.)Constantine Khroulev1+1-1
2011-09-13 17:00Removed the code converting geodetic coordinates to geocentric ones.Constantine Khroulev3+23-69
2011-09-08 16:37Fixed bug #18149 in the trunk.Constantine Khroulev6+224-156
2011-09-07 19:27Fixed bug #18312 in the trunk.Constantine Khroulev3+24-2
2011-09-07 18:48- added EISMINT-Greenland based example which compares enthalpy and temperature-based thermodynamics. These are the runs used in the paper "An enthalpy formulation for glaciers and ice sheet".Andy Aschwanden2+153-0
2011-09-06 23:06Added the Pism_MANUAL_BUILD CMake flag that disables *all* the scripts attempting to find essential libraries.Constantine Khroulev1+27-8
2011-09-06 19:21Added one line to PISM.i so that IceModelVec2S::sum() is wrapped properly.Constantine Khroulev1+1-0
2011-09-02 23:07Start of adding siple based inversion files to PISM.David Maxwell4+524-0
2011-09-02 23:02Mostly small documentation changes. hmelt_max -> bwat_max in util.pyDavid Maxwell6+40-17
2011-09-02 20:13Fixed bug #18313 by adding more command-line options.Constantine Khroulev5+42-40
2011-09-02 16:33Implemented reporting of discharge_flux, containing calving loss from -calving_at_thickness, -eigenCalving, -kill_Icebergs. Only gets reported with -ts_vars discharge_flux. At a first glance it looks ok, but needs further testing.Maria Martin6+109-1
2011-09-02 15:34Fixed the bug I introduced in r1900.Constantine Khroulev2+4-0
2011-09-02 04:58Fixed use of set_attrs() for dHdt diagnostic field.Ed Bueler1+1-1
2011-09-02 00:36- added maximum_diffusivity plot for force-to-thickness to browser, and updated scripts accordingly. - still not restartable, though.Andy Aschwanden5+76-32
2011-09-01 23:19Replaced foo.data() with &foo[0] throughout.Constantine Khroulev14+47-47
2011-09-01 23:17Fixed bug #18593 in the trunk.Constantine Khroulev1+0-2
2011-09-01 20:27Fixed up Storglaciaren 3d model so (e.g.) Marjike and David can use an example that is a whole ice mass, that is small scale, and that has realistic geometry. To use 3d model, see examples/storglaciaren/README. Several improvements to the model, already used in the flowline version, could be applied easily in this 3d case. Note 40m grid in use but that could be 10m or 20m with modest runs. Note sure about availability of observed surface velocities, but this is likely.Ed Bueler6+136-21
2011-09-01 19:48 - we now consider \lambda in force-to-thickness mechanism to be constant, thus it was removed as it would act only as a scalar multiplier of \alpha. Change default \alpha, and updated documentation. Made appropriate & necessary changes to EISMINT Greenland example. - replaced {*,/} secpera with convert(x,"inunit","outunit") in PISMSurface.cc. Changed units of \alpha in pism_const.cdl to yr-1, conversion to s-1 occurs in PISMSurface.hh - updated ice_bib.bib - todo: 1. report time scale 2. add software test for exact restartability with FTT. The test in test_ftt.sh indicates that exact restartability does not yet happen. Needs further testing.Andy Aschwanden9+196-87
2011-08-31 19:19Made high tauc configurable.Constantine Khroulev2+8-5
2011-08-31 17:53 - removed thick0, temp0 calculation from iMreport.cc - fixed a typo - updated User's Manual - added maximum diffusivity to stdout - removed thick0, temp0Andy Aschwanden5+26-41
2011-08-30 22:06Changed the way surface gradient is computed in pismo (both in SIAFD_Regional and SSAFD_Regional).Constantine Khroulev3+155-47
2011-08-30 13:29New timeseries diagnostic variable slice, which is sea-level relevant ice in sea-level equivalent. Not part of default output, needs to be asked for in runtime option -ts_vars slvolMaria Martin4+73-0
2011-08-29 21:26Fixed bug #18485 in the trunk.Constantine Khroulev3+20-18
2011-08-29 18:42Created IceModel::cell_interface_diffusive_flux().Constantine Khroulev5+51-35
2011-08-25 10:31A couple more refs.Ed Bueler1+20-0
2011-08-25 01:46Bed elevations are never actually accessed in IceModel::massContExplicitStep(), whether dirichlet boundary conditions or not.Ed Bueler1+0-2
2011-08-24 23:47Fixed the pismo re-implementation of IceModel::cell_interface_velocities().Constantine Khroulev1+21-19
2011-08-24 20:49Merge branch 'cleanup'Constantine Khroulev1+2-1
2011-08-24 00:24- implemented regional versions of diffusive flux cell interface velocities computation in SIAFD_regional::compute_diffusive_flux() and IceRegionalModel::cell_interface_velocities, respectively. - there remains an issue with differencing across processor boundaries at the boundary between model and no model area, probably in compute_surface_gradient or compute_driving_stress. More specificially, such a corner point tau_y has opposite sign compared to its neighbors.Andy Aschwanden2+52-2
2011-08-23 22:09* Cleaned up regression test numbers. * Fixed a documenting comment in iMgeometry.cc. * Removed ftt_modified_acab. See task #7435. * Re-implemented the dHdt diagnostic: now -extra_times 0:10:1000 -extra_vars dHdt -extra_file foo.nc will save dHdt computed as the *average* ice thickness rate of change over 10-year periods. The first (year 0) frame is filled with _FillValue. * Got rid of a compiler warning in src/base/stressbalance/THI.cConstantine Khroulev28+474-414
2011-08-21 22:41- added cumulative_* quantities to table of available scalar diagnostics in User's ManualAndy Aschwanden1+7-1
2011-08-19 20:40- minor changes to slr_show.py: - choose start and end year of plot with -a A and -e B, respectively, where A and B are integers in years from 2004 - first makes a high-contrast plot for the screen, then saves the figure with a white background insteadAndy Aschwanden1+50-6
2011-08-14 23:30* Replaced "taud" with "taud_mag" in output_medium config variable. But see bug #18506 pointing out that more needs to be done.Ed Bueler3+146-32
2011-08-01 21:22Completion of task #7428, and fix to associated '-o_size small' bug.Ed Bueler4+48-22
2011-08-01 15:15fixed rhs results in ssaBC documentation according to code and winkelmann et al 2010, eq. 2.3.6-7Torsten Albrecht1+4-4
2011-07-25 20:49- added ncpdq line to postprocess scripts to change dimension ordering to (time,y,x) for SeaRISE-compliance.Andy Aschwanden2+2-0
2011-07-24 06:34- changed "-calving_at_thickness" from 50 to 100m in searise-greenland spinup. - added option to do experiments with PIK marine ice dynamics: do $ ./experiments.sh N foo.nc 1 to turn it onAndy Aschwanden2+10-2
2011-07-22 00:19- added 5ka spinup with PIK options to searise-greenland. Needs testing.Andy Aschwanden1+27-4
2011-07-20 19:56- fixed two typos in error messagesAndy Aschwanden2+2-2
2011-07-20 14:49fixed a typo in experiments.shMaria Martin1+2-2
2011-07-20 14:41changes to searise-antarctica scripts: They should be ready to use now. Configuration parameters are set, and postprocessing is added, basically identical with the respective scripts from searise-greenlandMaria Martin5+279-71
2011-07-20 02:52- smalish increase in ftt alpha to better match observed surface elevation - added a script to visualize SeaRISE experimentsAndy Aschwanden2+68-1
2011-07-16 20:57A fix for the bug #18173.Constantine Khroulev4+64-32
2011-07-16 09:04converting the temperature anomaly data in future forcing for SeaRISE Antarctica from Celsius to Kelvin, such that the can be direktly used as input for PISMMaria Martin1+4-0
2011-07-15 18:38This commit adds a line or two to a readme. But it is an excuse to say: the files ar4_[precip|temp]_anomaly.nc at http://www.pism-docs.org/download are have the corrected ice-equivalent units. See previous email on this topic; I had failed to upload the corrected ones.Ed Bueler1+6-0
2011-07-15 00:32Fix to bug #18332 in trunk: vLongitude,vLatitude have a 'missing_at_bootstrap' attribute which is checked by compute_cell_areas(). PASeariseGreenland fails if they are missing_at_bootstrap. Also: updated URL for SeaRISE-Antarctica to valid.Ed Bueler4+30-1
2011-07-14 22:52- searise-greenland postprocessing second amendment: added grid_mapping attributes to acab/dHdt. No smoking pistols...Andy Aschwanden1+2-0
2011-07-14 01:15Fixes to *both* SeaRISE-Greenland and SeaRISE-Antarctica scripts to fix the units of precip to be ice-equivalent as needed by PISMs couplers. In the former: flipped ratio to give correct multiplier. In the latter: added lwe->ice-equivalent conversion. See email.Ed Bueler4+29-11
2011-07-14 00:33- searise-greenland postprocessing first amendment: added meta data to dHdt and acab, removed unneeded fields from final files.Andy Aschwanden1+13-5
2011-07-13 20:55- fixed a typo in searise-greenland preprocess script - yearly-averaged dHdt and acab are calculated during postprocessing using some ncap2 magic. Yearly output is kept in *_full.nc files, while submission-ready files only contain data every 5th yearAndy Aschwanden2+8-4
2011-07-13 17:36Added python based ssa regression tests.David Maxwell11+90-81
2011-07-13 15:09small changes in SeaRISE-antarctica scripts --- they du run now, but more testing and a final decision for the pik-parameters to be used for SeaRISE is neededMaria Martin3+47-77
2011-07-13 04:09- added new variable acab_cumulative to extra vars. postprocess script will need updating.Andy Aschwanden1+2-2
2011-07-13 04:04Tidying up scripts. No functional change.Ed Bueler2+15-15
2011-07-13 02:17Implemented reporting the cumulative surface mass flux.Constantine Khroulev8+78-3
2011-07-13 00:53- nc2cdo.py now always adds standard name to lat/lon fields - minor changes to experiment and preprocess scriptsAndy Aschwanden3+27-31
2011-07-12 22:36Fixed Python regression tests so that they fail if a script exits with the return code of zero even though it broke and Python printed a traceback.Constantine Khroulev1+5-1
2011-07-12 22:36Implemented '-atmosphere pik' (see task #7424)..Constantine Khroulev5+135-19
2011-07-12 21:59Fixed broken call to PISMIO.append_time.David Maxwell1+1-1
2011-07-12 18:28Minor updates to references.Ed Bueler1+10-1
2011-07-12 15:12added preliminary shell-scripts for an Antarctica-Spinup with constant climate (antspinCC.sh, later to be followed by antspinGC.sh for a glacial cycle spinup) and experiments.sh with the SeaRISE experiments, mostly based on Andys scripts for SeaRISE-Greenland. Rudimentary Documentation in README and in the shellscripts themselves.Maria Martin4+562-2
2011-07-12 02:29- added tauc, tempicethk_basal to extra_varsAndy Aschwanden2+2-2
2011-07-12 02:22- minor changes to searise-greenland, mostly to keep up with recent changes in PISM codeAndy Aschwanden3+13-13
2011-07-08 23:37- fixed a typo in manual. - renamed "-ocean_bc_file" to "-ocean_file", as a follow-up to commit 1848. Direct forcing options are now all consistently called "-{surface,atmosphere,ocean}_file"Andy Aschwanden2+2-2
2011-07-08 22:12Another fix for bug #18307.Constantine Khroulev2+12-18
2011-07-08 16:28introduced alternative preprocessing for future climate in SeaRISE Antarctica as suggested by Ed... it needs to be tested, however, whether anomalie files can be read in with option -surface pik,anomaly or something like this...Maria Martin1+29-2
2011-07-07 19:53This is the 'audit trail' for our ar4_..._anomaly.nc files. Note they are generated from the global form of the AR4 data at the SeaRISE site.Ed Bueler5+345-1
2011-07-07 16:09updated preprocess.sh for SeaRISE-antarctica to produce the scaled future climate data needed for Experiments C*Maria Martin1+15-0
2011-07-07 15:46new option -sliding_scale_brutal which simply scales the ssa-velocity (also on the shelves). This aims at providing an alternative to the much better justified option -sliding_scale, the only reason one might do this is to meet the SeaRISE requirements for experiment S more literallyMaria Martin3+20-1
2011-07-07 15:22changed the way the numerical input with option -essa is interpreted... now it is subject to the power of 1/n as in equation 9 of winkelmann_martin11. This makes the numerical input comparable to the numerical values given in Ma_Gagliardini10Maria Martin1+2-2
2011-07-07 15:08changed the default parameter for F_melt as in martin_winkelmann11 and introduced an option -meltfactor_pik to be used together with -ocean pik whenever a numerical value different to the default of 5e-3 should be usedMaria Martin1+10-1
2011-07-07 14:52changed the parameterization for surface temperature to match the one in the performance paper for PISM-PIK (martin_winkelmann11)Maria Martin1+1-1
2011-07-06 21:12Redo on code to fix bug #18307. And removed -o_size big from antarctica test script; the files are big enough already.Ed Bueler2+16-7
2011-07-05 00:27- a simple hack to deal with bug #18307. Not the desired solution, will need proper fixing.Andy Aschwanden1+3-1
2011-07-01 23:43Implemented "-ocean given" and renamed the -bc_file option: now we have "-foo_bc_file" for use with "-foo given". (I.e. "-surface given -surface_bc_file data.nc", etc.)Constantine Khroulev6+168-31
2011-07-01 08:41Added standard controls to searise-antarctica script, but no change in content except addition of 10ka (= no much) -no_mass before running. Clobbered pedantics in 2T.Ed Bueler3+45-19
2011-07-01 00:56Added a workaround for a bug in NetCDF. (Please see https://www.unidata.ucar.edu/jira/browse/NCF-88 for details.)Constantine Khroulev1+9-1
2011-07-01 00:32Implemented 'daily', 'monthly' and 'yearly' reporting that uses the Gregorian calendar to compute reporting times.Constantine Khroulev7+155-21
2011-06-30 19:27Fixed CMakeLists.txt (FFTW_INCLUDES was missing from the list of include directories).Constantine Khroulev1+1-1
2011-06-29 00:59Renamed "t" to "time" in output files.Constantine Khroulev23+112-179
2011-06-28 21:43Modified the code reading (and using) scalar and spatially-variable forcing data to use time bounds (if present).Constantine Khroulev8+239-148
2011-06-27 18:30Added a README and trivial simplifications to pgrn.cc as I thought about killing pgrn.Ed Bueler4+42-28
2011-06-26 21:55- refactored and isolated Eismint- and SearRISE-Greenland atmospheres. Now you can call "-atmosphere eismint_greenland", similar to "-atmosphere searise_greenland", without using pgrn.Andy Aschwanden11+322-276
2011-06-26 21:04* Changes to src/ are all in strings: stdout cleanup, esp. in case of ksp-diverged, and long_name for some scalar time series to clarify the areas are basal ice.Ed Bueler6+140-10
2011-06-25 01:12Cleaned up pism_[python,matlab].[py,m]. These now produce the same file according to nccmp.py, and run is more realistic. (A version of these runs can be show/test mass conservation on nonflat bed; see comments in pism_python.py.) Added groundingline.m from no-longer-public part of karthaus notes; needs work.Ed Bueler3+162-54
2011-06-24 23:09* Made pik iceberg process code less verbose. Now either -vpik or '-verbose 3' will give all those cool messages about 'cut off nose' or 'killed isolated one-box-iceberg' or 'lonely partially filled grid cell'. But they are off by default. Uses new config flag 'verbose_pik_messages'.Ed Bueler7+137-149
2011-06-24 22:19Shortened finest grid part of spinup to 5ka. Revised pikdemo.sh to be a better, simpler demo of pik options for Greenland experimentation.Ed Bueler2+25-28
2011-06-24 02:04Most changes relate to this page, which I have also updated: http://websrv.cs.umt.edu/isis/index.php/Output_Format * Changed output units (glaciological units) on iarea,iareag,iareaf,iareatemp,iareacold to be m2 instead of km2. * Scripts: - add dHdt (SeaRISE-required and no longer in bad set) - remove title at run stage; added in post-pro - mask values match SeaRISE (after I resolved inconsistency in above page) - removed bounds,coordinates,pism_intent attributes from all varsEd Bueler4+40-29
2011-06-24 00:17Fixed bug #18268.Constantine Khroulev1+134-52
2011-06-23 01:35These scripts do seem to work. They are more aligned to current PISM functionality and to output format SeaRISE webpage.Ed Bueler3+45-34
2011-06-23 01:06Replaced the IceFlowLaw::averagedHardness_from_enth() implementation with the one I can understand. (Works better, too.)Constantine Khroulev2+36-22
2011-06-22 20:53- fixed an unreported bug in calculating -ocean_kill flux - added cumulative fluxes such as cumulative_surface_mass_flux, etc. On by default.Andy Aschwanden4+211-1
2011-06-22 02:11Added the nuH diagnostic quantity. (Available with -ssa_sliding and -ssa_method fd.)Constantine Khroulev3+48-0
2011-06-22 00:48Added the "timeseries_buffer_size" configuraton parameter (default is 10000).Constantine Khroulev3+4-10
2011-06-21 23:20Re-factored scalar time-series.Constantine Khroulev17+1563-895
2011-06-20 22:59Fixed an minor bug in PISMBedSmoother::get_theta().Constantine Khroulev1+6-4
2011-06-19 05:02Added a ref and changed a phrase.Ed Bueler2+10-2
2011-06-17 23:56clarification of meaning of extra_vars velsurf,velbar,velbaseEd Bueler1+4-5
2011-06-17 23:42Minor fixes. (The postprocess.sh script now works for me, now that I have reinstalled netcdf4-python which generated the earlier error.)Ed Bueler2+3-2
2011-06-17 01:59- single-line bug fix in nc2cdo.py. Should be ported to 0.4.Andy Aschwanden1+1-1
2011-06-16 00:51Task #7414 done in the way I think it should be done for now: Diffusion of stored basal water, which was deleted in r1642, is back. It is off by default but can be turned on with -diffuse_bwat. No derived class of PISMYieldStress was built, but instead this functionality was added back to IceModel. (Logic: I decided that there should be, in the future when we know what it is supposed to do, a well-defined object for hydrology. Until then PISMYieldStress should be uncluttered. This suggests PISMYieldStress::basal_water_pressure() is mis-placed, and should be part of future hydrology object. Perhaps it should be in file src/base/iMhydrology.cc.)Ed Bueler8+119-4
2011-06-15 08:49Added ability to run quickstart with 'export METHOD=fem' instead of default METHOD=fdEd Bueler2+19-3
2011-06-15 08:31Implemented task #7413.Ed Bueler13+112-113
2011-06-15 08:12Removed purely historical setDefaults() method of IceModel. It remains the case that a few things are set in the constructor of IceModel; whether that should change is worth brief thought perhaps.Ed Bueler4+16-78
2011-06-15 07:02It turns out that a boolean which is always 'true' is not really needed as a member of IceModel. (Blame ... historical reasons.)Ed Bueler7+10-23
2011-06-15 06:49Clobbered pedantic warnings.Ed Bueler2+8-10
2011-06-15 01:41- updated postprocess_mask.py to conform with new PISM mask values. Should fix bug reported by Ed per email.Andy Aschwanden1+10-7
2011-06-15 01:12Trivial edit to .bib. For experiments.sh, added ability to use env vars PISM_SKIP and PISM_INITIALS.Ed Bueler2+30-11
2011-06-14 20:13- new scalar time series "gDmax" (maximum diffusivity), added to list of default quantities, and made a note in the User's Manual. Ideas for a better name?Andy Aschwanden4+24-2
2011-06-14 05:12- oops, needed to fix another bug in the script (my fault)Andy Aschwanden1+7-7
2011-06-13 16:52- renamed "forecast" -> "experiments" - todo: updated postpro the maskAndy Aschwanden4+253-253
2011-06-11 00:09OOPs. The last commit message referred only to the changes to examples/searise-greenland/preprocess.sh Now for the real commit message for r1792 src/:Ed Bueler1+1-1
2011-06-10 23:39Very, very minor cleanup of stdout.Ed Bueler3+102-72
2011-06-10 01:34- added FTT run to spinup - added previously missing postprocess_mask.py file - some minor modifications to SeaRISE scriptsAndy Aschwanden4+102-2
2011-06-09 22:56- fixed an unreported bug in Ocean model: glaciological unit was set for shelfbmassflux, but not converted at the writing stage - added new command-line option -shelf_base_melt_rate (m/yr) to the constant ocean model, added a note in the user's manual - added experiment M (shelf melt rate) to SeaRISE forecastAndy Aschwanden5+95-33
2011-06-09 07:08No change in functionality from these changes. But see the diff for pism_config.cdl; I think bug #18193 may be fixed simply by adding -ssa_eps 1e17 (or -ssa_eps 1e16) instead of the default -ssa_eps 1e15; I'll try it. If this fixes the bug then we will want to make the mechanism more robust and easier to diagnose. This issue ... takes me back to the good old days (=about 2007). Other changes: move ssaplay dir to right place, added minor ref, doc cleanup.Ed Bueler12+11-6
2011-06-09 02:01- fixed a typo in preprocess script - experiment S and C mostly ready, now working on M - added postprocess script, unaltered from ice repo - more coming soon.Andy Aschwanden3+124-49
2011-06-08 20:46* Renamed grid.{cc,hh} to IceGrid.{cc,hh}. * Added processor rank and (map-plane) processor grid size to -prof outputs. Added code to ignore events that took less than 0.001 seconds (max. over all processors).Constantine Khroulev29+826-801
2011-06-08 01:42Added basic regression tests for PISM's Python bindings.Constantine Khroulev3+63-1
2011-06-08 00:20- Some more updating to searise example: - uses config file for physical parameters - preprocessing creates scaled anomaly files - Needs testing. Not yet dealt with forecast scriptAndy Aschwanden4+201-95
2011-06-08 00:17Jed's comment on bug #18193 suggests this, to write the linear system to a petsc binary file. (Next: Will seek to reproduce error and provide him with the file.)Ed Bueler1+11-1
2011-06-07 23:51Update to SeaRISE-Greenland spinup script. Notable: equal-spaced vertical, no PDD params from Fausto, added tempicethk_basal and bwp to extras.Ed Bueler1+13-13
2011-06-07 00:35Added a footnote about ccmake on Ubuntu.Constantine Khroulev2+2-2
2011-06-06 18:33Updated site-packages/PISM/util.pyConstantine Khroulev1+2-2
2011-06-06 09:09Added PetscTruth puOnTop to avoid a couple of repetition for the redistribution of only small ice volumes (minHRedist < 50m in total?).Torsten Albrecht2+14-11
2011-06-06 08:49neglect implicit SMB in partially filled cells to hopefully avoid the occurrence of bug #18146Torsten Albrecht1+3-1
2011-06-04 01:47- Updated 'Getting Started' example figures and text to 10km version.Andy Aschwanden6+2-2
2011-06-04 00:22Clobbering pedantic compiler warnings from recent bugfix commit.Ed Bueler2+3-12
2011-06-03 23:13* PDF docs are not included in the .deb package any more (before this change it was possible to get a package with or without PDFs depending on the system configuration; not good). * Moved the check_maximum_thickness() call into the very end of IceModel::massContExplicitStep().Constantine Khroulev2+10-7
2011-06-03 22:49- changed (previously forgotten) -no_temp -> -no_energy in cheatsheetAndy Aschwanden1+1-1
2011-06-03 22:49Correction to text in Getting Started, in prev commit.Ed Bueler1+1-1
2011-06-03 22:46Adding 10km results to 'Getting Started' in User Man.Ed Bueler4+17-13
2011-06-03 22:44Minor clean-up of pismo.cc, with added docs.Ed Bueler1+27-11
2011-06-03 01:14Fixed bug #18176 by re-organizing the code allocating sub-models.Constantine Khroulev13+345-260
2011-06-02 05:55Adding references as I work on replying to Ward's excellent analysis.Ed Bueler1+36-1
2011-06-01 22:52In search of bug #18176, cleaned up stdout. Unable to reproduce bug on my laptop for now.Ed Bueler5+33-23
2011-06-01 07:18Minimal factoring of ex48.c solver. THI is a well-defined class already. It can be re-implemented in C++ to no particular benefit except familiarity for PISM developers.Ed Bueler3+1430-1369
2011-06-01 06:37Minor thing: PetscFinalize() was showing options_left when we did not want, e.g. for 'pismr -version', 'pismr -pismversion', 'pismr -usage'. All of these now work correctly.Ed Bueler2+9-2
2011-06-01 01:09- switched from absolute to relative tolerance in "squared-ness" test in iMcalving.cc. Set (arbitraily) to 1%. That is, for a 5km grid, the grid is considered square if abs(dx-dy) < 50m. - sweep through the manual, trying to give the tables a more consistent lookAndy Aschwanden7+110-105
2011-05-31 23:18Removed a fragile floating-point number comparison test.Constantine Khroulev1+8-5
2011-05-31 20:54Fixed bug #18173.Constantine Khroulev1+9-1
2011-05-31 20:38Small editorial changes to text committed by Torsten. Removal of duplicate PIK description, as rightly suggested by Torsten.Ed Bueler2+24-35
2011-05-31 10:44Added some lines to the manual about pik-improvements. Feel free to modify the text. The table in Sect. 5.11 seems redundant now.Torsten Albrecht2+39-23
2011-05-31 06:20Added documentation to PISMStressBalance and clarified a bit of externals of BlatterStressBalance. Much to do on latter ...Ed Bueler8+152-52
2011-05-30 22:03- replaced "-no_temp" with "-no_energy", updated userman accordingly, and changed $t -> $E reflecting the change from t (temperature) to E (energy). Just for clarity.Andy Aschwanden9+23-23
2011-05-30 21:28We have a Blatter solver! Just kidding ...Ed Bueler7+1746-32
2011-05-27 21:14A one-character change: 3->4 in CMakeLists.txt to identify the current PISM "trunk" as "post-stable0.4".Constantine Khroulev1+1-1
2011-05-27 19:51Minor documentation (browser) fixes.Constantine Khroulev4+10-3
2011-05-27 07:31Fixed stoopid -no_model_mask option handling. Then added no-modeling that I realize should have been there originally: tauc set to large value in no_model_mask==1 areas. Note these fixes are not in stable0.4, but that's o.k.; this is the first post stable0.4 development, I suppose.Ed Bueler1+134-58
2011-05-27 00:20Documentation edits in trunk.Ed Bueler3+438-446
2011-05-26 20:54Minor fixes mirroring ones in stable0.4: compressed .txt files in examples/storglaciaren, fixed CMake rules building source code browsers.Constantine Khroulev10+8-810
2011-05-25 22:55Updated the PISM Cheat Sheet.Constantine Khroulev1+88-70
2011-05-25 18:42Fixed a bug in the "-surface given" and "-atmosphere given" code.Constantine Khroulev1+14-0
2011-05-24 21:12Re-implemented the -ocean_kill mechanism and cleaned up some code.Constantine Khroulev22+161-133
2011-05-24 20:18A two-character change fixing the bug in the -force_to_thk output.Constantine Khroulev1+2-2
2011-05-24 02:25There is a bug in the output of the force_to_thk mechanism. I can't make it produce ftt_mask,ftt_target_thk in output files. Very mystifying. These small changes remove unused boolean and generally should have no effect; but they don't fix the bug either.Ed Bueler2+11-9
2011-05-24 01:32Made the force_to_thk mechanism produce more useful output by default, and documented it better. No change in functionality.Ed Bueler2+44-28
2011-05-23 22:31Changed the way air enthalpy is handled: instead of solving trivial equations we just set values. This seems to have a measurable effect on performance, a reduction of the time spent for those processors which own a lot of ice. In any case, the new code is just as clear.Ed Bueler1+11-8
2011-05-23 21:32Fixed reported bug in pismo caused by stupidity of yours truly.Ed Bueler1+9-7
2011-05-21 08:07Added something I never really wanted to know before: how many time steps did PISM take? Generates short verbosity=1 report at end, of number of steps and average step length. (Note 'pisms -count_steps -verbose 1 -log_summary >> foo.out' gives text report which is essentially compatible with Jed's petscplot script. That parser script is for looking at log_summary output as matplotlib graphs of solver performance. This gives rather different information than the -prof info, which is suited to the load-balancing question.)Ed Bueler3+15-0
2011-05-21 06:47The pythonization of PISM works for me; nice\! petsc4py install is mildly finicky, but that is not our fault. The rest of the install process is adequately documented, but it would be nice to have the 'export PYTHONPATH=...' stage, which requires the user to watch cmake output and set something accordingly, to be automatic as part of the cmake build (in the case where Pism_BUILD_PYTHON_BINDINGS is set); Constantine, is this possible? In examples/python I set svn:executable and I added a README, partly to see if mine was indeed the intended build and usage route.Ed Bueler7+43-2
2011-05-20 22:43Fixed examples/python/pross.pyConstantine Khroulev1+3-1
2011-05-20 22:43Implemented the "constant till yield stress model". Try it using "-hold_tauc" (to enable) and "-tauc" to set a constant till yield stress value (in Pa).Constantine Khroulev10+226-45
2011-05-20 18:12- updated spinup ts figure, still 20km grid though.Andy Aschwanden1+0-0
2011-05-20 00:46Minor stuff: cleanup of searise-greenland scripts and their compatibility with getting-started section; added grid report to mgpoisson.c.Ed Bueler5+11-29
2011-05-20 00:10Userman updates.Constantine Khroulev17+96-92
2011-05-19 23:11- updated SeaRISE Greenland Getting Started example. So far, only 20km runs were done. - replaced all occurrences of "triple" with "melting"Andy Aschwanden26+106-87
2011-05-19 20:00Fixed a bug reported by voice by Constantine and Andy: strict/not-strict inequality issue. The fix was to decide that we should give up on energy conservation for ice that gets 50% water fraction in one time step. Thus 'fully-liquified' warning may happen slightly more. Also removed badly-designed (by me) error checking in enthSystem.cc, which got in the way of fixing the other issue.Ed Bueler2+8-37
2011-05-19 03:33No functionality changes: Cleaned up browser build errors and omissions. Edited comments in PISMYieldStress; tests in searise-greenland certainly suggest no bad changes that I can see, from the creation of that very nice class. Added a script to test -pik -eigen_calving on searise-greenland; looks great from limited testing so far.Ed Bueler16+161-118
2011-05-19 00:46Renamed materials.{cc,hh}Constantine Khroulev22+169-172
2011-05-18 23:35Moved PISM's basal yield stress model into a separate class derived from PISMComponent.Constantine Khroulev17+930-742
2011-05-18 23:23Fixed some command-line options in scripts and documentation.Constantine Khroulev8+13-31
2011-05-18 23:15Fix to bug identified by Constantine: cold mode could generate negative bwat values. Also made bwat,bmelt choices agree between cold mode and (normal) enthalpy mode, by changing the cold mode to agree. W.r.t. SeaRISE-Greenland, compare fields from 'pismr -ocean_kill -e 3 -skip 5 -atmosphere searise_greenland -surface pdd -config_override config_269.0_0.001_0.80_-0.500_9.7440.nc -y 100 -i g20km_pre100.nc' with and without '-cold'. (Results differ in ways I understand, I think. Mostly from enhanced deformation according to Duval et al softening.)'Ed Bueler2+22-24
2011-05-18 19:57- fixed a typo - fiddled with front page figureAndy Aschwanden3+2-2
2011-05-18 08:22I think this is a reasonable version for testing what we want.Ed Bueler2+11-7
2011-05-17 23:44Fiddled with cover image. What do you think?Ed Bueler3+3-3
2011-05-17 23:26- new front page figure - further work on Storglaciaren exampleAndy Aschwanden7+37-14
2011-05-17 17:46Fixed a bug in src/base/iMbasal.cc: default_till_phi was used only if -plastic_phi command-line option is set.Constantine Khroulev3+12-17
2011-05-17 06:50Added an example.Ed Bueler2+117-1
2011-05-16 19:57Fixed bug #18146.Constantine Khroulev1+8-7
2011-05-16 18:43Added a new verification test (flow-line shelf setup using the van der Veen solution to test the CFBC implementation in SSAFD).Constantine Khroulev8+263-8
2011-05-16 18:43Added a man page for pclimate.Constantine Khroulev2+63-2
2011-05-16 18:43Deleted unneeded files in the debian subdirectory.Constantine Khroulev2+0-69
2011-05-15 08:21Some PIK-related changes because it seems to run fine on SeaRISE-Greenland!! Trying it on SeaRISE-Greenland is a robustness test; more soon.Ed Bueler14+180-102
2011-05-12 22:59Fixing scripts building the Debian package.Constantine Khroulev3+5-2
2011-05-12 22:20Deleting empty directory that git forgot ... so says Constantine.Ed Bueler0+0-0
2011-05-12 22:04- added missing figure to repoAndy Aschwanden1+0-0
2011-05-12 21:50filled-in some text for the Storglaciaren example.Andy Aschwanden10+153-156
2011-05-12 21:49Fixed doc/manpages/CMakeLists.txt (Debian installs Python scripts without the .py suffix.)Constantine Khroulev1+1-1
2011-05-12 21:39Added manpages for several PISM executables.Constantine Khroulev11+298-2
2011-05-11 22:30Updated the Installation Manual.Constantine Khroulev5+41-33
2011-05-11 21:31Several fixes: - Changed the name of the anomaly forcing to "anomaly". Use with -atmosphere ...,anomaly -anomaly_temp ... - Moved PetscLogEvent... stuff into PISMProf. - Removed Makefile; added code to set Href to 0 when initializing from a PISM output file that does not have it.Constantine Khroulev8+43-56
2011-05-11 21:03Fix to SSAFD for unreported issue (misused Mat type meant -ksp_type preonly -pc_type lu did not work on one processor). Fixed example in user man explaining role of -ksp_rtol.Ed Bueler2+3-3
2011-05-11 18:03Fixed a publication.Ed Bueler1+6-7
2011-05-10 22:48Various changes.Constantine Khroulev31+1146-895
2011-05-06 17:45Re-implemented "-surface given" and "-atmosphere given" code to make it more like "lapse_rate" modifiers and avoid code duplication.Constantine Khroulev13+440-1195
2011-05-05 22:29* Fixed a bug in IceModelVec2T. (Now it reads the time dimension corresponding to a variable instead of using "t" or "time" variable.) * Reverted the r1681:1682 change related to -skip. Also: added a lengthy comment explaining why the code does what it does and added a warning if both -skip and -no_mass are set.Constantine Khroulev4+69-16
2011-05-05 22:05I believe that this fixes another bug in the enthalpy code. That is, the first bubble in the flowchart in the enthalpy paper was not, but is now, correctly implemented.Ed Bueler1+5-2
2011-05-05 19:13Various fixesConstantine Khroulev5+24-8
2011-05-05 14:12If using -part_grid we have to make sure, that the surface mass balance is smaller than H_average. Otherwise The resultant Href is negative.Torsten Albrecht1+1-1
2011-05-05 13:48If using -part_grid we have to make sure, that the surface mass balance is smaller than H_average. Otherwise The resultant Href is negative.Torsten Albrecht1+14-2
2011-05-04 20:12Documented lapse_rate modifiers. This closes task #7387.Constantine Khroulev5+148-106
2011-05-04 18:04Implemented ice-surface temperature, ice-surface mass balance, near-surface air temperature and precipitation lapse rate corrections as "modifiers".Constantine Khroulev19+518-229
2011-05-04 01:20The version of ssa_forward.py in the repository was a wildy incorrect older draft. Whoops. Now fixed.David Maxwell3+160-218
2011-05-03 23:37Fixed a typo in src/pythonbindings/PISM.iConstantine Khroulev1+1-1
2011-05-02 21:05Fixed src/pythonbindings/PISM.i.Constantine Khroulev1+2-2
2011-05-02 20:49Created new classes to isolate checks used to compute and use mask.Constantine Khroulev48+502-459
2011-05-02 16:30Some more work on Storglaciaren example.Andy Aschwanden5+23-19
2011-05-02 16:28- just saving before switching to another branchAndy Aschwanden2+25-1
2011-04-29 21:53Removed FIXME. Not sure why D2 contribution to strainheating was zeroed out in ice shelves. Now it is computed everywhere, regardless of mask. Comparison in SeaRISE-Greenland, which does have some floating ice, reveals no differences.Ed Bueler1+5-17
2011-04-29 18:53Fixed unreported bug in enthalpy code. Bug exhibited very small effect: In locations where base was at the pressure melting temperature, but the ice above was cold, and the geothermal flux is low, and the conductive flux just above the base was high (because of advected colder ice above), a small refreeze rate could be generated even if there was zero stored water below (vHmelt). The generated refreezes were on order of fractions of millimeters per year.Ed Bueler1+8-2
2011-04-28 22:46Fixed a bug in the -extra_... mechanism.Constantine Khroulev1+16-4
2011-04-28 21:06Fixed a bug in pismr reported by Ed.Constantine Khroulev2+21-2
2011-04-28 20:42More SSA fixes.Constantine Khroulev2+40-18
2011-04-28 20:42Removed SSAFD_PIK regression tests.Constantine Khroulev3+0-86
2011-04-28 20:42Removed SSAFD_PIK.Constantine Khroulev10+2-492
2011-04-28 20:42SSA code cleanup.Constantine Khroulev2+88-93
2011-04-28 20:12examples/python/ssa_forward.py fixed to read command line flags for parameters affecting basal till. Seems to work well now.David Maxwell16+350-302
2011-04-28 13:14CFBC asks for mask instead of thickness now. Piktest examples with circular setups reproduce PISM-PIK output now (fixed a not reported bug). But make test breaks for some ssa-tests now, sorry. Added also a line in pism_const.cc to make PISM run on PIK cluster (is this ok?).Torsten Albrecht12+790-250
2011-04-26 19:23Clarified that some time series are more meaningful than others. Generally the time-derivative time series are bad unless they are generated more frequently than the fastest variability in the coupler inputs to the PISM core. Despite this clarification, yes we do need to return to the issue of good mass and energy accounting.Ed Bueler3+54-23
2011-04-26 16:53Cleanup.Constantine Khroulev1+10-9
2011-04-26 16:53Re-implementing direct forcing.Constantine Khroulev12+544-58
2011-04-26 16:53Added IceModelVec methods: get_grid(), get_ndims(), get_nlevels(), get_stencil_width(), has_ghosts().Constantine Khroulev4+19-16
2011-04-26 13:40vel_bc and bc_locations can now be set as Dirichlet boundary conditions for the SSA also in prognostic mode (pismr) with option -dirichlet_bc. Added vBCMask and vBCvel just for diagnostics, possibly redundant. Fixed a small line switch in energy.Torsten Albrecht9+78-3
2011-04-26 01:36* Fixed unreported bug in enthalpyDrainageStep(): K = k / rho is wrong and K = k / c is right, and they differ by about a factor of 2. This typo in iMenthalpy.cc meant that the base of cold ice saw about half the geothermal flux it was supposed to see. (The typo was *not* present in enthSystem.cc, which is unchanged on this matter.) Thus *all trunk runs since r1599* (and possibly further back) have this factor of 2 too-high conductivity at the base of cold ice. Sorry. This error would have been caught by verification on the enthalpy code (as opposed to all the verification we do have for the cold code).Ed Bueler5+98-54
2011-04-25 18:51Fix for bug reported by email, by Constantine. (Not completely clear what set of circumstances generates the bug. The mask check vMask.as_int(i,j)==MASK_ICE_FREE_BEDROCK is not obviously the same as the ks calculation in enthalpyDrainageStep(), but clearly it *should* be o.k.)Ed Bueler2+4-2
2011-04-24 14:49SSAFEM linearized forward and adjoint problems now working (not just in serial) code needs cleanup, documentation in a dire wayDavid Maxwell3+169-75
2011-04-24 02:39EnthalpyConverter::isTemperate() was not returning the right thing in cold mode, with effect that diagnostic tempicethk_basal was zero, quite inaccurately. Now polythermal and cold modes produce comparable values for tempicethk_basal. The new code is easier to understand and debug; efficiency probably about the same but who cares.Ed Bueler1+4-5
2011-04-24 01:34pacman: (1) puts wobble in the 15th decimal digit (probably last bit) in the ivol.nc output of test_12.sh, thus that test was failing; this make it pass, which is reasonable in this context, if I may say so myself ... (2) has a more-pedantic-than-thou setting, which complains about missing newlinesEd Bueler2+3-3
2011-04-24 01:09Correction of constant in GPBDL flow law. Fix typo in exactV.py.Ed Bueler4+9-9
2011-04-22 08:18looked at mismip and see improvements ... but things are not that great either, with ice shelf not looking the right shape (e.g. like van der Veen solution, which they should converge to in steady state); not clear if new PIK options improve mismip for nowEd Bueler4+18-18
2011-04-22 00:08Fixed a bug in PBPointwiseIsostasy.cc (topg_last was not initialized properly).Constantine Khroulev2+4-1
2011-04-21 20:09Code maintenance.Constantine Khroulev23+326-275
2011-04-21 05:04Added SSAFEM_Forward for computing SSA linearized forward and adjoint maps. Forward is working fine, adjoint needs some work.David Maxwell13+1017-69
2011-04-20 18:20Fixed a bug in IceModelVec2T that was affecting "-surface given" and "-atmosphere given" runs.Constantine Khroulev12+85-25
2011-04-20 16:30Cleaning up the SSA code; more work on exactV.py: reproduces PISM's versions of Figures 6b and 6c from Albrecht et al.Constantine Khroulev4+86-32
2011-04-19 20:40* Removed the Hmelt diffusion mechanism. Neither the SeaRISE-Greenland runs nor the efgis runs need it for their stable function. (FIXME: the PST derived class could re-implement it for historical reasons only.)Ed Bueler7+73-107
2011-04-19 17:29Fixed bug #18049.Constantine Khroulev3+31-31
2011-04-18 22:25Merged SSAFD_PIK and SSAFD. Run pismr with "-ssa_method fd -cfbc" to use CFBC.Constantine Khroulev13+393-251
2011-04-18 21:21Fixed an unreported bug I introduced in r1638.Constantine Khroulev1+1-1
2011-04-18 18:16Cleaning up mass continuity code.Constantine Khroulev3+123-118
2011-04-18 17:55Finished task #7377.Constantine Khroulev6+38-5
2011-04-17 07:16Removed diagnostic 'bueler_brown_f' from output_big list. It is o.k. to have code which computes it, I guess, but only an explicit request should run that code, because we are not using f(|v|) anymore anyway.Ed Bueler1+1-1
2011-04-17 03:33Deleted spurious debugging output causing regression test failures.David Maxwell1+0-1
2011-04-17 01:46Uninteresting merge keeping small changes up-to-dateDavid Maxwell18+1107-3439
2011-04-16 22:42Further clean-up of column system reporting and error reporting. Moved generic code for that purpose out of iMenthalpy.cc. Now for each of {age,enthalpy,temperature} solves, '-view_sys -id X -jd Y' will give a usable m-file.Ed Bueler9+159-155
2011-04-16 20:16Documented PADirectForcing.Andy Aschwanden1+58-49
2011-04-16 01:52Essentially a documentation commit. I did change the way columnSystemCtx::viewSystem() saves a text file, so that the result can be read into Matlab and examined; good debugging tool for changes to the enthalpy, age, temperature codes.Ed Bueler4+84-40
2011-04-15 19:56Fixed up Maxima code; replaced the guts of SSAFD::assemble_martix() with Maxima-generated C code.Constantine Khroulev9+414-220
2011-04-15 19:56Renamed c[01]+ in SSAFD and SSAFD_PIK assemble_matrix().Constantine Khroulev2+54-54
2011-04-15 04:13Fix of bug #18028. My fault; beta uninitialized in my code committed in r1617.Ed Bueler1+5-5
2011-04-14 22:45Added Maxima code and a .tex file documenting PISM's finite-difference discretization.Constantine Khroulev10+642-0
2011-04-14 22:45Fixed src/verif/tests/exactV.pyConstantine Khroulev1+3-5
2011-04-14 15:49little more text in tex-file describing the implementation of the SSA neumann boundary condition...Torsten Albrecht1+33-19
2011-04-14 10:24updated the description of the calving front boundary condition in src/base/stressbalance/ssaBCTorsten Albrecht1+76-62
2011-04-13 06:46Made some numbers associated to drainage mechanism into 'official' config parameters.Ed Bueler4+58-27
2011-04-13 04:36minor changes which may help with debugging, and killing pedantic messagesEd Bueler2+7-2
2011-04-12 18:32* Fixed modeling conflict caused by T_0=0 K in ICMEnthalpyConverter. Now you can do "pismv -test G -o foo.nc" followed by "pismr -i foo.nc -surface constant" and not get an error, for example.Ed Bueler7+224-183
2011-04-12 02:07Three trivial changes; essentially affect only documentation.Ed Bueler3+4-5
2011-04-11 21:57Added src/verif/tests/exactV.py, plotting ice thickness and velocity profiles plus the "theoretical position of the free boundary at time t" (equations 11, 12 and 13).Constantine Khroulev3+98-2
2011-04-11 18:41Started implementing verification test V (flow line setup from Albrecht et al, 2011 compared to the Van der Veen solution).Constantine Khroulev3+198-61
2011-04-10 19:02Added some basal resistance in ice-free-bedrock areas in SSAFD object; seems to work in full searise-greenland run on 20km grid. (SSAFEM object not touched on this topic.) The rest of this commit is a documentation, and config parameter naming convention, maintenance sweep; I think these docs are in pretty good shape for stable0.4.Ed Bueler8+143-117
2011-04-10 06:15Better details in the way the vertical conduction in the enthalpy problem is handled when there is a jump from the (larger) cold ice conductivity to the (smaller) temperature ice conductivity. And BOMBPROOF docs are updated to match paper.Ed Bueler2+103-290
2011-04-08 23:05Code clean-up. Just re-formatting so that I can see what is going on...Constantine Khroulev12+712-1114
2011-04-08 22:54Minor improvements to Getting Started material.Ed Bueler2+31-41
2011-04-08 22:53Five years after Jed made it work fine, I have finally made the error handling from the columnSystemCtx objects a little better. (Yes, it bit me again.) And fixed indents in iMenthalpy.cc.Ed Bueler8+91-91
2011-04-08 18:51Finished implementation of PADirectForcing, which works similar to PSDirectForcing, but using artm and precip instead, which are then handed down to the surface model. Use -atomsphere given -bc_file foo.nc to activate it. Documentation is coming soon. Thanks Constantine for your help!Andy Aschwanden2+10-20
2011-04-08 18:51- preliminary work on new PADirectForcing model, which should be similar to PSDirectForcing, but read precip instead of acab, then pass it on to the surface model. Compiles and runs, but still produces wrong outputAndy Aschwanden5+321-23
2011-04-08 04:37Drainage implemented which matches paper (but with updated D(omega) values ... O(1) is too big ...). No max on omega is enforced any more, but values above 0.01 are relatively rare. The flow law only sees 0.01 at max.Ed Bueler3+48-51
2011-04-07 22:37Fix to cold-version pisms bug in 1608.Ed Bueler2+2-6
2011-04-07 22:04Minor fixesConstantine Khroulev7+59-13
2011-04-07 20:59Code cleanup (boring).Constantine Khroulev10+349-342
2011-04-07 20:37Adopted EnthalpyConverter which matches AschwandenBuelerBlatter; works fine in searise-greenland. Cleanup in iMenthalpy.cc and (trivially) iceModel.cc. Went back to default unequal spacing in spinup.sh; works fine.Ed Bueler5+25-219
2011-04-07 19:50Some work on Storglaciaren, but more is coming soon. I've got a clearer picture now of what goes into the tutorial.Andy Aschwanden6+172-43
2011-04-07 19:50implemented the new convert mechanism into PSElevation. Nice stuff.Andy Aschwanden1+34-30
2011-04-07 19:12Isolated surface->update() and ocean->update() calls.Constantine Khroulev36+390-468
2011-04-07 19:11Implemented convert(value, "unit spec", "unit spec"). Try convert(0, "Celsius", "Kelvin").Constantine Khroulev3+47-5
2011-04-07 03:16Sorry about that: in r1599 I was not following our own advice to ourselves about the application of the basal boundary condition. With this commit we are. The result is pleasant in searise-greenland.Ed Bueler1+22-12
2011-04-06 19:14Minor cleanup: lat and lon valid ranges, more robust vertical grid extension code, build system fixes.Constantine Khroulev5+25-7
2011-04-06 04:44* IceModel::enthalpyDrainageStep() now basically matches paper, and produces reasonable behavior in runs tested so far. Code got shorter, too!Ed Bueler5+68-167
2011-04-05 21:08Fixed a bug in the BTU's update interval contiguity check (was sensitive to rounding errors).Constantine Khroulev1+11-1
2011-04-05 19:46Fixed a bug in src/coupler/PALapseRates.cc (same as the one fixed in r1596 in stable0.3).Constantine Khroulev1+2-0
2011-04-05 02:57Added option -acab_limits {list of 2 numbers} to PSElevation. With, e.g., "-acab_limits -3,0" can be made sure that the mass balance is 0 above h_max, and thus Storglaciaren can't grow above the current bergschrund. Added a few lines to the documentation. Just a toy so far.Andy Aschwanden3+38-6
2011-04-04 22:27Clean-upConstantine Khroulev4+78-13
2011-04-04 19:01Fixed util/flowline.py (now it supports files with different variable orders) and replaced "const char foo[]" with "string foo" in many places.Constantine Khroulev9+109-105
2011-04-04 18:44line added to CMakeLists.txt so 'make userman' works in build/Ed Bueler2+2-1
2011-04-04 18:03more work on Storglaciaren. Some docu, and adding scripts for 3-dimensional case as wellAndy Aschwanden7+105-67
2011-04-04 18:03Further work on the Storglaciaren example. Included 3-dimensional caseAndy Aschwanden9+1063-0
2011-04-03 01:41- initial work on Storglaciaren worked exampleAndy Aschwanden11+382-341
2011-04-03 01:41- got started on flowline-storglaciaren exampleAndy Aschwanden4+438-0
2011-04-03 01:41- redesigned PSElevation to make it compatible with the needs of a Scandinavian-type polythermal glacier. - Updated docu followsAndy Aschwanden3+53-65
2011-04-02 02:59Split iMgeometry.cc into iMpartgrid.cc and old stuff. No code changes, just rearrangement to tidy up.Ed Bueler4+651-621
2011-04-02 02:47I have looked at, and modestly-refactored, a bit of Torsten's code:Ed Bueler5+262-45
2011-04-02 00:49Fixed bug #17978.Constantine Khroulev5+47-40
2011-04-01 23:25 - documented PSElevetation - massaged console output of PSElevationAndy Aschwanden2+40-10
2011-04-01 18:29 - renamed option -bc_lapse_rate to bc_artm_lapse rate, and added new option -bc_acab_lapse_rate (meters per year per kilometer). - fixed a typoAndy Aschwanden4+36-9
2011-04-01 03:12 - implemented a new surface model PSElevation, which parametrizes artm and acab as a function of surface altitude. Parameters include equilibrium line altitude, temperature (in degC) at the equilibrium line altitude, temperature gradient, and mass balance gradients in the ablation and the accumulation area. The user can also set a minimum and maximum altitude below and above which a constant mass balance is applied. Documentation in the User's manual is coming soon. In the meantime, try it out with:Andy Aschwanden6+279-2
2011-03-31 22:43Looked through Torsten's code and made progress on generating regression tests. Everything looks good in r1579. Run run_part_dist.sh script, and see comments therein, as source of regression tests. Are these r1580 results good in Torsten's opinion?Ed Bueler5+89-146
2011-03-31 22:37Updated test_21.sh based on test K: in r1578 it was failing because it was better. Also removed invalid -zb_spacing option from searise script.Ed Bueler2+8-8
2011-03-31 21:11final part of fix to bug #17971, which was closed prematurelyEd Bueler1+2-1
2011-03-31 20:34Fixed bug #17974.Constantine Khroulev1+12-0
2011-03-31 20:24* in IceModel::step(), moved "dtTempAge=0.0" reset *INSIDE* a "(updateAtDepth && do_energy)" conditional; this fixes a major unreported bug in -skip N runs * replaced dtTempAge (member of IceModel) with pair t_years_TempAge,dt_years_TempAge; this is a sub-optimal solution because this pair must be kept compatible with pair grid.year,dt and because couplers are called with (potentially) non-sequential time stepsEd Bueler9+41-36
2011-03-31 01:30Updated some comments to reflect new BTU code. Happy with the new, smaller energy-related code. (But will report bug once I confirm it occurs in 1574.) Deleted delete_manifest ;-).Ed Bueler5+33-64
2011-03-30 23:42Various fixes related to the BTU code, its initialization, removing bedrock grid information from IceGrid and defining variables. Seems to work.Constantine Khroulev41+385-3068
2011-03-30 13:40Parts of the log text were missing in previous commit, sorry: Some changes with regard to option 'calving_at_thickness' to make sure, that only ice shelf grid cells along the boundary to ice free ocean neighbors are calved off. Method 'killEasyIcebergs' should be called when option 'kill_icebergs' is set and calving is applied. Otherwise isolated grid cells on the ocean containing ice (H or Href) should not occur.Torsten Albrecht1+1-1
2011-03-30 13:34Parts of the log text were missing in previous commit:Torsten Albrecht1+1-1
2011-03-30 13:31Some changes with regard to option calving_at_thickness to make sure, that only ice shelf grid celTorsten Albrecht4+28-11
2011-03-29 22:14Update of comments related to BTU and new enthalpy understanding.Ed Bueler5+72-64
2011-03-29 21:16added one line to set enthalpy in ice-free columnsEd Bueler1+1-0
2011-03-29 20:20Fixed a bug. (Automatic grid extension caused corruption of 3D diagnostic outputs.)Constantine Khroulev1+12-0
2011-03-29 19:44Minor fixes.Constantine Khroulev6+41-22
2011-03-29 18:40Fixed bug #17952 plus more work on the BTU object.Constantine Khroulev15+311-213
2011-03-29 08:21Added script by Torsten and Rica, posted for bug #17950, as a potential regression test of -part_grid -part_redist; bug #17951 currently blocks such use. Minor changes to manual to shorten by a couple of pages.Ed Bueler4+280-24
2011-03-29 01:51Updated the authorship of the installation manual.Ed Bueler1+9-16
2011-03-29 01:46Updated the look of the browser front pages.Ed Bueler2+24-34
2011-03-29 01:31Changed to 'the PISM authors' authorship for User's Manual. Please tell me what you think and what your one-line authorship entry should really say. Updated acknowledgements. First draft of documentation of new PIK options; please move around, correct, update etc.Ed Bueler3+79-21
2011-03-29 01:29Looked through Torstens code and like it. Moved config.get() calls out of loops. Alphabetized options for consistency, and fixed spelling of -calving_at_thickness for consistency. Updated related comments. Added single-call method killIceBergs() which manages sequence findIceBergCandidates(), identifyNotAnIceBerg(), killIdentifiedIceBergs(). (Should killEasyIceBergs() be called anytime -part_grid is selected?) We need regressions for these methods.Ed Bueler6+81-80
2011-03-28 19:25Squashed commit of the following:Andy Aschwanden3+59-4
2011-03-27 19:42minor stuff: systematically shortened long test names; clobbered some pedantic warnings; made a couple of slow tests fasterEd Bueler8+43-43
2011-03-27 01:40Code change which should fix bug #17950. That is, method IceModel::enthalpyAndDrainageStep_new() was fixed, which resolve things once the PISMBedrockThermalUnit is fully-attached.Ed Bueler4+20-5
2011-03-27 01:13I did these things because it was fun to look at the new Torsten-PIK code:Ed Bueler10+121-116
2011-03-25 21:24Various changes necessary to make it easier to have 3D fields using different vertical grids.Constantine Khroulev77+1034-1115
2011-03-25 21:23Added the new mask ("new_mask") as a diagnostic quantity.Constantine Khroulev5+128-3
2011-03-24 00:09Addition of three references. Clobbered pedantic compiler warnings; by GNU '-pedantic -Wall' standard, we are warning-free.Ed Bueler4+37-5
2011-03-23 23:51Missing text from previous commit: Added PIK-style ocean and surface coupler (POConstantPIK.cc and PSConstantPIK.cc (not really constant)). They are called with options '-surface pik' and '-ocean pik'. Ice Surface temerature 'artm' is parameterized according to latitude and surface elevation (Martin et al., 2011, Eq. 2.0.2). Accumulation rate 'acab' is read from file. Sub-shelf melting is parameterized analogous to Beckmann & Goosse, 2003 (or Martin et al. 2011, Eq. 2.0.6+7) with a tuning parameter, meltfactor, which is set to its default value, so far (no option yet). It's definitely much easier to couple and parameterize boundary fields, compared to stable0.2! :)Torsten Albrecht1+1-1
2011-03-23 23:24PIK simulation of the Antarctic Ice Sheet might be potentially reproduced with the current code version. For an Antarctica setup as preprocessed as for the SeaRISE-Antarctica example (but acab, artm) and bootstrapped from it, a prognostic pismr command with all the necessary options may look like this:Torsten Albrecht12+426-5
2011-03-23 01:09added the option '-e_ssa x', which enhances SSA-velocities (as suggested by Ying et al. 2010: Enhancement factors for grounded ice and ice shelves inferred from an anisotropic ice-flow model). The ssa_enhancement_factor just scales the staggered viscosity in SSAFD.cc. Default value is 1.0, of course.Torsten Albrecht3+26-1
2011-03-22 20:12Work around for NetCDF3 vs. NetCDF4 nc_type in PISM.i cleanup of ssa_testj.py example python scripts are now executableDavid Maxwell3+39-29
2011-03-22 00:55Added iMcalving.cc with strain-rate calculation (and eigenvalues edot1 >= edot2), and the application of the strain-rate based calving rate, C, at the floating calving front as C=k*edot1*edot2, edot1,2>0. Option '-eigen_calving k' calls the routine and sets k in [m s] as the calving law constant, which is thought to comprise material properties at the calving front.Torsten Albrecht9+466-21
2011-03-21 18:53Added improved support for python PISM.verbPrintf. Moved common code out of examples/python into the PISM python package.David Maxwell7+468-432
2011-03-21 18:21Switched to using the old form of add_test() to make regression tests work on marmaduke, which has CMake 2.6 patch 4 installed.Constantine Khroulev1+1-2
2011-03-21 18:00Fixed CMakeLists.txt: now it actually uses PETSC_COMPILER as a hint when looking for MPI.Constantine Khroulev1+1-1
2011-03-21 04:59Added support for ^C to python bindings with user specified behaviour: exit immediately or wait for the last call into C to complete.David Maxwell8+127-8
2011-03-20 17:16Polished the python bindings for working with command line options. Finished pross.pyDavid Maxwell11+515-310
2011-03-19 02:35More python bindings (e.g. SSAFD) Made the python SSATestCase, ssa_testj more pythonic Implemented pross as a (python) SSATestCase: errrors match C implementation so far, but am not yet reporting vs. riggsDavid Maxwell4+686-191
2011-03-18 05:14Synchronized NCVariable::define with NCConfigVariable::define. Abstract base class NCVariable::define had a default third argument write_in_glaciological_units=true Subclass NCConfigVariable interface for define did not specify the default argument explicitly; this is now done. The mismatch confused the SWIG generated python bindings, which thought that NCConfigVariable was still abstract.David Maxwell1+2-1
2011-03-18 01:46Commented out line in interface: IceModelVec3Bedrock::getValColumnPL(PetscInt i, PetscInt j, PetscInt ks, PetscScalar *valsOUT); This method isn't implemented, but the python bindings make code that link against it.David Maxwell2+4-4
2011-03-18 01:31First pass at python bindings for (a very, very small part of) PISM.David Maxwell14+4195-1
2011-03-17 08:17This one-character commit is needed so that doxygen bothers to produce any docs whatsoever about PISMBedThermalUnit. I hate this aspect of doxygen ... no good error reporting on matching LaTeX brackets.Ed Bueler1+1-1
2011-03-17 07:31enthSystemCtx object added, which is identical to iceenthOnlySystemCtx, actually. New PISMBTU-using enthalpyAndDrainageStep() added. C.K.: Now I think I've done all I can and I think you should do the rest, acknowledging it may break. Note I *have* run all the new code and it looks reasonable-looing result.Ed Bueler8+796-9
2011-03-17 05:09* temperature from IceModel seen as z=0 Dirichlet condition for PISMBedThermalUnit now has correct values at ice free land, ocean * PISMBedThermalUnit no longer needs any knowledge of the ice enthalpy ... this is goodEd Bueler8+185-133
2011-03-16 19:38Actually fixed bug #17908 (or at least I think I did.)Constantine Khroulev2+7-1
2011-03-16 19:22Fixed bug #17908.Constantine Khroulev4+38-29
2011-03-16 04:581) Added killEasyIcebergs() in iMicebergs.cc which might be useful when applying strain-rate based calving. Single grid cells of floating ice or single partially filled grid cells, as well as 'ice noses' arising from the ice shelf front can be cut of.Torsten Albrecht3+210-6
2011-03-16 02:57* IceModel now owns a PISMBedThermalUnit. It is hooked-up for most input and output, but there is a missing input-file initialization routine, and no attention to regridding. It uses tempSystemCtx_new with no apparent issues.Ed Bueler8+349-5
2011-03-16 01:03* Minimal code reorganization: age model in iMage.cc, columnSystem.[hh|cc] in correct location. * Two regression tests speeded up without loss of function.Ed Bueler14+1332-965
2011-03-16 00:06Eradicated last vestiges of MASK_SHEET from SSAFEM.David Maxwell2+12-37
2011-03-15 23:18Minor things:Constantine Khroulev3+4-6
2011-03-15 18:28Cleaning up the mask-related mess.Constantine Khroulev47+545-498
2011-03-15 08:48PISMBedThermalUnit works great. Try bturun.sh to see. Ready for attaching to IceModel and IceGrid, which will generate substantial simplification. bturun case with -Mbz 41 and -dt 0.1 a natural case for regression; also -Mbz 1 in same regression. Will work with C.K. on this. Other cleanup, including clobbering pedantic warning about pism_const.hh.Ed Bueler6+272-207
2011-03-15 04:16Added the missing iMicebergs.cc for -kill_icebergs option.Torsten Albrecht1+255-0
2011-03-15 01:13Floating Ice Shelf Patches detached from grounded area (icebergs) can be identified and killed with option -kill_icebergs. In this case also the vIcebergMask is written to nc-file (just for disgnostic purpose). No reporting of (negative) ice fluxes yet.Torsten Albrecht10+85-8
2011-03-14 21:17Found the issue: Exact Test K is a surprisingly tough test because of same issue that requires 'we redefine the initial state to have a finite eigenfunction expansion'. That is, I needed to read the documentation ... that I wrote. Now upward heat flux at base is evaluated with second-order difference formula.Ed Bueler4+161-87
2011-03-14 17:47Updated examples/eisross/rossplot.py so that it works with different output file variable orders. (I.e. rossplot.py transposes 2D fields if necessary.)Constantine Khroulev1+18-6
2011-03-14 10:23Added some comments and tidied up the partially filled grid cell scheme with residual mass distributionTorsten Albrecht2+94-84
2011-03-14 07:51btutest runs and gives vaguely reasonable number. Discussion with C.K. needed.Ed Bueler3+101-87
2011-03-13 21:11btutest pretty much as I want it for now ... yes it compiles and links ... I don't know if it runsEd Bueler2+52-13
2011-03-13 20:24btutest under developmentEd Bueler1+49-10
2011-03-13 19:58Test changes: * regression on ssafem_test_plug.sh speeded up by factor of 10 and non-equal tested * equal vertical spacing used in Test K by vfnow.py; helps with bedrock thermal layer testing; I believe I am fixing this bug where Test K produces crap on nonequal spaced vertical grids * upward heat flux output added to exactTestK.c; simpleK.c updated accordinglyEd Bueler6+50-22
2011-03-12 20:15BTU and its test much closer to what I want. Compiles. btutest not functional.Ed Bueler3+210-85
2011-03-12 03:38Partially filled grid cell scheme is capable of redistributing residual ice masses (after being fully filled with respect to Hav) in calculateRedistResiduals(), called in the end of massContExplicitStepPartGrids(). Residuals doesn't need to be model_state. Hav is Scalar now. Reporting needs to be fixed.Torsten Albrecht4+200-45
2011-03-11 23:54Modified IceModel::updateYieldStressUsingBasalWater() to compute tauc ghosts "redundantly".Constantine Khroulev2+6-7
2011-03-11 23:26Cleaning up...Constantine Khroulev28+382-1348
2011-03-11 20:12More work on the mass-continuity update...Constantine Khroulev4+226-213
2011-03-11 01:32Fixed SSAFEM indexing bug regarding computing hardness from enthalpy. IceModel now knows about SSAFEM for testing and debugging purposes.David Maxwell4+34-3
2011-03-11 00:33The new option -part_grid turns on the partially filled grid cell scheme for the advance of the ice shelf front (massContExplicitStepPartGrids). Href(i,j) accounts for the current ice mass in a partially filled grid cell and Hav(i,j) for the average over adjacent floating ice shelf grid cells. Both are model_state so far and written to the nc-file. In the case of an advancing water terminating outlet still thin ice shelves can evolve. The redistribution of residual ice mass (when partially filled grid cell is considered to be full) is not yet implemented.Torsten Albrecht7+397-39
2011-03-11 00:20Fixed eisross.py: now all Dirichlet boundary contition locations are specified in bcflag.Constantine Khroulev1+4-17
2011-03-11 00:20Fixed up btutest.ccConstantine Khroulev1+18-4
2011-03-10 23:45This seems to read the grid (from pisms output) but perhaps dz_fine is not set correctly?Ed Bueler2+10-50
2011-03-10 23:23Added what should have been in the previous commit: btutest needs to "prepare" a file before writing and avoid over-writing the file just created...Constantine Khroulev1+3-2
2011-03-10 23:17Added the code necessary to ask the "btu" to write its state.Constantine Khroulev1+23-11
2011-03-10 22:37Latest installment of PISMBedThermalUnit.Ed Bueler2+106-43
2011-03-10 21:41 CleanupConstantine Khroulev20+262-423
2011-03-10 18:33Fixed parallel crashes for SSAFEM: Counting bug in FEElementMap Made enthalpy, tauc, and Dirichlet data 'local' (i.e. with ghosts) in pross Minor cleanup of SSATestCase reporting outputDavid Maxwell5+36-21
2011-03-10 02:43Barely half-baked but improving BTU.Ed Bueler3+198-59
2011-03-10 01:06Fixed MASK_SHEET semantics in SSAFEM (i.e. MASK_SHEET in mask vs. MASK_SHEET in bc_locations) Dispersed SSAFactory stuff out of SSATestCase for use in other contexts. Added choice of SSAFD vs SSAFEM to pross. SSAFEM works for pross: (FEM chisq = 3794 vs. FD chisq = 3647)David Maxwell8+156-136
2011-03-10 00:09Added a subsection about modifying PISM code with pointers to the Installation Manual and the browsers, plus a mention of the role of regression tests and version control systems.Constantine Khroulev3+42-5
2011-03-09 22:41Finished task #7350.Constantine Khroulev12+77-27
2011-03-09 22:33Added the reference date, plus calendar and axis attributes to time axes used by diagnostic time-series.Constantine Khroulev2+62-28
2011-03-09 21:13Cleanup: (1) removed shelves_drag_too experimental kludge, which never worked well and is now being de-justified by David and Torsten's work; in any case new SSA implementations shouldn't even see it; (2) removed 'OpenIB WEIRDNESS' subsection from installation manual (PDF); if users report it as an installation bug then we can always look it up ...Ed Bueler7+1-60
2011-03-09 20:06 Modified PSDirectForcing; added the ability to change the reference date in output files.Constantine Khroulev10+110-17
2011-03-09 19:50Minor tweaks of source code browser settings.Constantine Khroulev1+5-5
2011-03-09 02:15These small changes to stdout help me understand what the SSAFD_PIK (and SSAFD) implementation is doing.Ed Bueler5+22-0
2011-03-08 22:05I'd like to avoid putting PDFs in the src/ tree. pdflatex works fine.Ed Bueler2+0-0
2011-03-08 21:32Fixed memory allocation bug for SSAFEM running in parallel. Cleaned up exact solution test cases. SSAFEM is passing all of these.David Maxwell6+76-24
2011-03-08 21:28Reverted accidental changes in examples/eisross.Constantine Khroulev2+34-35
2011-03-08 21:19SSA Neumann Boundary condition has been added in the same manner as in PISM-PIK code (still without partially-filled grid cell scheme). Diagnostic testing with pross -boot_file ross.nc -Mx 147 -My 147 -ssa_rtol 2e-3 converged. Some more details about the used stencil are given in src/base/stressbalance/ssaBC/Torsten Albrecht7+719-36
2011-03-08 03:24I can confirm that SSA-as-sliding-law is again 'on' in the PST mode. This commit to the PST script comes closer to my understanding of how to reproduce PST results under stable0.4ish semantics. But the basal melt rate is still not what is expected. In any case, basal melt rate is critical to reproducing basic PST results, and we are not reproducing the basic experimental idea unless the time zero field of bwat (i.e. bwat in P0A.nc) is similar to Figure 5(right) in Bueler and Brown (2009).Ed Bueler1+26-25
2011-03-07 22:26Another PST fix (I forgot that not all PST runs have SSA sliding).Constantine Khroulev1+5-5
2011-03-07 20:51Fixed PST (I think).Constantine Khroulev1+14-13
2011-03-07 20:34Added the (empty) new class for the PIK's CFBC implementation.Constantine Khroulev3+50-0
2011-03-07 19:02MISMIP fixes (compiles and runs; needs more testing).Constantine Khroulev3+43-45
2011-03-07 17:40Minor cleanupConstantine Khroulev4+17-16
2011-03-06 08:35Draft code to isolate function of bedrock thermal layer from rest of energy balance. This should clean up columnSystemCtx nonsense, and address some of the misery this code has caused us and Florian Z. Also this isolation is completely compatible with new, clearer view of enthalpy method. Draft code compiles, but not linked to any IceModel, and btutest does not run.Ed Bueler4+373-3
2011-03-06 04:49PIK-style (finite-volume-style) upwinding of SSA-generated velocity, as a term in the mass continuity equation, is adopted. See Albrecht et al 2011 (TC) and Winkelmann et al 2010 (TCD). Because Test E produces basal sliding velocity which is incorporated into mass continuity the same way an SSA velocity field would, it actually serves as 2D verification. See very slight changes to Test E regression results in test/regression/test_15.sh.Ed Bueler2+68-67
2011-03-04 23:13Cleaned up by removing two empty directores.Constantine Khroulev1+3-2
2011-03-04 22:58Fixed bug #17842.Constantine Khroulev2+40-41
2011-03-04 22:04Squashed commit of the following:Andy Aschwanden1+1-2
2011-03-04 20:31Code to do PIK-style upwinding of SSA-derived (or SIA_Sliding-derived ... in any case, NON-diffusive) velocity in IceModel::massContExplicitStep(). But turned off for now. Testing shows same processor-dependence as in SIA_Sliding object in r1477 and earlier; i.e. no fault of new code.Ed Bueler2+29-0
2011-03-04 19:36Switched to using CTest for regression testing.Constantine Khroulev74+1359-1701
2011-03-03 19:43More clean-up.Constantine Khroulev38+1131-1273
2011-03-02 20:51Added some more documenting comments for the Util. Browser.Constantine Khroulev5+139-23
2011-03-02 00:03Disabled compiler warnings for src/udunits/* and src/earth/cubature.cConstantine Khroulev6+14-11
2011-03-01 22:39Restored SIA sliding in EISMINT II experiments 'G' and 'H'.Constantine Khroulev12+696-593
2011-03-01 22:30Cleared up the meaning of "make browser_base". (See task #7341.)Constantine Khroulev1+8-11
2011-02-28 23:30Fixed the run-time log10(nuH) viewer. Also: some IceModelVec re-factoring.Constantine Khroulev10+271-317
2011-02-28 22:58Last round of SSAFEM documentation for now.David Maxwell4+219-125
2011-02-26 01:06Reverted recent MASK_SHEET->MASK_BC changes because this is the only recent SSAFD-related code change and some -ssa_sliding runs stopped converging.Constantine Khroulev13+71-42
2011-02-25 21:44MASK_SHEET -> MASK_BC in the finite element code; ssa_testj -ssa_method fem passes again. Minor documentation changes.David Maxwell8+29-42
2011-02-25 21:11Fixed SSATestCase.cc: I missed one spot where IceGrid::mapcoords was still used.Constantine Khroulev1+1-3
2011-02-25 20:31Updating the User's Manual.Constantine Khroulev10+224-89
2011-02-25 20:31Fixed max_timestep() in PSDirectForcing.Constantine Khroulev7+136-32
2011-02-25 20:31Various clean-up tasksConstantine Khroulev29+139-272
2011-02-25 02:24Ongoing documentation work. Moved SSAFEM_util.* to FETools.*. to reflect the overhauled contents.David Maxwell7+765-708
2011-02-24 08:22First installment of doxygen comments for the finite element code.David Maxwell4+266-203
2011-02-23 06:23Overhaul of FEM SSA code for clarity, maintainability. Still needs documentation; this is next.David Maxwell4+562-645
2011-02-23 05:57got some more pedantic warnings ...Ed Bueler5+14-14
2011-02-23 05:55This addresses a topic which came up in conversation with DM. I propose the following copyright message conventions: (1) if you create a new file with new concepts, put your name and the current year at the top; (2) if you modify an existing file in any substantial way then add your name and extend the date range to the current year; (3) if you copy a file and modify it, you have to decide what to put at the top. (Not that any of this really matters. We probably could just write 'copyright 2099 the PISM team' ... But the current convention provides PISM modifiers some sense of the history of the source. E.g. I checked the log to find out when Jed started the SSA FEM implementation.)Ed Bueler5+7-6
2011-02-23 03:23Addressing a few more warnings, both pedantic compiler and doxygen.Ed Bueler6+34-20
2011-02-23 03:20Addressing (why? ... I don't know ...) pedantic compiler warnings in C code exact tests. Mostly complaints about mixing declarations and code, and about malloc needing size_t.Ed Bueler7+68-59
2011-02-23 02:59Added-back additional subdirectories so 'base' browser picks up more of source code. Changed path on pism.sty to avoid error message when building browser. Attempt to address hard-to-understand doxygen complaints about tag refs in bedrough.txt; it looks o.k. ...Ed Bueler4+14-9
2011-02-21 21:24New directory for SSA test cases. New test cases added: convergence to constant, exponential, and plug flow solutions. Bug fixes to SSAFEM: non-equal grid spacing, hidden duplicate scaling of viscosity by H/2David Maxwell12+1355-578
2011-02-20 08:39Editorial changes to a single stdout line.Ed Bueler1+4-4
2011-02-20 03:27* Removed getWaterFractionLimited(), which wasn't needed anyway.Ed Bueler4+194-91
2011-02-20 01:58Changes helpful to new EnthalpyConverter, but these hav no current functional effect. ICMEnthalpyConverter has been cleaned up only; no change.Ed Bueler3+44-26
2011-02-19 01:25The changes, which only become activated with 'CHANGE_BASAL_MELT 1' setting, in iMenthalpy.cc, show much of what is needed to update the enthalpy code. Note new Test 0 will allow regression once the basal melt rate is corrected. (Next commit will include an alternate EnthalpyConverter, again not 'hooked up'.)Ed Bueler3+101-22
2011-02-19 01:07Undoing r1431 commit: -pdd_annualize not changing anything but lowering performance on 20km (coarse) grid. More thought needed.Ed Bueler1+2-2
2011-02-17 11:13Fixes for the previous commit.Constantine Khroulev2+2-5
2011-02-17 10:51Re-organized CMake scripts building PISM documentation.Constantine Khroulev11+179-141
2011-02-17 00:35I looks like these Makefiles are unmaintained; please restore but update if I'm wrong. I was confused about how the new/split browser was being built ... until I saw which cmake flags to choose. Not sure whether doc/COOKBOOK or doc/README* are worth maintaining, but at least now they don't confuse about where the docs are built or appear.Ed Bueler5+7-102
2011-02-16 20:19More doxygen comments. Also: documented pismebm, some clean-up.Constantine Khroulev7+235-44
2011-02-16 09:05Clobbering compiler warnings. More doxygen comments. More IceModelVec re-factoring.Constantine Khroulev31+271-256
2011-02-15 23:28Minimal maintenance on examples/pddtune/. Indeed the -pdd_annualize does not seem to affect outcome at all, because the runs are integer years anyway, and because of how pclimate works.Ed Bueler3+43-35
2011-02-15 20:46Maintenance on EISMINT-Ross. Now rossplot.py will show results properly at any resolution. Try './quickstart.sh 2 15 80' and './quickstart.sh 2 80 15' to see that results are rather strongly *in*dependent of grid resolution. Looking at the results I think the largest errors come from principal strain direction being perpendicular to artificial, straight calving front. This means we should see improvement with Torsten/PIK-based mods of calving front stress direction mechanism, I think.Ed Bueler2+36-31
2011-02-15 19:00Minor edits on front pages of Installation Manual. Debian: I love it\!Ed Bueler1+11-13
2011-02-15 18:32Re-factored IceModelVec3 and updated the Installation Manual.Constantine Khroulev5+293-374
2011-02-14 12:24Updated regression test 20 (test J) plus minor doc. fixes.Constantine Khroulev3+27-28
2011-02-14 12:24Updating documentation and fixing up little things.Constantine Khroulev38+2129-259
2011-02-14 10:36Fixed the shift issue in test I and corrected spelling in ssa_test?.cc.Constantine Khroulev3+11-20
2011-02-13 18:01Steps to make Jed's SSA FEM code work in the new stressbalance framework: Added (not yet properly doxygenized) comments. Fixed bugs related to element indexing, grid periodicity, and Dirichlet conditions. Added framework for SSA test case drivers, drivers for tests I and J. Test case J functions, case I still in progress....David Maxwell1+8-1
2011-02-13 18:00Steps to make Jed's SSA FEM code work in the new stressbalance framework: Added (not yet properly doxygenized) comments. Fixed bugs related to element indexing, grid periodicity, and Dirichlet conditions. Added framework for SSA test case drivers, drivers for tests I and J. Test case J functions, case I still in progress....David Maxwell1+136-0
2011-02-13 18:00Steps to make Jed's SSA FEM code work in the new stressbalance framework: Added (not yet properly doxygenized) comments. Fixed bugs related to element indexing, grid periodicity, and Dirichlet conditions. Added framework for SSA test case drivers, drivers for tests I and J. Test case J functions, case I still in progress....David Maxwell7+1293-184
2011-02-12 05:09Fixed bug #17709. And comments-only update to Test O code.Ed Bueler2+27-21
2011-02-10 22:49stdout_report() on SSAFD solve was not available in ssafd_test and pross; now it is. In pross: use -verbose 3 to see nonlinear (outer) iterations and use -ssa_rtol 1e-6 to tighten over default -ssa_rtol 1e-4, for example.Ed Bueler3+24-2
2011-02-07 05:09stdout appearance changes only.Ed Bueler2+7-7
2011-02-07 05:02SeaRISE-Greenland spinup now using annualized PDD; probably a small change but makes time-stepping fixed to integer frac of year (is this intended?). Trivial change to pst.sh, which still does not produce good results ... but no one is actually caring for now.Ed Bueler2+14-6
2011-02-06 04:53Added stencil width (WIDE_STENCIL=2) to vtauc; a FEM implementation of SSA will usually need some width, unlike the FD implementation. For now vtauc is not computed on the ghosted points, so before using the latest vtauc values do beginGhostComm(), endGhostComm() pair (from IceModelVec).Ed Bueler2+4-1
2011-02-06 02:41New very-simple basal melt rate verification test (Test O) hooked up. PISM does poorly ... next is revising enthalpy and temperature conservation of energy methods with EFGIS paper and Test O as tools.Ed Bueler8+191-62
2011-02-04 21:33Added, but not hooked-up, an utterly straightforward basal melt rate calculation as 'Test O'. See browser for exactTestO.c file. Changes to test K are trivial appearance only.Ed Bueler6+248-18
2011-02-03 19:29Cleanup of rarely-used PETSc-technicalities examples directory.Ed Bueler8+141-1268
2011-02-01 22:37Re task #6318: Decided to just do it. Now when the ice is floating the SIA 'diffusive flux' is *not* included in the mass continuity equation, even though the SIA flux is small there. Even when the SIA is on, the mass continuity equation on an ice shelf only sees the SSA contribution.Ed Bueler1+6-4
2011-02-01 22:06Added PISMSurfaceModel::ice_surface_liquid_water_fraction(). See task #7163.Constantine Khroulev7+39-10
2011-02-01 20:39Removed a "FIXME" in PISMBedSmoother: now it computes smoothed thickness as thk_smooth = (usurf - topg_smooth) for grounded areas only. For floating areas thk_smoothed = thk.Constantine Khroulev4+43-25
2011-02-01 19:26Added compile-time flags turning on writing NetCDF-4 output files (Pism_NETCDF4_OUTPUT) and NetCDF-4 variable compression (Pism_NETCDF4_COMPRESS).Constantine Khroulev5+128-23
2011-01-31 21:59Fixed bug #17617 (again). The previous fix (r1419) works too and this one is *not* more elegant, but current code avoids allocating T3 (which stays allocated until the end of the run) if PISM is running in the polythermal mode. (Temporary storage is used instead.) Also: now -init_from_temp and -init_from_temp_and_liqfrac work even if an input file does not contain the "enthalpy" variable, which might be the case if it is a "small" output file of a cold run.Constantine Khroulev11+146-114
2011-01-31 20:14Re-committed the r1410:1411 code change (a fixed version of it, that is). See bug #17615. I believe that this does *not* represent a change in PISM behavior (compared to r1410) if -pdd_annualize is omitted. We do need to check the "reasonableness" of results (e.g. SeaRISE-Greenland spinup results) before closing task #7296. (I agree that we need a PDD regression test, too.)Constantine Khroulev4+77-2
2011-01-31 18:47Fixed bug #17597.Constantine Khroulev3+28-9
2011-01-31 08:11Probably kludgish solution to bug #17617. Also, removed situation in tests where pismv is used to generate an input file. (Generally this is bad because pismv uses a different EnthalpyConverter than other runs. The situation is fine, and even appropriate, for verification. Warning about this in User Man is a good idea.)Ed Bueler8+31-16
2011-01-31 05:02Revision of flowlaw_test.cc and regression tests so that values of enthalpy are not exposed at regression. New tables have stress,temp,omega as inputs and put out a flow law coefficient. Thus EnthalpyConverter can have a different reference temperature (= temperature at which enthalpy is zero by definition) without breaking these regressions. Also delete ThermoGlenArrIceWarm regression (=test_31.sh) as unnecessary. Renamed test_32.sh -> test_22.sh for superficial numbering appearance.Ed Bueler9+187-235
2011-01-30 20:06Reversing r1410:1411 changes because I do not see how to fix them. (I.e. 'svn merge -r1411:1410 .') Note that I am trying to test enthalpy converter changes in the searise example.Ed Bueler3+2-63
2011-01-30 08:24Removed pism_config:water_melting_temperature and replaced with pism_config:water_triple_point_temperature. Value is unchanged at 273.15.Ed Bueler10+55-46
2011-01-30 06:41Editorial changes to bombproof/enthalpy docs.Ed Bueler2+14-51
2011-01-29 07:20Removed a FIXME: Made SSAStrengthExtension configurable by addition of new constants min_thickness_strength_extension_ssa and constant_nu_strength_extension_ssa. Cleaned-up its documentation.Ed Bueler8+73-51
2011-01-28 23:37A bit of code clean-up: IceGrid can and should contain arrays of grid point coordinates. This is cheap and allows getting coordinates of a point (i,j) as (grid.x[i], grid.y[j]), among other things.Constantine Khroulev6+98-52
2011-01-28 01:30* Started addressing task #7297: added mass_held_in_surface_layer() and surface_layer_thickness() methods to PISMSurfaceModel (both return 0 and are not in use). Also, tried to identify places in the code that will need to be changed to finish this task. Search for "FIXME task #7297" in src/.../*.cc. * Some minor clean-up.Constantine Khroulev16+140-112
2011-01-27 22:56Finished task #7296 (use "-surface pdd -pdd_annualize" to turn "annualization" on). The new mode limits time steps and requires hitting every year since -ys, i.e. "-ys 10.3" means that it will compute mass balance for the year (10.3,11.3) at the beginning of the run and use it until PISM hits 11.3 years, which is when PISM will compute acab for (11.3,12.3) and use that for a year, etc.Constantine Khroulev3+63-2
2011-01-27 20:23Added the -ebm_vars command-line option to PSExternal (specifies variables needed by an EBM as a comma-separated list).Constantine Khroulev2+38-14
2011-01-27 18:58* Replaced PetscEnd() with PISMEnd() throughout the code. This is necessary to make PISM exit cleanly when PETSc was initialized on a subset of all the allocated processors. * Finished working on coupling-through-files. Split the code: the base PSExternal class reads both acab and artm from a file created by an external model, PSExternal_ALR uses a lapse rate (option -artm_lapse_rate) to correct artm. * Moved the line "grid.year += dt / secpera" to the very end of the time-step. Previously when going from year 0 to year 1 boundary models were asked for conditions corresponding to the time-step (1,2) instead of (0,1).Constantine Khroulev50+433-296
2011-01-26 23:08More coupling-through-files code changes. Seems to work.Constantine Khroulev5+67-44
2011-01-26 19:19Fixed a couple more things in SSAFEM.cc.Constantine Khroulev1+96-57
2011-01-26 00:30* More work on the SSAFEM class; compiles, looks OK.Constantine Khroulev9+784-784
2011-01-25 20:15* Added a "fake EBM" script util/ebm. * Some more work on "coupling through files". Works in some cases, needs more testing. * Added an IceFlowLaw method computing both effective viscosity and its derivative with respect to the second invariant (by copying old code from r831). * SSAFEM.cc compiles. (Obviously needs more work, though.)Constantine Khroulev8+251-196
2011-01-25 01:49- fixed a typo in figure, updated script accordingly, and switched to pdf instead of png. - build directory must be cleaned to see changeAndy Aschwanden3+20-7
2011-01-24 22:15* Some more work on the coupling-through-files. * Clobbered some compiler errors in SSAFEM.cc. Needs more work.Constantine Khroulev7+285-142
2011-01-24 01:17Jed's FEM implementation from revision 831 is revived and de-PIMPLed. Does not yet compile; uncomment the line in src/base/stressbalance/CMakeLists to try.Ed Bueler3+1055-0
2011-01-23 09:02Separated the SSA as a ShallowStressBalance from its finite difference implementation. The separation is not perfect because, for example, some quantities are computed diagnostically (e.g. nuH) that would not necessarily be exposed/available in a different (e.g. SNES FE) implementation. But this version passes all regressions I can think of, including '(cd test/software && make)' and 'ssafd_test' output and the SeaRISE-Greenland first-use-of-full-physics after end of prespinup (see examples/searise-greenland).Ed Bueler8+935-815
2011-01-23 02:24Round of prettifying and documenting as I think about stress balances.Ed Bueler9+170-89
2011-01-20 20:18* Working on the two way coupling through files. * Minor clean-up, etc.Constantine Khroulev13+470-224
2011-01-19 07:16Added PISMVars *variables to PAYearlyCycle so derived classes of it can register their diagnostics. I don't think this is the right way to do it, or at least not in the right place in the hierarchy, but it works. Constantine?Ed Bueler2+4-1
2011-01-19 02:02Prettified tests so they fit on one line and have fewer colons, to facilitate 'visual grep'. That's all.Ed Bueler15+19-18
2011-01-18 02:36Fix to r1395 bug #17523. Minimal patch; clearly beddef in IceModel::init_diagnostics() is not always a valid pointer, but I am not sure if it should be made valid, etc.Ed Bueler1+3-1
2011-01-12 22:12* Updated howto.txt * Removed the SeaRISE-Greenland regression test. * Added some code to allow creating diagnostic quantities in surface, atmosphere, ocean and bed-deformation models. * Minor tweaks of the build system. * Removed many useless diagrams from the source code browser to reduce its size.Constantine Khroulev12+161-86
2011-01-11 23:34Finished working on packaging PISM (so far).Constantine Khroulev5+9-34
2011-01-11 22:56Finishing touches... more testing is in order, but I think we can produce usable Debian packages now.Constantine Khroulev3+4-3
2011-01-11 21:33Some more tweaks...Constantine Khroulev3+9-8
2011-01-11 21:10More work on the package; next step: testing.Constantine Khroulev3+8-14
2011-01-11 20:31Minor tweaks of the build system.Constantine Khroulev2+46-62
2011-01-10 23:29Some more work on the Debian package.Constantine Khroulev5+97-21
2011-01-07 23:32Added a paragraph about managing I/O in classes derived from PISMComponent.Constantine Khroulev1+28-0
2011-01-07 21:42Very small edits to very useful new 'related' page.Ed Bueler1+23-19
2011-01-07 21:32Re-created the pism-toolkit package "the right way"; some more work on the "pism" package.Constantine Khroulev19+81-51
2011-01-07 21:16Added doc/browser/howto.txt (as an attempt to preemptively answer some questions about extending PISM).Constantine Khroulev3+231-0
2011-01-05 22:35More work on the .deb package.Constantine Khroulev8+94-2
2011-01-04 17:19* Source code tree re-organization; I hope this will reduce the clutter a little bit. * Added the "debian" directory with the files necessary to create the pism-toolkit meta-package.Constantine Khroulev80+3086-3033
2010-12-20 19:29Mods to pismo error messages and std out, and replace with IceModelVec2Mask. Update to spinup for SeaRISE-Greenland; what do you think?Ed Bueler3+41-92
2010-12-17 20:42* Completed task #6839 by adding a -regrid_bed_special to our Friday menu.Constantine Khroulev5+153-0
2010-12-16 19:58* Added CMake switch adding compiler flags necessary for profiling with gprof. * Completed task #7271.Constantine Khroulev5+35-22
2010-12-14 19:25Switched to computing a time-step-average of artm and acab in PSDirectForcing (instead of a mid-time-step snapshot).Constantine Khroulev3+22-4
2010-12-13 22:20Finished implementing the "surface model" reading B.C.s from a file.Constantine Khroulev3+33-6
2010-12-10 00:41* Fixed PISMDiagnostic::set_attrs() (now it does not reset attributes to "" if an empty string is given as an argument.) * Added a preliminary implementation of PSDirectForcing, a surface model reading time-dependent top-surface B.C. from a file.Constantine Khroulev4+246-4
2010-12-10 00:20Fixed the no_model_mask metadata to conform to other PISM mask style; thus it has no units but it has flag_{values,meanings} attributes. Added msg to stdout explaining that double initialization messages appearing (at stdout) are intrinsic to delete of old stress balance.Ed Bueler1+23-6
2010-12-09 21:50Added a CMake switch to change variable storage order in an output file (t,x,y,z versus t,y,x,z versus t,z,y,x).Constantine Khroulev3+36-0
2010-12-09 19:55Fixed bug #17279.Constantine Khroulev4+15-13
2010-12-09 05:52Update to bmr_enhance_scale based on playing with it in Jakobshavn context. Unsuccessful attempt to add units of '1' to schoofs_theta, which is conforming in cf 1.4 and which is the only thing that makes compliance checker fail.Ed Bueler4+7-6
2010-12-09 01:00Fixed some pedantic compiler warnings; added more profiling events.Constantine Khroulev7+29-7
2010-12-08 23:40* Added a config and command-line switch to turn "off" the SIA stress balance (-no_sia to turn "off", -sia (the default) to turn it "on"). * Added more profiling events. * Fixed the computation of the vertically-averaged velocity field (changed the threshold used to determine if a cell is "ice-free").Constantine Khroulev9+67-22
2010-12-08 22:58Added a profiling event counting time spent in the SSA update; fixed SSAFD (it was not saving the initial guess).Constantine Khroulev2+14-0
2010-12-08 22:27More work on sub-models and I/O; I think I'm done with this.Constantine Khroulev17+209-100
2010-12-08 22:09pismo gives correct stdout msgs re presence of either option -no_model_strip or input variable no_model_maskEd Bueler1+10-0
2010-12-08 06:44Because executable pismo has no well-defined semantics without it, it now *requires* option -no_model_strip to even run. It stops if the option is not given.Ed Bueler2+15-6
2010-12-08 02:02Added initialization messages to SIAFD, SSAFD and regional versions of these classes. Resurrected stdout reporting of stress balance computations.Constantine Khroulev4+14-2
2010-12-08 01:42Fixed a sign error in SSAFD::compute_basal_frictional_heating() (plus some code changes in boundary models).Constantine Khroulev6+75-7
2010-12-08 00:39Fixed a typo.Constantine Khroulev1+1-1
2010-12-08 00:34Fixed NCSpatialVariable::define() (now it does not try to define a variable if it exists already).Constantine Khroulev2+14-2
2010-12-08 00:09* Yet another change to PISMVars: keys() gives a set of short names, not standard_names if present, short otherwise. * Added taud_mag diagnostic quantity.Constantine Khroulev7+151-41
2010-12-07 22:25* Fixed pismo.cc (SIAFD_Regional and SSAFD_Regional were not "on"). * Cleaning up PISMComponents and boundary models (the way they write diagnostics, mostly).Constantine Khroulev19+273-293
2010-12-06 21:33Some clean-up; fixed a bug in NCTool (::close() needs to reset def_mode to "false".)Constantine Khroulev6+18-23
2010-12-06 19:46Added some more code to address task #7265.Constantine Khroulev5+90-55
2010-12-04 01:12* Finished the reorganization of PISM's spatial diagnostic computations. * Fixed the IceModel::write_variables(): now it defines all the variables prior to writing them. (Writing -o files should take a lot less time now, especially when Mx, My, Mz and Mbz are large.) Also: this change fixes a problem that prevented PISM from switching to writing NetCDF-4 files. * Map-plane and sounding viewers are back. Removed "slice" and "surface" viewers: temperature at the surface is available as the "tempsurf" map-plane field, etc. * Minor code changes/reorganizations; fixed a memory leak.Constantine Khroulev20+770-749
2010-12-03 00:59* Fixed a memory (de-)allocation issue. * Added more diagnostic quantities; I think we're back to the original set. * A step towards fixing run-time viewers.Constantine Khroulev6+349-118
2010-12-02 23:08Fixed the bug reported by Ed on Dec 1 (or at least I think I did).Constantine Khroulev1+7-2
2010-12-02 22:33More work on diagnostic outputs. Almost done... (most run-time viewers are broken, though).Constantine Khroulev16+1104-724
2010-12-01 21:53Doing the "midday commit" just to back-up the code. Lots of changes, all of them related to changing the way PISM handles diagnostic quantities.Constantine Khroulev50+1555-1015
2010-11-29 09:40Work on building jacobian. Not functional yet, but bodvardsson still works as well as before with -snes_fd and -snes_mf.Ed Bueler1+96-21
2010-11-29 03:11Bug fix so it works equally well in parallel.Ed Bueler2+119-52
2010-11-28 10:19Added ability to try reasonable initial conditions.Ed Bueler4+52-89
2010-11-27 23:58Added 2nd-order upwinding for SSA+mass continuity, and it works a charm.Ed Bueler4+53-23
2010-11-24 00:48* Removed IceROSSModel and the old pross executable: pross_new will replace both. Updated the EISMINT-Ross regression test. * Fixed some typos in the User's Manual. * Fixed a bug in climate (my fault; a conflict was not resolved correctly). * Added the "compute_grain_size_using_age" config parameter. * Moved pismo.cc and friends to src/regional. * Moved mapcoords(...) to IceGrid. * More work on the stress balance code. * Added IceModelVec::get_state_counter() and inc_state_counter(). These two will allow updating certain fields only if their dependencies were updated since the last computation. SIAFD::update() uses this. * Implemented the SIA sliding (verification test 'E') mechanism as a "ShallowStressBalance" derived class (see SIA_Sliding.{hh,cc}). Works; isn't used anywhere except in pismv. * Removed the old ssa_test executable. * Removed all the old stress balance code and replaced it with the new. All the software tests pass; more work remains to provide desired diagnostic quantities.Constantine Khroulev67+2383-5883
2010-11-18 21:43* Removed the interpolation mask from the IceModelVec regridding code. * Added a profiler to IceGrid (that way it is available pretty much everywhere in the code). * Fixed the pclimate issue (task #7248) by fixing PISMVars: methods like get_variables() are dangerous because iterating over pointers is a bad idea.Constantine Khroulev21+178-194
2010-11-18 19:08Squashed commit of the following:Andy Aschwanden1+11-8
2010-11-18 01:55pclimate should compile now. The parallel issue reported in task #7248 remains. Sorry about that. (Note there is no regression for pclimate so I missed the compile-time failure from inattention.)Ed Bueler1+3-0
2010-11-17 23:43Fixed part of task #7248, but new parallel issue revealed.Ed Bueler3+18-12
2010-11-17 21:49Updated SeaRISE-Greenland example in pism-dev to be v1.1. That is, I think stable0.4 should be mostly up to date with what we are doing at UAF.Ed Bueler2+6-4
2010-11-15 23:17Cleaned scaling with Jed's help. Recalled FD by coloring. (But I can't see much affect from it. Is it aware of dof=2?)Ed Bueler3+99-91
2010-11-15 21:59* Some minor code cleanup. * Further SSA code testing: implemented pross_new, an EISMINT-Ross example implementation (similar to pross) that does not use IceModel. * Moved some documentation comments out of iMssa.cc and iMsia.cc and into base/stressbalance/{SIA,SSA}FD.ccConstantine Khroulev13+883-732
2010-11-14 21:42First results from coupled mass-continuity+SSA PETSc/JFNK scheme using Bodvardsson for verification.Ed Bueler4+98-37
2010-11-14 10:31Viable. Shows convergence under grid refinement, but uses upwinding and does not yet have Picard or Jacobian.Ed Bueler1+127-61
2010-11-14 07:00Draft of combined, steady mass continuity plus SSA solver which should be verifiable by Bodvardsson soln. Pretty clearly has equation scaling issues for now.Ed Bueler5+414-44
2010-11-12 23:37* Reclaimed authorship of PISMStressBalance.hh. * Implemented siafd_test, an executable performing a diagnostic non-sliding SIA computation using verification test F initial data. Shows convergence on par with "pismv -test F -y 0", see below.Constantine Khroulev9+601-42
2010-11-12 07:45Corrected and added refs.Ed Bueler1+27-4
2010-11-12 00:43Updated, good-looking values and interpretation for Test N.Ed Bueler3+33-29
2010-11-11 22:03ssafd_test is an example of using the SSA solver without IceModel. Uses test J; gives a pismv-style error report.Constantine Khroulev4+248-105
2010-11-11 21:50Test N is reinterpreted Bodvardsson (1955) solution. Significance of this solution, and use as a verification, will follow eventually.Ed Bueler3+121-102
2010-11-11 19:14* More work on the stress balance code. Added ssafd_test.cc (does not quite work, but serves as an example). * Minor code changes in the IceModelVec code. * Added the code writing the SSA ssytem to the ssa_test executable.Constantine Khroulev10+383-39
2010-11-10 21:54Sped up test 14 so that it is not at all the slowest one.Ed Bueler1+2-2
2010-11-10 21:36I am cleaning up old code which I wrote, which no one else has ever modified, and which I think is either stupid or totally superceded by Constantine's recent work.Ed Bueler10+25-595
2010-11-10 18:43* Added more configuration flags: "write_ssa_system_to_matlab" and "shelves_drag_too". * Modified the top-level Makefile so that one can run "PISM_STATIC=1 make" (or "make PISM_STATIC=1") to build PISM with static libraries. This is a temporary solution; should be removed once building with CMake is well-documented. * Added a new option to the CMake PISM build system: Pism_ADD_FPIC to add "-fPIC" to C++ flags. (This is needed on pacman and might be needed on other machines requiring static libraries.) * More stress balance code... time to write software tests.Constantine Khroulev9+553-30
2010-11-08 20:08* Minor code changes to get rid of pedantic compiler warnings. * Added CMake options: Pism_DEBUG: defines PISM_DEBUG (to enable debugging checks in the code) and Pism_ADD_FPIC: adds "-fPIC" to C++ compiler flags. * More stress balance work: all of the SIA code is in, as well as most of the SSA...Constantine Khroulev27+650-219
2010-11-07 07:56Added old reference: Nye on plastic reconstruction.Ed Bueler1+11-0
2010-11-05 06:42An include line needed to build on my laptop. Not so on my 8core. But it does make sense that SSAFD.hh should include petscksp, yes?Ed Bueler1+1-0
2010-11-05 00:28* More stress balance code... still incomplete. What is committed compiles (but would not run yet). * Fixed some typos in the User's Manual. * Some code clean-up. * Software tests now print the directory "pismr" was found in, not the one tests are located in (to avoid confusion about what PISM we're testing).Constantine Khroulev26+754-163
2010-11-04 22:31Spelling fixed.Ed Bueler1+2-2
2010-11-03 21:38* Fixed a bug: -plastic_phi option had no effect. * More stress balance code...Constantine Khroulev7+101-19
2010-11-03 20:34Removing empty bin/ and lib/ directories from version control. Completes task #7235.Ed Bueler0+0-0
2010-11-02 23:48Added comments to util/pism_[matlab.m,python.py] to show usage including minimal pismr call which will successfully bootstrap. Fixed examples/pddtune/linesearch.py to be a bit more flexible; no intended change in function.Ed Bueler3+48-13
2010-11-02 23:40Working on the stress balance code...Constantine Khroulev7+307-83
2010-11-02 20:10* Fixed a bug in FindPETSc.cmake. * Added a build option Pism_LOCAL_INSTALL: if "ON" (the default), PISM is installed in prefix/bin, prefix/lib and prefix/doc. Otherwise in prefix/bin/pism, prefix/lib/pism and prefix/share/doc/pism (to avoid cluttering the system, think prefix==/usr/local). * More work on the stress balance code.Constantine Khroulev21+825-118
2010-11-02 17:30Changes only to docs and stdout msgs in PISMSurface.cc. Changes to MISMIP were in an attempt to enforce the MISMIP stopping criteria more literally, but with no real improvement.Ed Bueler4+78-72
2010-11-01 19:16- added command line option "-e_age_coupling" which ties enhancement factor to the age of ice (age calculation is automatically turned on if this option is seen). Set e=1 if age<11'000 years, e=3 otherwise. See e.g. \ref Greve for a justification.Andy Aschwanden2+27-2
2010-10-31 21:59Improvements, including better line search, ability to reproduce a file from its name, and better docs.Ed Bueler4+85-53
2010-10-30 08:04Implemented a lapse rate, with latitude, for the pdd_std_dev quantity. Not well-evaluated yet.Ed Bueler7+247-164
2010-10-29 21:36Added some (re-factored) stress balance code. Only interfaces so far, still lots of work to do. Also see .pdf figures for an idea about the general structure.Constantine Khroulev12+422-10
2010-10-29 02:31linesearch.py replaces runcase.sh and weights.py. It still does not do line-searching (for zero smb difference) but that can be implemented easily now.Ed Bueler6+146-240
2010-10-29 02:27Changed 'snowprecip'->'precip', and removed '-pdd_fausto', so EISMINT-Greenland PDD parameters are in use. FIXME: Add tuned params to this example once done. Also: Simplified preprocessed data file.Ed Bueler2+4-7
2010-10-28 23:58* Switched to the thickness-relative-to-the-smoothed-bed in IceModel::massContExplicitStep(). Let's see if that makes a difference... * Added Andy's pism_python.py so that we have an example of using Python to prepare a bootstrapping file.Constantine Khroulev2+114-9
2010-10-28 21:12* Added bueler_brown_f (a diagnostic quantity). * NetCDF attributes are written only once now, when a variable is defined. * Added a IceModelVec2Stag method averaging a vector field onto a regular grid. (staggered_to_regular(); uses different offsets for u and v components). * Added magnitude() and magnitude_squared() methods to PISMVector2. Now one can do thing like "vel_ssa(i,j).magnitude_squared()" * Isolated the computation of bueler_brown_f(). * Now we flush time-series when an automatic backup is written. * A bunch of SIA-related code. Just separating ideas; this "new" code is not in use and most of it never will be.Constantine Khroulev10+576-86
2010-10-28 20:53Draft of line search which will replace a long list in dotune.sh. I think the 'special sauce' is a longitudinal lapse rate for the pdd_std_dev variable, to parameterize the variability-of-climate that seems to come into the temperature fields.Ed Bueler3+167-119
2010-10-28 07:10Searches are a bit simpler because I have tied the snow and ice DDFs: DDFice = 2.0 * DDFsnow.Ed Bueler4+143-31
2010-10-28 01:49fixed a bug in script. Should be back-ported to stable 0.3 too, I believe.Andy Aschwanden1+4-2
2010-10-26 18:47Revised interpretation of DegreeDayFactors.refreezeFrac: if the PDD sum is not sufficient to melt all the snow, the *all* the snow refreezes (conceptually as ice lenses within snow), so the refreezeFrac is only active in locations where all of the snow has melted. This presumably corrects a misunderstanding (by me) of the meaning of refreeze fraction. In any case it is very compatible with the runoff pattern observed in Ettema et al data.Ed Bueler1+19-14
2010-10-26 18:38Is this the correct brief install instructions now that we have cmake? I.e. is it correct? *and* is there a better thing to tell users?Ed Bueler1+22-15
2010-10-26 02:03Yes, this commit fixes one line in a readme. But it has already caused a user some trouble ...Ed Bueler1+1-1
2010-10-25 18:03* Re-wrote software test #32 (it uses a relative tolerance of 1% now). * Split IceModel::surfaceGradientSIA() into three methods implementing eta-transformation, Haseloff and Mahaffy approaches. * Some more work on the automatic backup code: it should always write "-o_size small"-style files. Also, we don't need two time-stamps in the history string.Constantine Khroulev6+270-184
2010-10-24 18:03 netCDF3 and netCDF4 module seem to behave slightly differently, causing objective.py not to work on my mac with netCDF4 module. I believe my fix should work also under netCDF3, but let me know if not.Andy Aschwanden1+3-3
2010-10-23 08:35Minor stuff, but squashes compiler warning on pacman.Ed Bueler3+20-13
2010-10-23 01:30Tuning all set up. That is, given the current uncertainty as to what is the objective (the function ...), the scripts are all in place to try out various schemes and objectives fairly quickly. See README for a recipe.Ed Bueler10+423-86
2010-10-22 20:01Finished working on the calculation of the basal temperate ice layer thickness; now we have two fields: tempicethk (total amount of temperate ice in the column, meters) and tempicethk_basal (basal layer only, uses linear interpolation for more precision).Constantine Khroulev3+29-24
2010-10-22 03:16Squashed commit of the following:Andy Aschwanden1+1-1
2010-10-22 02:36It turns out substantial simplifications are possible, once I change myself to being less dumb.Ed Bueler4+26-129
2010-10-22 00:40Added a new method of computing the thickness of the basal layer of temperate ice.Constantine Khroulev3+91-3
2010-10-21 23:05* Added a script that runs a subset of software tests. Try "cd test/software && ./run_test.sh 29 32" to run test_32.sh and test_29.sh in test/software/pismConstantine Khroulev26+435-317
2010-10-21 18:39O.k. I have a PDD parameters tuning result. Need to show to Regine.Ed Bueler6+94-67
2010-10-21 17:09Squashed commit of the following:Andy Aschwanden2+6-6
2010-10-21 07:58Progress on PDD parameter tuning. The script dotune.sh can be run to see the results of controlling the PDD parameters; the actual objective evaluation script objective.py is not seriously started, but a placeholder is there.Ed Bueler5+141-24
2010-10-21 07:56Removed 'snow' from 'snowprecip' name, throughout PISM.Ed Bueler9+84-63
2010-10-21 02:35Work on PDD parameter tuning tools. Things don't run so don't bother trying.Ed Bueler6+176-166
2010-10-20 22:17Renovated PAForcing class: * removed the mean annual semantics, so that anomaly file simply provides time-dependent 2m air temperature and precipitation differences, and when we need mean annual quantities we run code that computes annual averages as needed * removed the 'snowprecip' semantics; with r1280 we now treat precipitation as precipitation and we use air temperature to determine if it is rain (and if it is rain it does not contribute to surface mass balance ... but all of this is appropriate to handle in the PISMSurfaceModel) * removed 'ma_' from names like '-anomaly_temp_ma' and 'temp_ma_anomaly' * removed 'snow' from names like 'snowprecip'Ed Bueler7+182-160
2010-10-19 01:57Cosmetic change, but requires thought and then a bit more coding to fix FIXMEs.Ed Bueler5+94-48
2010-10-17 22:21* Added the final fix for the add_vars_to_output() code. * Removed unnecessary code from pclimate.cc (NCConfigVariable::write(filename) opens and closes 'filename'.)Constantine Khroulev2+3-4
2010-10-16 02:08More cleaning/clarifying. Only added functionality is new config parameter pdd_positive_threshold_temp, which defaults to 273.15 but could be 268 if we believe van de Broeke et al 2010.Ed Bueler8+260-202
2010-10-16 00:43To pclimate, added ability to write pism_overrides variable into output file. Made pclimate a bit more verbose than other PISM executables, w.r.t. pism_overrides. Constantine: did I do this right? Please fix if not.Ed Bueler3+40-24
2010-10-15 22:07Major cleaning and clarifying (I hope) of PDD code, but without changing the results or actually adding functionality. Main accomplishment: un-spaghettied the Fausto/Greve choices of degree day factors from the core PDD scheme functionality. Also, I documented the meaning of recently-added PISM model outputs acab, saccum, smelt, srunoff.Ed Bueler6+410-242
2010-10-15 20:53Fixed a stupid bug in "void add_vars_to_output(...)" methods of some couplers.Constantine Khroulev6+25-17
2010-10-14 20:58Various little fixes, including climmask.py.Constantine Khroulev10+58-29
2010-10-14 03:41New thread. My plan is to use the Ettema precip (=snowprecip) and runoff fields, and the Fausto temp parameterization, to run a quick sequence of pclimate runs to attempt to find values of PDD parameters which yield close agreement with the Ettema smb field. It would be helpful to have the Ettema melt field, but I don't have it. (The others just mentioned are in Greenland_5km_v1.1.nc already.) This is in preparation for RCP runs. For now my standard for agreement will be agreement of srunoff resulting from 'pclimate -atmosphere searise_greenland -surface pdd -pdd_fausto' with the runoff field, e.g. using a least-squares approach. Of course I will use the -config_override mechanism to change parameters as described in that section of the User's Manual. There will be an actual tuning script once I get a first run working. I seek help and advice.Ed Bueler3+251-0
2010-10-13 00:31Cleaned up semantics and ghost computations for no_model_mask. Better wvelsurf results for sure.Ed Bueler1+23-23
2010-10-13 00:24Added the "automatic backup" feature: running PISM with -o foo.nc writes foo_backup.nc every hour. The backup interval can be changed using the -backup_interval option or the backup_interval config. parameter (units: hours).Constantine Khroulev5+92-0
2010-10-12 22:38* More work on the coupling-through-files code; added examples/coupling/psexternal.py. Runs, but does not work right yet. * Added a mechanism allowing boundary models write more if -o_size (or -save_size) is given. PSForceThickness writes ftt_modified_acab with -o_size big. * Now PSForceThickness writes ftt_mask only if it was present in the -force_to_thk file. * Moved boundary models to a separate PISM library (see src/CMakeLists.txt).Constantine Khroulev13+381-93
2010-10-12 21:15* Fixed unreported bug in -force_to_thk mechanism: now surface mass balance is *actually* the combination of the original and a part proportional to the difference of thickness from target thickness.Ed Bueler5+44-46
2010-10-12 07:10Attempt to add new variable ftt_modified_acab to written fields when using PSForceToThickness. I'm missing something; need to talk to Constantine. See FIXMEs in PSForceToThickness write methods.Ed Bueler2+58-14
2010-10-12 01:44Despite the large number of lines changed, this is a documentation and cosmetic update only: (1) Changed PSLocalMassBalance to PSTemperatureIndex, because of name clash with LocalMassBalance class. (2) Improved source code (doxygen) documentation of PISMSurfaceModel classes. (3) Improved run-time stdout reporting on same. (4) Added 'Making HARD Modeling Choices' section to User's Manual, and wrote first page of same.Ed Bueler8+358-250
2010-10-11 16:43* Some clean-up: typos, etc. * Fixed the code I committed earlier today. (Everything compiles now.) * Switched from calendar = "none" to "365_day" in the metadata for the time dimension in PISM output files. * Added scripts in util/ to the list of files installed by "make install".Constantine Khroulev10+56-26
2010-10-11 15:42Work-in-progress; some code may not even compile.Constantine Khroulev4+323-2
2010-10-07 08:14Added new executable 'pismo', for 'PISM outlet glacier mode', with new mask 'no_model_mask' with following semantics: In a strip around the edge of the computational domain, overwrite computed surface gradient and driving stress; zero them out in that strip. pismo.cc contains derived class IceRegionalModel : public IceModel. See option -no_model_strip to specify width at bootstrapping. Note FIXMEs related to ghosts in derived surfaceGradientSIA(). Need discussion Dani+Constantine+Andy+me on Friday. Will commit UAF-regional/run.sh mods which use it; Works. I love CMake.Ed Bueler3+311-3
2010-10-06 23:50Added write of ftt_mask to PSForceThickness, so that the mask used in Dani's regional model is carried forward for longer runs.Ed Bueler3+37-11
2010-10-01 23:47Minor stuff. In writing up my annual report I again counted lines in PISM: about 44K lines of source code, I think.Ed Bueler2+9-8
2010-10-01 22:25Merged 1258:1259 changes from stable0.3.Constantine Khroulev2+8-6
2010-10-01 22:05* Removed config/... * Some more work on packaging and related issues.Constantine Khroulev4+15-53
2010-10-01 19:14Minor fixes; some work on the debian package.Constantine Khroulev6+50-13
2010-09-29 22:17* Removed the old (written-by-hand-Makefile-based) build system. (PISM docs can still be built by running "cd doc && make userman" and similar.) * Note that the "build" directory can be used with CMake; running "make install" in the top-level directory attempts to configure, build and install PISM in place (similar to running "make" before this commit). * Re-organized the source code browser and added it to the CMake build system. Not very pretty (ahem). Oh well. * Replaced PISM_PREFIX with PISM_INSTALL_PREFIX to avoid confusion. * Minor cleanup.Constantine Khroulev25+2169-2378
2010-09-29 18:58- updated CMakeLists.txt to make git working - added Hock & Holmgren (2005) referenceAndy Aschwanden3+13-5
2010-09-28 18:32* Changed PDDMassBalance::getMassFluxFromTemperatureTimeSeries() to produce accumulation, melt and runoff. * Added "extra" variables "saccum" (accumulation), "smelt" (melt) and "srunoff" (runoff). See task #7128 and try "pclimate -surface pdd -atmosphere searise_greenland". * Some minor updates and fixes.Constantine Khroulev9+195-116
2010-09-27 20:51* Removed old flow law code. Replaced the old flowlaw_test with a new version. * Added flowlaw_test, bedrough_test and ssa_test to the list of executables built by default. * Added the -r command-line option to nccmp.py (switches to using the relative tolerance). * Fixed a bug: Enth3 was not initialized at the beginning of a cold-mode run (my fault).Constantine Khroulev22+93-1464
2010-09-27 01:28There is no change to running code in this commit, only comments. The comments in localMassBalance.cc suggest formulas to implement the melt and runoff aspects of task #7128. Presumably new fields--suggested names smelt,srunoff--would by written by a call to PISMSurfaceModel::write_fields().Ed Bueler3+38-9
2010-09-27 01:23Change of names of verification tests to remove double 'with', and to clarify if SIA or SSA. And removal of historical comment in iMgeometry.cc.Ed Bueler2+15-16
2010-09-24 22:19Bit of work on a new manufactured flowline solution of SSA plus mass continuity.Ed Bueler4+233-2
2010-09-24 21:05More work on flow laws; added flowlaw_test_new to compare old and new flow laws.Constantine Khroulev8+498-19
2010-09-24 19:32* More software tests (flow law regression). * Removed the unused (and misnamed) method IceFlowLaw::effectiveViscosityColumn().Constantine Khroulev9+560-202
2010-09-23 20:30Added forgotten src/base/flowlaws.{cc,hh}.Constantine Khroulev2+371-0
2010-09-23 18:56* Some work on flow laws. * Fixed the CMake building instructions in the Installation Manual. Also: running "make" in the build directory places executables right there. * Added a couple lines to allow specifying the install location by running "PISM_PREFIX=/some/path cmake .." in the build directory (this is easier than "cmake -DCMAKE_INSTALL_PREFIX=/some/path ..").Constantine Khroulev6+27-9
2010-09-22 21:01Lots of small (cosmetic) changes. Getting ready to re-factor flow laws (once more)...Constantine Khroulev20+155-98
2010-09-22 16:34Replaced -boot_from with -boot_file everywhere in the PISM source tree so that all command-line options (except -i and -o) taking a file name have the _file suffix. Added a check for -boot_from and a error message.Constantine Khroulev39+115-112
2010-09-21 20:53* Updated the software test #3 and added a new software test (#27, comparing initialization from a PISM output file using "-i" to initialization from the same file with "-boot_from ... -regrid_file ...". * Documented building PISM using CMake and installing PISM on Mac OS X + Fink (see task #7132). * Added the (forgotten) doc/browser/Makefile. * Fixed a non-bug (see bug #16307) by making enthalpy "diagnostic" when -cold is given. This way PISM warns if -regrid_vars enthalpy is given with -cold. (The problem was that "pismr -cold -regrid_file foo.nc -regrid_vars enthalpy" does NOT bring in the energy state from foo.nc, because energy balance computations are done using temperature.) Added both enthalpy and temperature to the output_medium list. * Fixed an unreported bug: "pismr -boot_from foo.nc" did not set time to the value found in foo.nc. * Fixed a typo in src/base/iMIO.cc * Fixed an unreported bug (test #27 without bedrock failed).Constantine Khroulev10+164-50
2010-09-20 21:03* Re-organized the doc/ subdirectory. * Improved util/flowline.py.Constantine Khroulev22+1966-1929
2010-09-17 20:02Squashed commit of the following:Andy Aschwanden1+1-1
2010-09-16 23:23* util/flowline.py should be usable now (see task #7133). * Minor fixes, including pedantic compiler warnings and an update of FindPETSc.cmake.Constantine Khroulev3+119-30
2010-09-15 23:43* Updated CMake/*.cmake using versions from Jed's github repository. * Resurrected the CMake build system. (Hmm. One could call this going back to task #6521.) I like it! * Clobbered some compiler warnings. * Added the -periodicity command-line option (task #7133).Constantine Khroulev33+386-252
2010-09-13 20:57* Fixed a nasty bug that made pism? -age crash. * Added util/flowline.py (unfinished, see task #7133).Constantine Khroulev3+123-2
2010-09-12 02:53Another reference.Ed Bueler1+10-0
2010-09-09 23:13Squashed commit of the following:Andy Aschwanden3+31-21
2010-09-09 22:56* Fixed the Cheat Sheet (-split_snapshots -> -save_split). * Updated eisgreen.py and fill_missing.py so that they work with the current version of netcdf4-python.Constantine Khroulev3+3-4
2010-09-09 22:42The initial one year run set a bedrock thermal layer thickness of 2000 m *but* the script (bootstrap.sh) did not specify the number of levels in the bed, Mbz; the default number was therefore used, namely Mbz=1. Note that the EISMINT-Greenland description PDF (http://homepages.vub.ac.be/~phuybrec/eismint/green-descr.pdf) gives values of the bedrock thermal conductivity and heat capacity, so we may assume that, in EISMINT, a bedrock thermal layer is called for if the model can do it. So now I have set -Mbz 21.Ed Bueler1+1-1
2010-09-09 22:29Fixed message reporting on initialization of bed smoother object.Ed Bueler1+5-5
2010-09-09 19:17Merged 1224:1225 changes from stable0.3 to trunk.Constantine Khroulev5+10-10
2010-09-09 18:25Very minor updates.Constantine Khroulev2+12-12
2010-09-09 06:28A couple of references.Ed Bueler1+24-0
2010-09-08 21:20Some work on the code restricting the force-to-thickness mechanism to a subset of the model domain.Constantine Khroulev2+22-2
2010-08-30 20:42Changed calculation of precipitation-which-fell-as-snow to use N-1 not N subintervals, to avoid overcounting. (I.e. to fix off-by-one error.) Also lengthened long names of variables written by PAYearlyCycle and PAForcing; note that with options '-atmosphere searise_greenland,forcing -anomaly_temp_ma foo.nc' we get five different temperature fields; they are justified but disambiguation is needed.Ed Bueler4+14-14
2010-08-30 20:18modified (and simplified PDDMassBalance::getMassFluxFromTemperatureTimeSeries()Andy Aschwanden2+20-22
2010-08-30 19:01Modified/improved comments to PDDMassBalance::getMassFluxFromTemperatureTimeSeries(). Cleaned computation in same method, to clarify and reduce number of operations.Ed Bueler4+65-55
2010-08-30 18:14finished updating new accumulation calaculation, ready for testingAndy Aschwanden1+14-13
2010-08-30 04:58Minor maintenance: docs, stdout messages, and addition of bwp to outputs of SeaRISE-Greenland.Ed Bueler5+22-16
2010-08-29 22:57 A first attempt to implement task #7127: compute accumulation from precip and temp:Andy Aschwanden2+36-4
2010-08-25 00:12 - forgot to update userman in previous commitAndy Aschwanden1+4-4
2010-08-25 00:08 - changed ivol{cold,temp} to ivol{cold,temp}f and iarea{cold,temp} to iarea{cold,temp}f, i.e. variables are now either volume or area fractions. I think this is more usefull than absolute values. Absolute values can still be computed, but have been removed from pism_config.cdl:pism_config:ts_variables.Andy Aschwanden3+86-1
2010-08-17 22:27Added Ockendon ref.Ed Bueler1+10-2
2010-08-12 22:08Knoll & Keyes added.Ed Bueler1+11-0
2010-08-10 18:33Added acab to pism_config:output_medium. Changed pism_intent attribute of acab to be honest: IceModel::acab does not know its pism_intent, but it is the one that is written to file. The pism_intent attribute of PSLocalMassBalance::acab is now 'diagnostic' and for PSConstant::acab it remains 'climate_state'. Is this right and can it be improved? Other very small changes to docs.Ed Bueler4+12-10
2010-08-01 08:29Added ssa_test.cc, a completely IceModel-free SSA solver. It reads a PISM output file and solves the SSA diagnostically. It uses the surface elevation, ice thickness, bed elevation, mask, and yield stress tau_c fields only. For now it uses constant hardness and does not read either temperature or enthalpy. For now it only uses an X PetscViewer to show computed velocity, but that could be updated. Does not read stored uvbar or other mechanism to impose Dirichlet boundary conditions. Envisioned as playground for: (1) trying hydrostatic on same geometries, (2) studying effect of SSA inversion, or (3) regression-testing new StressBalance object, when available. Completely subject to revision and play. Currently, ssa_test.cc is not the basis of a regression test.Ed Bueler2+807-2
2010-08-01 07:29SIA code was not setting diffusivity in H==0 areas. Fixed.Ed Bueler1+2-1
2010-08-01 07:27I cannot find the implementation of IceModelVec2V(const IceModelVec2V &original), and it (finally?) caused a link-time error for me. Huh? I have commented out its prototype.Ed Bueler1+1-1
2010-07-28 08:23Fixed diagnostic diffusivity (of SIA) calculation: now it happens only in one place, namely velocitySIAStaggered(). We save diffusivity on staggered grid, in memory also used for 'hardav' in SSA; this is now IceModelVec2Stag vWork2dStag. Also a bit more reporting on bed smoother initialization.Ed Bueler8+85-81
2010-07-28 01:04Added and documented option -bed_smoother_range. Added slightly to performance stats put in history string.Ed Bueler3+30-13
2010-07-27 06:45Generally bed smoother/roughness-parameterizer ready to go. Thickness details (margin) fiddles done. Derived classes turn the mechanism off. Half-width of smoothing domain set to 5km, but not that there is always one grid point in each direction if the mechanism is on at all. Remains to add option adjusting this range.Ed Bueler8+150-30
2010-07-26 20:10Important fix inside velocitySIAStaggered() so that theta is correctly on the staggered grid point.Ed Bueler2+5-9
2010-07-26 16:22Improved bed roughness parameterization scheme by recognizing 4th order Taylor polynomial was convex like the function it is approximating. Updated tools to reflect that.Ed Bueler7+91-88
2010-07-26 00:46Clean up of r1200 commit: revert changes to iceModelVec2.cc (why can't global go to proc0?); fix bedrough_test.cc so test 26 passes; clobber 'possibly uninitialized poly_ice' compile warning for flowlaw_test.cc. Test 22 still fails, but this failure un-observable by eye.Ed Bueler4+15-13
2010-07-25 23:50Work on bed roughness parameterizer. Fails a couple of tests even when turned off.Ed Bueler12+266-165
2010-07-24 02:28Working PISMBedSmoother, still not attached to IceModel (that is next). Added min() and max() capability to IceModelVec2S, but have not used it in many possible applications in IceModel.Ed Bueler10+298-151
2010-07-23 23:44bed roughness parameterization test program runs minimally, but PISMBedSmoother is not attached to IceModelEd Bueler2+47-9
2010-07-23 17:17Added flowlaw_test, an executable similar to flowTable. flowlaw_test accepts the -ice_type command-line option but does not compute GK parts. It is used by the flow law regression tests 24 and 25.Constantine Khroulev4+209-3
2010-07-22 22:30Initial implementation of bed smoother. Compiles, but PISMBedSmoother is not known to IceModel, and test program trybedrough.cc not complete.Ed Bueler7+303-68
2010-07-22 08:35Work on implementing Schoof (2003) strategy. New C++ code compiles but is incomplete. docs are pretty complete.Ed Bueler4+334-12
2010-07-21 21:34More work related to bed roughness, and the addition of option -paleo_precip into spinup script for SeaRISE Greenland, as it was used.Ed Bueler3+48-20
2010-07-20 02:31Cleanup and uniformity of names in src/base/ directory.Ed Bueler16+2-2
2010-07-20 02:25More tools and text related to implementing Schoof (2003) bed roughness parameterization technique. The Matlab/Octave codes in src/base/bedroughplay/ come from Ed's Karthaus/McCarthy pedagogical codes.Ed Bueler12+673-31
2010-07-15 02:43a bit more docs re bed smoothing Schoof (2003) techniqueEd Bueler2+47-7
2010-07-14 23:34Modest start on bed roughness improvements related to Schoof (2003).Ed Bueler3+82-30
2010-06-29 22:19Testing on marmaduke reveal that r1187 produces small differences on different machines for tests 22 and 23; may be libraries? In any case, tolerances lowered to generally 10^-7 with several exceptions down to 10^-5.Ed Bueler5+62-46
2010-06-29 22:12Squashed commit of the following:Andy Aschwanden4+21-8
2010-06-29 06:15Completely superficial changes to look of software tests. Stop me before I fiddle again ...Ed Bueler17+17-17
2010-06-29 06:07Superficial cleanup of comments and line wrap in materials.hh, iMssa.cc. Removal of dead/unnecessary code in materials.cc|hh; note that the HybridIce law *should* work; my bad for saying it should be disabled.Ed Bueler3+84-127
2010-06-29 03:35Added regression test for SeaRISE-Greenland. Adds only 1MB.Ed Bueler6+57-1
2010-06-29 01:52Removed regression software test using verification test B; it is redundant given tests 12 and 16. Re-arranged test numbers. Coming test 22 will be searise-greenland regression.Ed Bueler3+53-97
2010-06-29 01:47Added regression to r1182 output of PST P1. Binary files only use 1.4MB.Ed Bueler3+39-0
2010-06-28 20:30Fixed a bug in IceModel::compute_hardav_staggered() and IceModel::velocitySSA(). (Yes, it was my fault.)Constantine Khroulev1+17-7
2010-06-27 23:37Cleaned up some metadata and titles: pclimate and IceModel long names for acab, artm now agree. simplified call to update_nu_viewers(). improved titles for viewers to use short_name when no long_name. added titles for nuH, diffusivity viewers. added -prof to pst.sh and spinup.sh.Ed Bueler10+26-17
2010-06-25 23:40* Added IceFlowLaw::effectiveViscosity(hardness, u_x, u_y, v_x, v_y), computing viscosity (not! nu*H). Note that HybridIce (G-K flow law) does not have this factorization and should be modified so that HybridIce is *not* derived from ThermoGlenIce. * Updated IceModel::computeEffectiveViscosity() to use ice hardness computed in IceModel::velocitySSA() and stored in IceModel::hardav (on the staggered grid). Tests I and J seem to work fine. Please test and close the task #6997 if this code does work the way it should.Constantine Khroulev3+59-54
2010-06-25 19:57* Minor clean-up. * Fixed a bug in IceModel::computeEffectiveViscosity(). Luckily it only affected PISM's cold mode. * Added IceModel::compute_hardav_staggered() and the new field IceModel::hardav. * Added the averagedHardnessFromEnth(...) method to the GPBLD flow law.Constantine Khroulev7+124-14
2010-06-24 20:39The final fix needed to make r1175 work. Sorry.Constantine Khroulev1+1-0
2010-06-24 20:24Added the IceModelVec2Stag code I mentioned (and forgot) a minute ago.Constantine Khroulev1+15-0
2010-06-24 20:22* Renamed Enthnew3 -> vWork3d. Removed Tnew3 and taunew3. (Temperature and age code uses vWork3d now.) * Added IceModelVec2Stag::{get,set}_components().Constantine Khroulev8+62-81
2010-06-23 22:02Moved the code allocating cell_area to IceModel::createVecs(). Now cell_area is always allocated and is set to either dx*dy throughout or cell areas computed using lat and lon fields and WGS84 datum. This means that PISM uses a little more memory (in the cases that don't really matter anyway) but we can get rid of code duplication and if-clauses in lots of IceModel::compute_... methods.Constantine Khroulev6+133-278
2010-06-23 18:54Added test/software/pism/test_00.sh to address task #7108.Constantine Khroulev1+60-0
2010-06-23 16:46* Finished task #7109 (flushing time-series when a snapshot is written). * Profiling data no longer includes CPU times. * Minor clean-up and bug-fixes. * Added (commented-out) code necessary to produce NetCDF-4 files (with or without compression) once PISM uses NetCDF-4.Constantine Khroulev11+113-112
2010-06-22 18:40- added new scalar and spatially-varying output fields (ivolcold,ivoltemp,iareacold,iareatemp,tempicethk)Andy Aschwanden5+313-2
2010-06-22 06:04Added option -sliding_scale to do experiment E1 for SeaRISE-Greenland. The picture from using this in the PST experiments shows that membrane stresses matter, sliding velocities are a *model output*, and that there is some speedup of the right magnitude but it is complicated as the thermal state, basal water pressure, and geometry all change. Also added documentation of pseudo plastic basal law, and cleaned up some stuff.Ed Bueler7+108-27
2010-06-21 06:49Very small changes to clean up usage messages.Ed Bueler7+29-29
2010-06-19 00:43fixes to example documenting force-to-thkEd Bueler3+50-0
2010-06-16 06:12Minor: Update manual reference in usage message. Clean up MISMIP example.Ed Bueler3+7-9
2010-06-04 00:54Improved documentation on -force_to_thk mechanism.Ed Bueler3+39-20
2010-06-02 09:17Merge of r1160:1161 changes from stable0.3. MISMIP looks good.Ed Bueler4+31-21
2010-05-25 21:23Rewrote some of PISMProf to make it independent of PISMIO (that way I can use PISMProf to profile PISM input and output).Constantine Khroulev6+81-67
2010-05-24 20:37Merged 1157:1158 changes from stable0.3.Constantine Khroulev2+6-4
2010-05-22 01:19Closing task #7065: written diagnostic vertical velocity fields are correctly relative to the geoid, and have metadata which reflects that.Ed Bueler2+4-7
2010-05-22 00:40Work related to bug #15886: The three recently-introduced time series (whacked,nuked,fried) are renamed and given metadata. Now there are six fluxes in time series: surface_ice_flux basal_ice_flux sub_shelf_ice_flux nonneg_rule_flux ocean_kill_flux float_kill_flux. But the semantics of nonneg_rule_flux are still not clear; what sum should be zeros? Also sign convention consistent for these fluxes: positive means mass gain for ice sheet.Ed Bueler5+44-41
2010-05-21 20:52* Almost done with the task #7065. * Some more work on the bug #15886.Constantine Khroulev8+85-53
2010-05-21 18:30* Some work on the task #7065. * Some work on the bug #15886. * Changed examples/searise-greenland/spinup.sh to use -gradient instead of -no_eta.Constantine Khroulev8+150-18
2010-05-21 07:20Removed no un-called computeMaxDiffusivity() after Constantine fixed the code to do the right things. Massaged comments to describe new situation, in which diffusivity is computed as a side-effect inside velocitySIAStaggered(), except when desired as a diagnostic output; see compute_diffusivity() for that situation.Ed Bueler5+33-128
2010-05-20 21:07merge of r1150:1151 changes from tags/stable0.3Ed Bueler2+6-3
2010-05-20 05:05* Moved max. diffusivity computation to velocitySIAstaggered(). -view_map diffusivity should work still. * Replaced test B,E,L outputs in software test scripts. * Added $(PETSC_INCLUDE) to CFLAGS * Removed #ifdef LOCAL_GHOST_UPDATE and related preprocessor directives. * Implemented PISMEvent and PISMProf, a class for PISM profiling. Added some events. (Try pismv with "-prof", for example.) * Removed "inline"s.Constantine Khroulev21+557-117
2010-05-17 18:00* Fixed bug #15984 (I'm sure it was my fault). * Some minor clean-up.Constantine Khroulev4+23-21
2010-05-14 08:53Added the case Marianne contributed, which shows strongly the difference among the three -gradient {haseloff|mahaffy|eta} methods. Note the eta method does really badly here. Demonstrates and fixes bug #12650.Ed Bueler2+14-0
2010-05-14 06:59Added an implementation of the Marianne Haseloff surface gradient scheme that modifies the surface slope at margin points where ice-free bed elevations are above surface elevations. Fixes bug #12650. This method is the new default in trunk, but fully-testing will require ongoing use in 'real' modeling.Ed Bueler6+156-59
2010-05-12 18:21Very minor stuff. But on 1km whole-Greenland grid I need more digits in 'model year per proc hour'.Ed Bueler2+3-3
2010-05-10 18:14Merge of r1139:1140 and r1143:1144 changes in stable0.3.Ed Bueler4+13-19
2010-05-07 07:37Merge of r1141:1142 changes from stable0.3/doc/Ed Bueler1+37-35
2010-05-06 23:31* Fixed a bug in pclimate that I introduced a revision or two ago. * Fixed a pedantic compiler warning.Constantine Khroulev2+3-1
2010-05-06 00:19* Finished task #7049. * Added new command-line options: -Nx, -Ny (numbers of processors in the x,y direction) and -procs_x, -procs_y (widths of processor domains in the x,y direction). * Fixed a typo in the installation manual.Constantine Khroulev14+155-42
2010-05-05 00:13Additional small change to diffusivity calculation, to 'top-off' the integral using the interval [z(ks),H]. Effect on SIA regressions is entirely minor, and improvements; the only exception is the test L volume, even though test L maxH and avH improve. Also, cleaned-up spinup.sh in SeaRISE-Greenland.Ed Bueler7+67-65
2010-05-04 19:53Third strict simplification of IceModel::velocitySIAStaggered() by removal of inactive code: K[] removed. Changes documented, and role of both D=Dfoffset and I[] clarified.Ed Bueler1+48-28
2010-05-04 18:46Second strict simplification of IceModel::velocitySIAStaggered() by removal of inactive code: Dfoffset calculation had an extra difference and product; removed.Ed Bueler1+2-3
2010-05-04 18:13Strict simplification of IceModel::velocitySIAStaggered() by removal of inactive code: J[] was only evaluated at k=ks, so only that level is computed.Ed Bueler1+5-7
2010-05-01 01:58Merge of 1128:1129 changes from stable0.3.Ed Bueler1+5-3
2010-04-29 19:32Storglaciaren-related references.Ed Bueler1+23-0
2010-04-29 17:41Bibliography changes only.Ed Bueler1+3-4
2010-04-29 17:10Added a reference to doc/ice_bib.bib.Constantine Khroulev1+11-0
2010-04-28 13:41- updated reference in ice_bib.bibAndy Aschwanden1+5-4
2010-04-27 20:53* Fixed a bug in the PISMVars class. * Merged 1120:1122 changes from stable0.3.Constantine Khroulev9+12-8
2010-04-26 07:45Tools for least-squares (alternative) bed gradient calculations.Ed Bueler4+228-0
2010-04-26 07:01Created methods diff_x_stagE(), diff_y_stagE(), diff_x_stagN(), diff_y_stagN() for differencing IceModelVec2S onto staggered grid. No change in calculational method; still the centered/Mahaffy versions. Lots of code could use these diff_?_stag?() methods if we want. Only currently used in IceModel::surfaceGradientSIA(), however. Assert that in all of PISM, horizontal differencing of vbed only occurs through vbed.diff*() methods, though regular grid and staggered grid gradients are both in use.Ed Bueler4+43-10
2010-04-25 22:01Small improvements to has_nan() debugging tool for IceModelVec s. Also see note in IceModel::run() about fixed length pre-time-step and derived classes; with this change IcePSTexModel runs, but it is a kludge.Ed Bueler5+71-7
2010-04-23 08:23Minor browser-related improvements, esp. merge of -r1118:1119 from stable0.3.Ed Bueler4+8-5
2010-04-22 08:38O.k. here is a bulge limiter. It allows advection bulges with enthalpy values equivalent to 30 K colder than surface enthalpy. It never even fires on SeaRISE-Greenland v0.93 SSA for 20km and 10km. But I think it will stop the problems with the 'extreme' CReSIS topg. I will move it into stable0.3 only after you (Andy) have used it a bit in trunk.Ed Bueler4+36-24
2010-04-22 00:27merge of 1114:1115 changes to stable0.3/doc; this means pism-dev/doc/installation.tex says 'pism0.3' instead of 'pism-dev' in various places, but surely pism-dev users can handle that; at stable0.4 a simple search-and-replace should workEd Bueler1+54-35
2010-04-21 07:33Made total_basal_mass_flux calculation respect include_bmr_in_continuity because that makes things clearer.Ed Bueler2+7-5
2010-04-21 04:37merge of r1110:1111 from stable0.3Ed Bueler1+7-2
2010-04-20 22:03Fixed bug #15929. Merged 1099:1100 changes from stable0.3 to trunk.Constantine Khroulev2+15-11
2010-04-20 20:10Merge of changes r1106:1107 on stable0.3.Ed Bueler4+38-2
2010-04-20 01:27merge of r1103:1105 from stable0.3Ed Bueler2+5-5
2010-04-20 00:44Some people don't read PDFs or pay attention to websites ... so here's the minimal install scoop.Ed Bueler1+26-9
2010-04-20 00:13Minor things, to be merged with stable0.3: PST does run but still with error messages about diagnostic timeseries; better error reporting in iMenthalpy; minor doc stuffEd Bueler5+27-22
2010-04-19 20:34Added new userman figures; fixed some typos.Constantine Khroulev9+239-335
2010-04-19 07:50Almost final User's Manual, I think. Figures in Getting Started need to be generated from NetCDF files.Ed Bueler7+177-89
2010-04-16 21:49* Added pdd_limit_timestep to pism_config.cdl with a doc. string discouraging its use. * Modified ice_mass_bookkeeping(): it uses corrected cell areas now. * Fixed viewers of 2D vector fields. * Minor userman fixes.Constantine Khroulev5+65-30
2010-04-16 20:13More fiddles; removed some old figures, replaced rossquiver.pdf with rossquiver.png (which is smaller and renders faster).Constantine Khroulev14+74-326
2010-04-16 19:32* Updated two figures in the EISMINT-Greenland section (visualizing climate inputs...) * Added a little bit of code to limit time-steps used by pclimate to 1 year if PDD is used (just as pcctest did).Constantine Khroulev5+18-2
2010-04-16 18:40User's Manual: updated error plots, removed old error reports, added a new (r1084) error report. Fixed the order of subfigures in Figure 1.Constantine Khroulev17+6-60
2010-04-16 18:23Minor User Man stuff to make it build for me, and removing FIXMEs, and slight content update in Getting Started.Ed Bueler4+12-19
2010-04-16 18:04* Modified Python scripts so that every one of them takes the --help option and is documented in the "Packages" tab of the PISM browser. * More work on the Manual; EISMINT-Greenland and Ross sections are OK now. * Removed check_polarstereo.py * Updated eisgreen.py * Tweaked the cheatsheet layout.Constantine Khroulev18+438-526
2010-04-16 13:38Squashed commit of the following:Andy Aschwanden2+3-1
2010-04-16 10:13Our options for spinup were missing '-ocean constant,forcing ' so I think we were not getting sea-level forcing. No big deal for User Man tutorial runs already done. But fixed; I am rerunning on 20km.Ed Bueler1+1-1
2010-04-16 10:11Some work on practical usage; I like our User's Manual ;-)Ed Bueler3+46-69
2010-04-15 23:04Some changes to flow law and stress balance parts of User Man. Why does Table 8 waste a page?Ed Bueler1+32-43
2010-04-15 19:47* Updated test/vfnow.py so that --userman works again. * Fiddling with the User's Manual.Constantine Khroulev10+168-105
2010-04-14 23:54Consolidated pross.cc into src/eismint/. Cleaned section 3 'Quick tests of the installation' of Installation Manual; included concept software tests there.Ed Bueler4+141-146
2010-04-14 21:19Squashed commit of the following:Andy Aschwanden1+9-0
2010-04-14 20:22* Finished task #7026. * Minor edits of the cheat-sheet. * Moved rules necessary to build pross to src/eismint/pross_make, which is included by build/Makefile.Constantine Khroulev12+43-39
2010-04-14 19:08* Finished working on the cheat-sheet. * Minor changes (User's Manual, typos, comments...).Constantine Khroulev6+370-126
2010-04-14 12:15 - new python script nc2cdo.py added, which adds CDO-required variables and attributes to a netCDF file. Script has doxygen documentation - slightly modified doxygen docus of other python scriptsAndy Aschwanden4+235-5
2010-04-14 07:11Fix to enhancement factor in pre-spinup stage of searise-greenland.Ed Bueler1+2-2
2010-04-13 23:12Various minor tweaks to manual.Ed Bueler8+15-15
2010-04-13 17:03* More work on the User's Manual. * Realized that the cell area correction idea was right, but the code was wrong. Fixed it.Constantine Khroulev6+199-215
2010-04-13 04:06Dabbling in User's Manual. Added script to searise-greenland example which does control run only; no future climate speculation in part because that part of SeaRISE has not stabilized.Ed Bueler5+163-45
2010-04-12 20:48* Updates software test #21. * Actually finished task #7025 by fixing a bug I introduced in r1073.Constantine Khroulev4+14-9
2010-04-12 19:01* Some more work on the User's Manual. * Finished task #7025 (it turned out to be easier than I thought).Constantine Khroulev15+275-78
2010-04-12 18:33More tidying-up of old/unmaintained/bogus SSA-related code. Note exactTestM.[h|c] remains in src/verif/tests/, and can be run on IceExactSSAModel/IceModel code with 'pismv -test M', but with no expectation of success. Also fixed communication bug for test M in IceExactSSAModel, and removed all mention of test M from User's Manual.Ed Bueler8+69-41
2010-04-12 04:53* Rewrote SSA to use DAGetMatrix and MatStencil and avoid the local storage of the solution (SSAXLocal) and the VecScatter and the method moveVelocityToDAVectors(). And it works! Tests I and J regression passed, EISMINT-Ross just fine, SIA+SSA hybrid searise-greenland run completely normal.Ed Bueler2+150-147
2010-04-11 06:52Improved and documented the -ssa_matlab mechanism. Fixed doxygen formulas in docs for computeMaxDiffusivity().Ed Bueler2+61-45
2010-04-11 04:18Fixed bug #15833.Constantine Khroulev2+27-15
2010-04-11 01:38* Moved unmaintained code related to SSA into src/base/ssaplay_nobuild/. This stuff includes ad-hoc-inversion and ssa-by-SNES stuff, none of which we are currently maintaining, but which we do want to keep in the trunk. * In stable0.3 tag we will remove these subdirectories: src/base/ssaplay_nobuild/ doc/old/ For the contents of these two directories, I would not want to explain to any user what they do or how to make them work. A quick look through the PISM sources suggests none others to remove. There *are* pieces of code which are not compiled, but these are mostly/all of the "we might have a user who can be helped by compiling/running this to diagnose something" category.Ed Bueler15+28-114
2010-04-10 05:42* Realized something I caused long ago. (At the time I didn't think it could make things work, and it did not then, probably because of subtle interaction with the infamous "w[0]" bug.) Namely, that the diffusivities computed in computeMaxDiffusivity() are computed differently from the same quantities in velocitySIAStaggered(). Fixed that. A efficiency improvement can be made because I have introduced a double call to surfaceGradientSIA().Ed Bueler4+99-57
2010-04-09 22:43Added Test #22: enthalpy symmetry near the base (pisms -no_cold).Constantine Khroulev1+65-0
2010-04-09 21:12* Some more work on the Manual. * Fixed bug #15819.Constantine Khroulev8+241-173
2010-04-09 08:37User's Manual changes which, I hope, at least did no harm.Ed Bueler5+88-73
2010-04-08 22:49* Some more work on the User's Manual. * Renamed pcctest->pclimate.Constantine Khroulev15+633-617
2010-04-08 21:00Bit of work on Getting Started.Ed Bueler2+38-62
2010-04-08 19:06More work on the User's Manual.Constantine Khroulev7+213-156
2010-04-08 18:26Added variables to some saves for uniformity of viewing results. Added -eta; still don't know if performance effect is as I think, and gives no improvement on 20km grid, but high res is the relevant case.Ed Bueler1+8-5
2010-04-08 03:00 - title page updated - link colors changed to blueAndy Aschwanden3+21-7
2010-04-08 01:37* More work on the User's Manual. * Fixed bug #15817. * Removed "coordinates" and "grid_mapping" from attributes of "lat" and "lon" (in an output file).Constantine Khroulev10+387-365
2010-04-07 23:38All small stuff. Turned back on liquified-cell reporting if any detected; they turned out to almost all be related to bug fixed in r1054. Added nice through-the-little-ice-age climate demo to searise-greenland.Ed Bueler4+14-6
2010-04-07 17:54* Added the hyperref version string to the \usepackage command in manual.tex. * Moved a bunch of config.get(...) calls out of PDDMassBalance and PAYearlyCycle methods that are called in i,j loops.Constantine Khroulev5+25-13
2010-04-07 00:17Lots of changes, all related to the User's Manual. * Renamed several figures to avoid using _ in file names. * Added a rule to doc/userman/Makefile to create ice-bib.bib (same reason; old ice_bib.bib is still in place, because it is used in lots of places). * Split the index into a) General index and b) Command-line option index. This requires the amsmidx package *and* a recent version of hyperref. * Started using the "underscore" package (lots of changes are related to this). * Removed all the PetscOptionsHead calls; we use PetscOptionsBegin/End instead now (it seems more stable). * Added a check to NCConfigVariable::get_string() so that it complains if a string attribute is unset.Constantine Khroulev46+442-412
2010-04-06 22:41* Fixed unreported bug, the symptoms of which were high bedrock temps below locations with high basal melt rates. The fix is to remove IceModelVec3Bedrock::getValColumnQUAD(). (I think it should stay removed; the stability benefits of PL are large and, because the bedrock problem is pure diffusion and not advection-dominated, the errors from PL are quite tolerable.)Ed Bueler6+14-100
2010-04-06 01:51Moved a call to config.get_flag() out of an i,j loop. This saves 5 to 10 percent of total time for thermocoupled SIA.Ed Bueler1+4-3
2010-04-05 19:33* Fixed bug #15785. (Yes, it was a variable initialization problem...) * Renamed pismd->pross; updated the files in PISM src/ and doc/ trees to reflect this change.Constantine Khroulev15+162-167
2010-04-05 18:53* Removed IceModel::diagnosticRun(). * Added IceModelVec2::view() and IceModelVec2V::view() (instead of only having IceModelVec2S::view()). * Changed the contents of pismd.cc; now it only does the EISMINT-Ross computation. Will rename to something like "pross.cc" later. Other diagnostic runs should use "pismr -y 0 -o_size big".Constantine Khroulev13+178-193
2010-04-05 07:22(1) Moved SSA object allocation and destruction into iMssa.cc where it is easier to think about; eventually the SSA solver will be an independent-of-IceModel class. (2) Documentation improvements, especially relating to assembleSSAMatrix(). Also fixed update and use of revision.tex in doc/, I think.Ed Bueler7+153-86
2010-04-05 06:37Added two examples which show matrix assembly using DAGetMatrix and MatValuesSetStencil. The second example (lapdof2.c) has dof=2. These examples fill a gap in PETSc examples, I think: use of good DA-based tools to solve a linear problem without requiring DMMG or SNES concepts.Ed Bueler5+500-66
2010-04-03 02:24All small stuff: fix bed_def scripts, change liquified reporting, clean up mass accounting commentsEd Bueler4+40-33
2010-04-03 01:15resolved Conflicts: doc/doxyfile_browserAndy Aschwanden4+77-19
2010-04-02 23:51* Finished refactoring the bed deformation code (see task #6630). New command-line option: -bed_def <keyword>, where keyword can be one of "none", "iso", "lc". * Fixed PDD code (now if computes the PDD sum at most once per time-step).Constantine Khroulev17+340-467
2010-04-02 18:47User Manual refactoring and cleaning.Ed Bueler9+895-948
2010-04-02 17:11 - replaced figs in getting started section with (possibly too) fancy onesAndy Aschwanden9+2-2
2010-04-01 23:44First draft of new Getting Started section.Ed Bueler8+95-94
2010-04-01 20:58In preparation for replacement of Getting Started section, here is the old version separated out.Ed Bueler2+162-154
2010-04-01 19:29Fixed bug #15762.Constantine Khroulev1+6-3
2010-04-01 19:16* IceGrid manages the parallel distribution of the grid now (instead of asking PETSc to compute the number of processors in the x and y directions). This allows us to handle cases like "mpiexec -n 16 pismv -Mx 3 -My 3" more gracefully. * Combined vuvbar[0],vuvbar[1] -> uvbar (IceModelVec2Stag).Constantine Khroulev19+186-189
2010-04-01 17:44 - modified doxygen in include files in util/ - python scripts check_stationarity.py and fill_missing.py updated, they have now doxygen-readable documentation and show up as 'packages' in the browser. - other python scripts will follow laterAndy Aschwanden3+108-84
2010-04-01 17:34* Optimization in enthalpyAndDrainageStep(): fast resolution of ks==0 case where there is no ice in the column. (This optimization will likely only pay off if number of grid points in most-ice-filled processor domain is less than in ice free domains.) * I believe code clarity is increased by removing method drainageToBaseModelEnth() and integrating it into enthalpyAndDrainageStep() main loop. No (intended) change in semantics.Ed Bueler2+86-69
2010-04-01 02:06Added refreeze mechanism to improve time-evolution of basal strength. Fixed unreported bug in ocean values of bmelt. Added bmelt,bwat to time-series for searise-greenland/spinup.sh.Ed Bueler5+75-12
2010-03-31 23:46Minor doc and cleanup in IceEISModel to clarify derived class init issues, so it is a 'model citizen'.Ed Bueler2+14-6
2010-03-31 23:44Changed the way time series ienthalpy is computed so it scales correctly with ice volume not number of grid points. Deleted time series dienthalpydt; it can be computed easily by post-processing, or by new implementation, but the deleted implementation was incorrect.Ed Bueler2+25-40
2010-03-31 21:23* Made the computational grid used in test/vfnow.py -t K finer, so that test/vfnow.py -n 8 -t K works. * Combined vubar,vvbar -> vel_bar. Realized that we don't need to communicate vel_bar (and that it can be "global"); made it global and removed corresponding communication calls.Constantine Khroulev16+96-138
2010-03-31 20:12Fixed a typo in the previous revision.Constantine Khroulev1+2-2
2010-03-31 20:07* Disabled DAGetOwnershipRages calls if PETSC_VERSION_MAJOR < 3 so that PISM compiles and works with PETSc 2.3.3. * New config. parameter: pism_config:do_cold_ice_methods. Note: one can run eisII experiments in the polythermal mode with "pisms -no_cold".Constantine Khroulev20+63-44
2010-03-31 07:31Updated some example scripts to use new SSA options. Also changed searise-greenland spinup script to use more-conservative basal strength and enhancement parameters.Ed Bueler2+4-4
2010-03-31 01:15* New command-line options: -ssa_sliding and -ssa_floating_only. Options -ssa and -plastic are gone; PISM prints an error message and stops if either of them is used. * Improved test_12.sh. * Configuration flag do_superpose is "true" by default now. * Fixed IceModel::bootstrapFromFile(): now the "mapping" variable is read in correctly again.Constantine Khroulev4+56-28
2010-03-31 00:13Realized that I was wrong about NetCDF variable storage order; undid all the changes related to it in 1027:1028. PISM writes temp(t,z,y,x) and similar now. Note: PISM output files are just as usable as before (ncview reads them just fine). And we follow the corresponding CF recommendation, too! :-) (As an additional bonus, we avoid the clutter in the code that I introduced in r1028.)Constantine Khroulev17+88-141
2010-03-30 22:43* Added a config parameter that allows changing the NetCDF storage order of 2D and 3D variables. * Resolved the IceModelVec2T::create(...) ambiguity (or at least I think I did). * Added another version of IceGrid::createDA(...). It should be useful later (once I get to load balancing). Also: added IceModel::compute_proc_ice_area(), computing proc_ice_area (the number of cells containing ice in a processor's domain, at the end of the run). * Tiny IceModelVec3 refactoring (reduces code duplication). * Fixed bed deformation code (vHlast needed a wide stencil).Constantine Khroulev27+260-115
2010-03-29 08:12Added config parameter summary_volarea_scale_factor_log10 which can be set to something other than 6 so that vol and area in stdout are scaled by e.g. 10^0 instead of 10^6. (See application in Storglaciaren.)Ed Bueler4+28-8
2010-03-29 06:09Minor cleaning in docs and scripts.Ed Bueler3+12-15
2010-03-29 06:06Added 'using' declarations to deal with warnings coming from use of -Woverloaded-virtual flag. I think one of these warnings is serious; see FIXME in src/coupler/iceModelVec2T.hh.Ed Bueler2+12-0
2010-03-28 20:55It has always irritated me that the Ross ice shelf from EISMINT-Ross doesn't show correct front; now at least it looks better. Also minor doc changes to EISMINT-Ross; I think that section is fine for stable0.3. Also clobbered 'undefined refs' LaTeX in building User Man.Ed Bueler8+46-27
2010-03-27 20:03Remove 'BPsacr' report at verbosity level 2; BOMBPROOF page is adequate doc on this tradeoff. Slight reduction of frequency of basal map write in searise-greenland pre-spinup.Ed Bueler2+2-2
2010-03-27 18:44Entirely changes related to doxygen HTML browser: Now my browser includes IceModelVec, which it did not in prev revisions. The key change was to move 'ifndef __IceModelVec_hh' lines *ahead of other includes*. Also note three slashes has doxygen meaning (sometimes) and so a few uses were changed to just two; maybe this does not matter. And edstressbalance code is not compiled but does get doxygen sweep; can't exclude it by obvious mechanism ... odd. Also: removed useless 'An overview of C++ classes in the PISM source code' page and cleaned-up 'PETSc: An overview ...' page in Related Pages tab.Ed Bueler5+49-108
2010-03-27 02:37PetscOptionsHead() and Tail() were stopping -topg_to_phi option from getting read; fixed but the fix may be temporary because I don't know why they were there exactly. Also minor stuff: changes to doc strings in test_12.sh and in pism_config.cdl. And bedrock thickness and fineness change in searise-greenland/spinup.sh.Ed Bueler4+12-12
2010-03-27 00:22Finished task #6998.Constantine Khroulev1+6-0
2010-03-27 00:09* Test 12: mass conservation within rounding error for nonsliding SIA moving margin (an imperfect implementation). * Test 14: test E instead of test A. * Test 17: test L instead of test F. * Fixed test L.Constantine Khroulev4+76-32
2010-03-26 17:22Added regression (software) tests based on verification tests.Constantine Khroulev9+377-1
2010-03-26 07:15Reordered searise-greenland/spinup.sh reporting to stdout so that user could clearly see which were set-able environment variables. Also minor fiddles with eisgreen/try_poly_cold.sh: if we want the full capability to compare -cold runs to polythermal runs we need to handle the case of '-cold -i foo.nc' where foo.nc was written by a polythermal run sans 'temp'; currently we fail. (And we may want to delete try_poly_cold.sh from stable0.3.) And fixed spelling of 'flotation'; my fault.Ed Bueler4+81-84
2010-03-25 23:16* Renamed ssavel->vel_ssa, basal_vel->vel_basal. * Switched to using a wide (2 grid points) stencil again. Grep for LOCAL_GHOST_UPDATE and WIDE_STENCIL to see the changes.Constantine Khroulev21+290-185
2010-03-25 22:57Changed name of library containing verification tests: libtests.so -> libverif.so. (I realized that the use of 'tests' in PISM was unnecessarily ambiguous.) Remove lib/libtests.so to avoid confusion. Also removed IceScandModel, which was an unmaintained stub; revive it if needed from svn repo.Ed Bueler5+12-244
2010-03-25 20:18* Fixed test C and the way Enth3 is used in iMreport.cc (again). * Config flag renamed: do_plastic_till -> use_ssa_when_grounded. * Reapplied the optimization of the quad. interp. code. * Removed the "vote by neighbor" code from update_mask(). * Put IceModelVec2Stag back in.Constantine Khroulev23+225-115
2010-03-25 18:28Undid the r1007:1008 changes.Constantine Khroulev37+479-625
2010-03-25 02:47Fix to IceGrid constructor which quashes compiler warning on midnight. Very minor editing of User's Manual and scripts.Ed Bueler4+64-81
2010-03-25 01:39Small changes which allow temppabase to be correctly computed in extra_file.Ed Bueler2+2-2
2010-03-24 23:59Minor cleanup for stable0.3: penth removed entirely; flowTable moved out of the way.Ed Bueler4+83-101
2010-03-24 20:11* Config flag renamed: do_plastic_till -> use_ssa_when_grounded. * Mask is updated locally, "vote by neighbors" code is gone. * Combined vubar,vvbar -> vel_bar. Renamed ssavel->vel_ssa, basal_vel->vel_basal (for consistency). * Optimized local quadratic interpolation from the storage grid onto the fine computational vertical grid: now it only interpolates *within* the ice. Values above the ice are set using constant extrapolation from the topmost level *in* the ice. This code is used inside temperatureStep and enthalpyAndDrainageStep, which use similar extrapolation when transferring the solution back onto the storage grid, so it does not require new assumptions. * Implemented ICMEnthalpyConverter, a converter used in IceCompModel. This fixes test C (was broken in r1006). * Fixed a bug introduced in r1006 (compute_temp_pa used Enth3 instead of Enthnew3 as temporary storage). * Added more documenting comments in IceModel::velocity(). * Implemented IceModelVec2Stag, a class for storing and accessing internal staggered-grid 2D quantities (surface gradient and SIA velocities). It is almost identical to IceModelVec2V, but does *not* support I/O and its components are not called "u" and "v". Will do some tests and remove it if it slows PISM down. * Work vectors, tau3, vh, vH, topg, vMask, vHmelt, vel_ssa, vel_ssa_old, Enth3, vdHdt, vuplift have stencil width = 2. This allows updating stencil width = 1 ghosts locally in many cases and avoid communicating eta, surface gradient, vuvbar[0,1], Sigmastag3[0,1], Istag3[0,1]. I need to do more tests to see if vel_ssa should have w=2 to avoid communicating nuH to get ghosts.Constantine Khroulev37+625-479
2010-03-23 23:02Fixes to comments in iMtemp.cc and iceEISModel.cc. Fixes to scripts. PST experiments probably do not run yet; I'll work more on that.Ed Bueler4+82-60
2010-03-23 00:14* Task #6868: IceModel only allocates T3 if doColdIceMethods==true. * Separated mask and surface elevation updates; surface elevation ghosts are updated locally. * New class: IceModelVec2V (for storing/reading/writing/regridding 2D vector fields). Also: re-structured IceModelVec inheritance: IceModelVec2 is a common base cl ass of IceModelVec2V and IceModelVec2S (S for "scalar", the old IceModelVec2). Global replace in the code: "IceModelVec2" -> "IceModelVec2S". * Combined: vubarSSA,vvbarSSA -> ssavel; vub,vvb -> basal_vel. * Updated software tests. * Fixed IceROSSModel.Constantine Khroulev70+1698-1226
2010-03-18 19:32Fixed unreported bug in way temp_pa is calculated with option -cold; 273.15 was subtracted twice. Also trivial change to comment re vertical velocity.Ed Bueler2+13-6
2010-03-18 19:25Added a couple of references, and tricially fixed a browser page.Ed Bueler2+29-2
2010-03-18 19:19 - r982 introduced compilation error on pingo (nc_util.cc), fixed - added comment to bombproof - check_stationarity script revised, now calculates rate of change of p-norm of a variable (2-norm by default)Andy Aschwanden5+121-88
2010-03-17 21:33* Fixed two small memory leaks I introduced a while ago (see IceModelVec3 and IceModelVec3Bedrock constructors and destructors). * Various optimizations.Constantine Khroulev21+490-490
2010-03-16 18:22* Finished task #6984: "age" is a special case now: it is "diagnostic", but is written out if -age is given. When a -i foo.nc -age run is initialized, age is set to zero if we can't find it in the input file. * Fixed some stupid bugs I introduced earlier (see pism_const.cc). * A little bit of work on the manual. A little bit of cleaning-up.Constantine Khroulev19+217-443
2010-03-15 06:12Added page to Related Pages tab on the change of vertical coordinate. Changed wvel->wvel_rel, and long name thereof, because it is the vertical velocity relative to the base of the ice. IceModel::vertVelocityFromIncompressibility() is *not* changed; the fix to bug #13974, as finalized in r987, is correct in my opinion. That is, the w[0] issue is resolved.Ed Bueler10+106-47
2010-03-14 18:57Very minor cleanup in old docs. This directory doc/old/ should be kept in trunk, but deleted from stable0.3.Ed Bueler3+22-2
2010-03-14 05:07Made sure ice temperature (T3) is updated at every energy step. (This should not be needed unless -cold is given. Instead, ice temperature is diagnostic in polythermal runs. That is, task #6868 should optimize away the need for a temperature update at every timestep of a polythermal run.) Also clobbered a compiler warning about PISMIO::open_for_writing().Ed Bueler4+11-0
2010-03-13 23:29* Annihilated IceEnthalpyModel. Executable penth remains only so that a NON-functioning penth is built in your bin/ directory (instead of holding over the old one). Will be deleted in a few commits.Ed Bueler12+679-897
2010-03-13 09:41* Fixed compute_temp_pa so that diagnostic output temp_pa is in degrees celsius; this is wrong in r993.Ed Bueler9+63-113
2010-03-13 08:36* Moved the write of liqfrac and cts from IceEnthalpyModel to IceModel::write_model_state(). Note that temp_pa and liqfrac are written either if user asks specifically for them or if doColdIceMethods==false, while cts is written only if user asks specifically. This maintains previous semantics.Ed Bueler6+134-201
2010-03-13 05:29Moved six procedures which convert IceModelVec3 thermal fields from IceEnthalpyModel into IceModel and new file src/base/iMenthalpy.cc. (Conversions are among temperature, liquid fraction, and enthalpy.)Ed Bueler5+250-231
2010-03-13 04:55* Moved creation of IceModelVec3's "Enth3" and "Enthnew3" into IceModel. This implies a performance hit for all PISM executables, until the conversion to enthalpy-based runs are complete and we remove T3 and Tnew3. Deleted IceEnthalpyModel::createVecs().Ed Bueler8+61-104
2010-03-13 03:24Added an EnthalpyConverter* to IceModel, and removed local creation of it in various IceEnthalpyModel methods. Moved content of IceEnthalpyModel::init_physics() to IceModel. Changed regularization warm_base_flux_enthalpy_fraction to 0.0001 from 0.02. (This is the first of several changes which will remove IceEnthalpyModel as a separate class.)Ed Bueler7+90-91
2010-03-13 01:19Lots of (boring) changes related to PISM command-line option processing overhaul (task #6493).Constantine Khroulev43+734-566
2010-03-13 00:38Re-added 100 year SIA smoothing step to searise-greenland/spinup.sh, which is needed for reasonable -no_mass performance. Also fixed filename bug (PRE4NAME) in same script. And cleaned eisgreen/try_penth.sh.Ed Bueler2+21-8
2010-03-12 23:10Fix of bug introduced by Bueler in commit r920. This commit was supposed to fix bug #13974, but introduced a different problem. This bug fix is highly-relevant to any evaluation of results from trunk versions 920--986 inclusive.Ed Bueler1+2-2
2010-03-12 02:18 - usermanual updated - some work on the cheat sheet - fixed eisross example - fixed typo in some python files (was my mistake)Andy Aschwanden9+112-104
2010-03-10 20:59 - added new extra_vars to userman - added enthalpybase, enthalpysurf to extra_vars - minor modification in check_stationarity.pyAndy Aschwanden5+74-2
2010-03-10 02:40Squashed commit of the following:Andy Aschwanden6+626-1
2010-03-09 22:33Added regularization which allows some heat flux into base when the ice is slightly warmer than pressure-melting. Also fix to allow bootstrapping runs, without false error message.Ed Bueler4+46-12
2010-03-09 20:05* Improved error checking in the regridding code. * Re-factored NCTool. NCTool is split in two parts: a minimal NetCDF wrapper (new "NCTool") and a class for reading/writing PETSc Vecs that needs to know about IceGrid ("PISMIO"). * Moved IceGrid defaults into pism_config.cdl. * Finished task #6978.Constantine Khroulev38+1237-1208
2010-03-08 23:12Squashed commit of the following:Andy Aschwanden3+25-2
2010-03-06 03:36* For IceEnthalpyModel, in case of downward velocity at a temperate ice base, added source terms and horizontal advection terms to "outflow boundary condition". This seems to not change the basal melt rate very much relative to r979. In the Storglaciaren runs and Greenland runs I did, anyway. (What bmelt is right? Some precise knowledge is needed to guide choices. I cannot think for now of credibly-important thermodynamics which is missing from the code.)Ed Bueler3+51-29
2010-03-05 08:40New material in User's Manual looks good. Here are small edits. Also a figure had not been committed; thisone is PDF from inkscape applied to the corresponding SVG.Ed Bueler3+32-48
2010-03-05 02:15Fixed a bug in PISMSurface that caused a crash if -surface constant and -extra_... were used at the same time.Constantine Khroulev4+38-31
2010-03-04 23:35 - added ienthalpy (volume integral of enthalpy, units Joule) to timeseries as a default output variable - modified Makefile(s) to avoid clashes with different versions of PISM and run scriptsAndy Aschwanden6+57-8
2010-03-04 19:29Updated Schoof references.Ed Bueler1+9-2
2010-03-04 18:53* Added configuration parameters hmelt_diffusion_distance and hmelt_diffusion_time. * Replaced max_hmelt with hmelt_max (for consistency).Constantine Khroulev9+40-30
2010-03-04 03:16Modified error reporting when EnthalpyConverter dects ice that is fully-liquified; now it does not break but indicates with error code. IceEnthalpyModel counts volume of ice that was fully liquified in one time step.Ed Bueler5+84-26
2010-03-04 01:04* Added better error messages if an argument of -atmosphere, -surface or -ocean is mistyped. * Moved grid.{ice,bed}_vertical_spacing == EQUAL check from IceEnthalpyModel into IceModelVec3[Bedrock]::getValColumn(). * Some work on the User's Manual.Constantine Khroulev9+247-54
2010-03-03 19:51In r971 the basal water layer thickness (bwat=vHmelt) was not being updated from the basal melt rate, as opposed to drainage. Fixed. Also added script to compare pismr and penth results for EISMINT-Greenland.Ed Bueler2+21-2
2010-03-03 18:48* Dealt with all bugs identified in r967 commit message.Ed Bueler7+175-45
2010-03-02 01:47* Fixed L[0] out-of-bounds array access in iceenthOnlySystemCtx::solveThisColumn(). Thanks for finding this one, Constantine.Ed Bueler6+61-6
2010-03-01 22:32* Preliminary work on the bed deformation model (i.e. re-factoring). * Option -help shows possible -atmosphere, -surface and -ocean choices. * Fixed pgrn (it suffered from a bug similar to #15420).Constantine Khroulev19+478-21
2010-03-01 02:25 quick hack in practical-usage.text to make userman compilingAndy Aschwanden1+1-1
2010-02-27 07:40* enthColumnSystem.hh|cc removedEd Bueler9+332-451
2010-02-27 00:23IceEnthalpyModel code now in intended condition. Only combinedSystemsCtx.cc needs implementing, following BOMBPROOF docs. (On next commit I will do that, and turn compilation of penth back on.)Ed Bueler2+115-51
2010-02-26 08:18* getValColumnSmart() method added to IceModelVec3 and IceModelVec3Bedrock; chooses between PL and QUAD based on equal-spacing flagEd Bueler9+290-189
2010-02-25 03:07New classes iceenthOnlySystemCtx and bedrockOnlySystemCtx suggest my current design intention: they configure themselves as much as possible but they make no nontrivial decisions (e.g. w.r.t. cases about ice base and bedrock as in BOMBPROOF docs). IceEnthalpyModel code merely compiles; NOT DONE. enthSystemCtx DEPRECATED. Need to add: combinedSystemCtx which handles cold base plus bedrock case (size = Mz + Mbz + 1). See BOMBPROOF doc page for all ideas. Constantine: You might look at iceenthOnlySystemCtx and bedrockOnlySystemCtx to make up unit test. We should talk.Ed Bueler10+573-60
2010-02-25 00:05Maintenance; getting ready to approach task #6630.Constantine Khroulev21+934-840
2010-02-24 18:23Three more references.Ed Bueler1+38-0
2010-02-24 04:26* Added more names to the list of available diagnostic fields in the User's Manual. * Removed output_variables config variable; replaced with output_medium and output_big. Now -o_size big writes 38 2D variables and 5 3D variables. * Fixed software test #1.Constantine Khroulev14+95-76
2010-02-23 23:00* Changed test J refinement path in test/vfnow.py * Started addressing bug #15420. (Testing will show if this helps at all.) * Un-spaghettied temperature and age computations. (Note: test G error report is the same.) * Cleaned up IceModel::assembleSSARhs().Constantine Khroulev15+44-105
2010-02-23 20:50For BOMBPROOF docs, inclusion of no-condition boundary condition at temperature ice base in case of downward vertical velocity. And improvement of floating case description.Ed Bueler1+59-56
2010-02-23 00:12* Added IceModelVec2 methods diff_x(i,j), diff_y(i,j) approximating x and y derivatives using centered differences. "Periodic" (_p) versions respect grid periodicity and use one-sided differences at grid edges, if necessary. Need to think more; see task #6211. * Fixed IceModel::computeDrivingStress(): eta transformation was applied at interior points and not at the edges. (Does not matter in most cases, since we usually try to keep domain edges ice-free. Does matter for MISMIP, though.) * More MISMIP maintenance (see task #6955). Running "mismip.sh <number> SM" works fine. Restarting from a result of a run with 10m uniform initial thickness (such as "mpiexec -n 4 pisms -model 1 -mismip 1a -step 1 -My 3 -Mz 15 -Mx 151 -extras") does not work well at all (there are convergence issues plus results are asymmetric).Constantine Khroulev6+143-122
2010-02-18 23:01* More re-factoring: separated code implementing the cosine yearly cycle using mean annual and mean July temperatures from parameterizations. This will make it easier to implement SeaRISE-Antarctica atmosphere model and other parameterizations using this yearly cycle. * Moved SeaRISE paleo-precipitation correction from "forcing" to SeaRISE-Greenland atmosphere model, as it is SeaRISE-specific. NOTE: -atmosphere greenland is replaced with -atmosphere searise_greenland (for clarity). * PISM now allows -atmosphere constant,forcing without -dTforcing (gives a message that forcing is inactive). * Renamed -regrid_from to regrid_file (for consistency with -extra_file, -ts_file, -save_file). * Mostly fixed MISMIP: runs and produces something. Needs testing (i.e. comparing to results circa r387).Constantine Khroulev23+366-342
2010-02-16 23:47i* Added command-line options -save_size and -o_size (choices are: small, medium, big). See task #6949. Still need to decide on meanings of "medium" and "big" ("small" means "only fields necessary for restarting"). * Removed some old and unused code. * Added two figures for the User's Manual. * Mostly done with task #6898 (boundary models and -extra_vars).Constantine Khroulev28+526-2372
2010-02-16 03:15Not exciting.Ed Bueler20+1345-66
2010-02-15 02:08Added diagnostics for column-wise tridiagonal problems.Ed Bueler3+107-23
2010-02-14 23:50BOMBPROOF doc page rethought. And minor stdout changes.Ed Bueler3+65-82
2010-02-13 19:43 - updated Mac OS X part of the installation manual - fixed typo in usermanAndy Aschwanden2+10-6
2010-02-12 20:19* Switched IceModel to using new "boundary models". * Updated software tests. * Implemented a simple lapse rate correction for air temperatures (try -atmosphere temp_lapse_rate -lapse_rate <number>).Constantine Khroulev60+961-1102
2010-02-11 05:46Adding flexibility to -topg_to_phi option, as suggested by Maria Martin.Ed Bueler2+20-7
2010-02-11 02:57Fixes/updates to User's Manual preamble and references.Ed Bueler3+40-30
2010-02-10 22:50Update to MISMIP scripts so it runs at all. See follow-on bug report; medium-low priority.Ed Bueler2+14-13
2010-02-10 02:46* New mask values. The current list of values is: MASK_UNKNOWN, MASK_ICE_FREE_BEDROCK, MASK_SHEET, MASK_DRAGGING_SHEET, MASK_FLOATING, MASK_ICE_FREE_OCEAN, MASK_OCEAN_AT_TIME_0. * Clean-up and more work on boundary models.Constantine Khroulev27+422-283
2010-02-07 02:21One more little change closes bug #15288.Ed Bueler1+2-2
2010-02-06 21:05* Checked (and fixed) the code handling -ts_times, -ts_vars, -save_vars, -save_times, -extra_vars. (see task #6899). * Ported the force-to-thickness code (it is a PSModifier derived class now). * Ported the EISMINT-Greenland parameterization. * Switched all the boundary models to using interval mid-points. * Fixed a bug in IceGrid::get_dzMIN_dzMAX_spacingtype() (please see bug #15288 and re-run bugreport.sh) * Some more work on putting the PISMComponent class hierarchy to work. * Some clean-up.Constantine Khroulev15+557-50
2010-02-06 02:42So now IceEnthalpyModel compiles. The Mbz==1 case has a diagnosed and fixable (but not fixed) bug. The Mbz>1 case seg faults; will work more on that ... See bug #15288.Ed Bueler7+282-322
2010-02-05 19:11Ongoing small changes in enthalpy code; still not complete. BOMBPROOF docs have the plan. Changes in src/coupler/ are for comments only.Ed Bueler7+267-341
2010-02-04 03:00Substantive mods to test 13 (which still works; strange hyperslab behavior from ncks, though). And some trivialities.Ed Bueler4+14-13
2010-02-03 22:12Added forgotten header files.Constantine Khroulev4+529-0
2010-02-03 22:04* Added software test #13: temperature continuity at the ice-bedrock interface. It needs more work. * Added the code intended to replace PISMClimateCoupler and its derived classes. Run "make ../bin/pcctest3" to get a new version of pcctest which uses the new boundary model hierarchy. Note: this code is not used in PISM yet. Also, please see FIXME notes src/coupler/*Constantine Khroulev13+1997-7
2010-02-03 07:53Addition of some minor facilities: issounding() and a version of check_option() that takes a bool instead of PetscTruth.Ed Bueler4+15-1
2010-02-03 07:38Documentation improvements including happiness with BOMBPROOF docs.Ed Bueler4+174-148
2010-02-01 00:33IceEnthalpyModel does not compile. But making progress I hope.Ed Bueler4+131-89
2010-01-31 09:02Documentation changes. Removal of garbage, mostly, except addition of searchability on browser mainpage, text changes there, and beginning of improvements to test in BOMBPROOF docs.Ed Bueler11+165-1413
2010-01-30 20:43these changes affect IceModel (and all derived): * ColumnSystemCtx base class now holds column indices and can reset system * some updated/corrected comments in grid.cc, doc/browser_pages.txtEd Bueler11+205-495
2010-01-29 22:53* Fixed the "showEnv" make target (it was not printing PISM_CONFIG). * Added a little bit of code to allow flushing time-series by sending USR2 to a running PISM process (or processes). * Fixed the history string handling. * Fixed IceModelVec3::getPlaneStarZ() (another bug caused by using uninitialized memory). * Improved IceModelVec3::has_nan().Constantine Khroulev9+79-18
2010-01-29 08:49* removal of BasalTypeSIA (a type in materials.cc|hh) and BasalTypeSIA *basalSIA from IceModel * removal of unused variables from IceModel::vertVelocityFromIncompressibility()Ed Bueler10+60-74
2010-01-29 07:29* traded in getEffectivePressureOnTill() for getBasalWaterPressure(); the former is easy to calculate from the later if neededEd Bueler9+197-198
2010-01-26 21:52THIS COMMIT MESSAGE SHOWS MOST r927:928 CHANGES AS WELL AS r928:929 (sorry ... bad command in prev commit ...)Ed Bueler10+53-29
2010-01-26 21:01Suggested move of initBasalTillModel() from misc_setup() to model_state_setup(). This seems to work, but moving bed deformation setup seems not to work; why does it seg fault? See preprocessor directives in src/base/iMinit.cc.Ed Bueler23+211-121
2010-01-26 03:16Added executable permission to nc2mat.py.Ed Bueler0+0-0
2010-01-26 03:15Minor updates to EISMINT-Greenland outputs now that year 0 diagnostics are valid.Ed Bueler3+5-4
2010-01-22 23:29Added initializations for two more IceModel parameters.Constantine Khroulev1+3-0
2010-01-22 19:51* Fixed a bug in IceModel::SigmaSIAToRegular() (look for "<=" instead of "<" around iMsia.cc:516). Yes, this one-character change seems to fix pismv -test K, the "pisms; pismr -i simp_exper.nc" sequence and software test 12 (penth). * Modified util/nccmp.py to handle files with NANs more gracefully. * Improved the "diagnostic" step and model re-initialization in IceModel::run(). * Added PISMVars::check_for_nan() and IceModelVec::has_nan() (for debugging). * Slightly improved command-line option processing (options -ice_type, -z_spacing, -zb_spacing).Constantine Khroulev9+100-22
2010-01-21 23:02 - modified buil/Makefile to make it work with petsc-dev - renamed env variable CONFIG -> PISM_CONFIG. It invokes additional OS-specific options into Makefile, e.g. used for OS X. Renaming avoids possible clashes with other software (such as petsc-dev). Modified installation manual accordingly.Andy Aschwanden2+6-6
2010-01-21 08:14 - fixed (hopefully) bug in python scripts which create a new netCDF file. Using netCDF4 module caused creating a netCDF4 file by default. This was changed to NETCDF3_CLASSIC instead.Andy Aschwanden6+6-6
2010-01-21 07:18Fixed summaryPrintLine() to deal correctly with -no_temp runs.Ed Bueler1+19-4
2010-01-21 03:16* Florian is distinctly right about bug #13974. I have fixed the code, I believe, in pism-dev. We need to see the consequences in the dev branch first. Then we (at UAF) are committed to going back and fixing this in stable0.2. For now, users of stable0.2 should use the patch Florian has posted in the bug report. We need a simple verification test with *both* a bed slope and basal sliding.Ed Bueler7+49-103
2010-01-19 22:53* Created IceModel::step(...); moved almost all the contents of the time-stepping loop there. * Modified the initialization sequence so that all the code initializing model state variables is in (or is called from) IceModel::model_state_setup(). Still need to check derived classes of IceModel. * Modified IceModel::run() so that PISM 1) does one time-step 2) re-sets time and re-initializes model state variables 3) goes on to the time-stepping loop. This way a -y 0 run is (should be) equivalent to a "diagnostic" run. Also, this extra time-step makes it possible to compute rates-of-change at the beginning of the run. * Fixed the util/fill_missing.py file permission issue (an output file was only readable by its owner).Constantine Khroulev5+167-121
2010-01-19 21:00* Moved "config" option database out of IceModel and into main(). This guarantees that it (the config database) is initialized *before* it is used. * Changed build/Makefile to import PETSc make variables (and not rules). Will revert to r917 if this does not work. * Removed the code initializing mapping parameters from IceModel::setDefaults(): there is no sensible default I know of.Constantine Khroulev39+173-128
2010-01-19 08:47 - added -init_from_temp_liqfrac option for debugging Storglaciaren melt rate bug. Option is similar to -init_from_temp, but calculates enthalpy from temperature and liquid water fraction found in the input fileAndy Aschwanden2+111-26
2010-01-15 03:17Made IceEnthalpyModel method of computing effective pressure on till, potentially-using basal melt rate, into standard IceModel version. Renamed a method to updateYieldStressUsingBasalWater(). Fixed IceEnthalpyModel::enthalpyAndDrainageStep() to correctly treat case where k0+ks==0. Re-ordered initialization procedures in iceEnthalpyModel.cc to follow actual initialization order. (None of these seem to help with bugs.)Ed Bueler6+154-238
2010-01-14 20:52Fleshed out init sequence flow chart.Ed Bueler1+43-12
2010-01-14 08:23In attempt to track down penth bug, I have improved nccmp.py (I think). And I have added -init_from_temp option to IceEnthalpyModel.Ed Bueler4+94-19
2010-01-13 23:04Now that Constantine has made the EISMINT-Ross example work again, I have done some additional minor maintenance.Ed Bueler9+188-223
2010-01-13 19:25Minor maintenance: prepend to history string in fill_missing.py; fixed messages in iMtimeseries.ccEd Bueler2+5-5
2010-01-12 23:15Test I was broken in 908:909 changes, I think. Turns out Test I was using uninitialized vubarSSA,vvbarSSA. Now have_ssa_velocities is explicitly set to false by IceExactSSAModel.Ed Bueler1+9-9
2010-01-12 22:55Only minor fixes to stdout messages, documentation, and the long name of acab.Ed Bueler6+10-12
2010-01-12 13:58* Fixed EISMINT-Ross and rossplot.py (max computed speed reported: 1378.185 (m/a), average relative error in vector vel = 0.07279, chi^2 = 3684.510) * IceModelVecs (-2 and -3Bedrock) are initialized to NAN at allocation time: this way we can easily see it a field was not initialized at the model setup stage. Added _FillValue as a default NetCDF attribute. Modified iceEISModel.cc and iceCompModel.cc to initialize tillphi explicitly. * Renamed software test #1 to "pismr (SIA only) exact restartability".Constantine Khroulev9+34-32
2010-01-12 07:20Looked at EISMINT-Ross example; definitely broken from deprecated use of 'fundamental transpose', which can now be eliminated. TO DO, that is; I have left some FIXMEs to help on some issues, too.Ed Bueler4+24-2
2010-01-12 00:45Attempting to make sure that basal melt rate (bmelt=vbasalMeltRate) is properly-initialized. Also trivial stdout change for penth.Ed Bueler5+14-10
2010-01-12 00:41Doc improvements only: editing of BOMBPROOF notes and revision of -verbose option docs.Ed Bueler2+74-100
2010-01-12 00:40Fixes to grid parameter reporting and verbose option.Ed Bueler4+67-61
2010-01-11 03:33Cleaned up and simplified the standard out flags and summary. Now that we have diagnotic viewers, less crap should be dumped to stdout.Ed Bueler13+158-171
2010-01-09 00:00Constantine has indeed fixed bug #13113. (For three years I have failed to have this relation clear in my head ... *thanks*.) Fixed test_12.sh to use penth correctly, and now *all 12 software tests pass*. Other minimal maintenance is trivial.Ed Bueler6+12-10
2010-01-06 19:59* IcePSTexModel now a "model citizen", though not well-tested. Uses custom DiagnosticTimeseries for output, not stdout.Ed Bueler12+298-151
2010-01-04 23:54* IceScandModel, with executable "pscand", contains almost all of what was "pisms -eisII S". Andy: If you want to remove the whole thing (i.e. the entire folder src/eismint/scand/) then please do. It was factored out so that IceEISModel would be a clean implementation of EISMINT II; next. * IceEISModel is now a "model citizen" for stable0.3, as is IceGRNModel.Ed Bueler13+474-317
2010-01-04 05:12Minimal and incomplete maintenance and tidying: cleaning of EISMINT II and PST stuff; added refs; Install Manual mods based on recent experience with karmic (Ubuntu); minor on vfnow.py.Ed Bueler11+122-511
2009-12-20 04:44Added another reference, and very minor mod to way temppabase is computed (i.e. signature only). (Sorry that I am finishing up various bits of work on various machines, before leaving town ...)Ed Bueler3+20-8
2009-12-20 03:27Added reference.Ed Bueler1+10-0
2009-12-20 03:21Start on modularization of SSA, i.e. the move out of IceModel.Ed Bueler3+373-0
2009-12-20 03:12Name change verifynow.py->vfnow.py reflected in script.Ed Bueler1+3-3
2009-12-20 02:09Squashed commit of the following:Andy Aschwanden2+28-1
2009-12-18 20:32Fixed empty -extra_times seg fault (from Andy typo as bug). Tried out Constantines PetscScalar**-free technique in src/eismint/; love it.Ed Bueler3+33-31
2009-12-16 23:18Changes in several variables to improve appearance of diagnostic output. Addition of fill_where_floating() and fill_where_grounded() methods to IceModelVec2. Not-very-elegant attempt to fix bug #14960. Improved (I hope) appearance of std out report on computational domain and grid.Ed Bueler6+83-18
2009-12-16 05:37* The age mechanism now has the right flag, namely pism_config:do_age. Default value is "no"; option "-age" sets it to "yes".Ed Bueler10+36-43
2009-12-15 22:04* Added "/" to the command detecting PETSc 3 in build/Makefile. * Implemented IceModelVec2Mask, a derived class intended to simplify using the mask. Also, removed PismIntMask. PismModMask is used in exactly one place now (vote-by-neighbors case in updating surface elevation and mask). * Added operator() to IceModelVec2, so that it is possible to call vH.begin_access(), for example, and then say "if (vH(i,j) > 0)..." or "vH(i,j) = foo" (i.e. defining a double pointer variable and calling get_array(...) can be avoided in most cases). * Replaced @make with @$(MAKE) in the top-level Makefile.Constantine Khroulev24+212-187
2009-12-15 02:52* Changes to accomplish item 2 in task #6868, the -age option. Tried to make all accesses to tau3 and taunew3 protected by check_option() for -age. (Is this the ideal approach?) Now IceModel does not allocate tau3 or taunew3 without -age set. Very minimal doc of this option for now. * New IceModel methods created for this purpose: countCFLViolations(), energyStats(), ageStats(). Eliminated: energyAgeStats().Ed Bueler20+262-148
2009-12-14 20:58Fixed PISMConstOceanCoupler to use config mechanism.Ed Bueler4+57-54
2009-12-14 08:41marmaduke has more pedantic than you ...Ed Bueler1+1-1
2009-12-14 08:35New attempt to resolve stupid PETSc 3.0.0 clean:: (w double-colon) target. Now there is no clean target in PISM's Makefiles. GNU make apparently has default clean behavior that removes the .o files in build/ If you want to clean dependencies (.d) in build/, do 'make depclean'.Ed Bueler1+6-7
2009-12-14 00:50Squashed commit of the following:Andy Aschwanden1+1-1
2009-12-13 08:26* Removed environment variable PISM_HAVE_PETSC3 from build/Makefile, because PETSc folks have apparently fixed their 'clean::' typo(?). Note PETSC_VERSION_MAJOR is still available for deciding which PETSc API call to make, between 2.3.3 and 3.0.+, when needed.Ed Bueler2+4-8
2009-12-12 00:28Fixed a stupid bug I introduced in the previous revision (PetscGlobalSum arguments cannot be the same.)Constantine Khroulev1+13-12
2009-12-11 21:33* Added more scalar time-series (ice mass, ice volume and mass rates of change, total surface, basal and sub-shelf ice fluxes). * Fixed the tillphi communication bug. * Scalar time-series use linear interpolation and don't affect time-stepping. * Remover IceEnthalpyModel::initFromFile (this means that a penth input (-i) file will have to have the enthalpy variable).Constantine Khroulev13+213-77
2009-12-10 03:58Clear out old/bad scripts from test/ directory.Ed Bueler2+0-379
2009-12-10 02:11MY mistake on SeaRISE-Greenland. The Fausto paper points out that the standard deviation of the PDD (=pism_config:pdd_std_dev) is quite critical in determining mass balance. Somehow we never changed the default away from the EISMINT-Greenland of 5.0 K; Fausto recommends half that (2.53, actually).Ed Bueler2+3-4
2009-12-09 09:54Added Weertman ref to bib. Trivial comment fix.Ed Bueler2+11-1
2009-12-08 03:39Added vfnow.py, the new version of verifynow.py (with the same look-and-feel), moved vnreport.py to vnreport_old.py and vnreport_new.py to vnreport.py. Using vfnow.py shows that test I does work (see bug #14837).Constantine Khroulev4+525-374
2009-12-07 22:17Simplified -force_to_thk options to conform to SeaRISE-Greenland use, and to simplify, and to stop when not possible (i.e. in PISMConstAtmosCoupler). Only controllable parameter is -force_to_thk_alpha.Ed Bueler3+49-49
2009-12-04 21:24Clean-up of all sorts.Constantine Khroulev13+87-53
2009-12-03 20:33While waiting for SeaRISE-Greenland runs, I played with EISMINT-Greenland. Used ts_file and save_file to greatly clean up ssl2.sh and ccl3.sh. Also added options -ts_append and -extra_append to append to files, instead of default move-to-tilde behavior. Also renamed -split_extras to -extra_split for uniformity with all -extra options.Ed Bueler5+85-106
2009-12-02 22:59* Made PISM mask out (using GSL_NAN) diagnostic quantities that are meaningless in ice-free areas. * IceModelVecs can correspond to time-independent quantities now: setting foo.time_independent to "true" will ensure that the corresponding NetCDF variable does not depend on 't' (and does not waste space). Made lat,lon,tillphi,bheatflx,rank time-independent. Note that derived classes can easily override this.Constantine Khroulev6+81-17
2009-11-30 23:50* Fixed software test #3. * Climate couplers write their fields to -extra_file. * Options -ts_times and -extra_times force the model to hit requested times (see task #6871). * All the -extra_vars are written in 'float' (see task #6872). * Minor code clean-up. * More diagnostic quantities: {u,v,w}velbase, tempbase, tempsurf.Constantine Khroulev14+312-76
2009-11-29 00:09Modest work toward item 1 of task #6868: added averagedHardness() method to IceFlowLaw in preparation for the item 1 optimization, and added compute_hardav() to show the barB(T) value in the output file as a diagnostic 2D variable. Also removed integratedStoreSize(),integratedStore(),integratedViscosity(), which were not called because of the removal of src/base/ssaJed/* in r832; integratedStore() did essentially what averagedHardness() does now.Ed Bueler5+95-69
2009-11-25 21:47Fixed the pismr -boot_from -pdd segmentation fault bug (I think).Constantine Khroulev2+4-3
2009-11-25 07:41Code to read -force_to_thk_alpha added to src/coupler/pccoupler.cc. (Sorry. I realize I had not committed it.)Ed Bueler1+27-6
2009-11-25 07:13Cleaning out old, useless code; python script rossplot.py does fine.Ed Bueler3+0-337
2009-11-24 21:26Fixed EISMINT-Ross scripts (they still had the "transpose" in them).Constantine Khroulev3+17-17
2009-11-24 00:14Finally implemented and tested future climate forcing using snow temperature and snow precipitation anomalies (options -anomaly_temp_ma and -anomaly_precip). Works with PISMGreenlandAtmosCoupler only.Constantine Khroulev4+247-219
2009-11-20 20:23* Moved start_year and end_year parameters to IceGrid. * Removed PISMAtmosphereCoupler::get_force_to_thickness_time_from_options() * Modified IceModelVec code so that having nested begin_access()... end_access() pairs with the same IceModelVec is OK.Constantine Khroulev20+132-176
2009-11-20 05:50Added reference. Slight change to comments showing example of -force_to_thk.Ed Bueler2+10-2
2009-11-20 03:54-force_to_thk mechanism functioning and documented; see comments on PISMAtmosphereCoupler::updateSurfMassFluxandProvide() for example based on EISMINT-GreenlandEd Bueler2+70-64
2009-11-19 17:57* Updated software tests. * Minor fix in pGreenlandAtmosCoupler.cc * Undid the "fundamental transpose" in most places. Pros: x, y, u and v have consistent meanings everywhere and we still use the array[x][y] indexing. Cons: diagnostic viewers are transposed (oh well) and one has to be careful when using DALocalInfo structs.Constantine Khroulev9+54-71
2009-11-19 05:40Fixes to comments to explain changes made in r861.Ed Bueler2+67-4
2009-11-19 04:08Major: implement -force_to_thk mechanism but it causes a seg fault currently; example use on EISMINT-Greenland files is (pismr -y 1 -ys 0.0 -i green20km_Tsteady.nc -pdd -force_to_thk green_SSL2_250k.nc). Minor: Fixed basal water pressure calculation. Fixed units (need seconds) on timestep restrictions from couplers, inside IceModel::run. Resolved timestep restrictions from coupler base class in PISMGreenlandAtmosCoupler::max_timestep(). Massaged appearance of pcctest stdout.Ed Bueler7+254-44
2009-11-17 23:59* Added dhdt (surface elevation rate of change; feel free to remove it to save space, or for any other reason) and bwp (basal pore water pressure) diagnostic variables to the PISM output. * Fixed bug #14780. * IceModel::run() asks couplers and modifies maxdt_temporary if necessary. * Got rid of some compiler warnings on midnight.Constantine Khroulev11+239-109
2009-11-17 03:05Change to PISMAtmosphereCoupler::writeCouplingFieldsToFile(): vsurftemp now written as is, without attempt to shift and shift back. This matches semantics of vsurftemp=artm in other places: it is updated by updateSurfTempandProvide() and that includes adding a -dTforcing shift to it. The PISMConstAtmosCoupler should probably not allow -dTforcing because it is hard to know if artm read from file was or was not shifted.Ed Bueler1+5-8
2009-11-16 00:39* Modified software test #11 to use equal vertical spacing. * Made IceModelVec::destroy() protected: ~IceModelVec() calls it when necessary, so calling it "by hand" is unnecessary. * Implemented IceModelVec2T::average() (using the trapezoidal rule). * Modified some IceModelVec2T code to get rid of pedantic compiler warnings.Constantine Khroulev13+102-106
2009-11-14 03:25Documentation only. Added two refs, fixed one, fixed poorly-written (by me) comment in PISMGreenlandAtmosCoupler.Ed Bueler2+41-18
2009-11-13 20:06Added doc/userman/Makefile I forgot to svn add before.Constantine Khroulev1+17-0
2009-11-13 20:04Started reorganizing the User's Manual and the doc/ subdirectory in general.Constantine Khroulev62+2604-2603
2009-11-13 17:54* Removed the code allowing Chebyshev vertical grid spacing in the ice. * Command-line option change: -quadZ, etc are gone; '-z_spacing quadratic' sets the quadratic spacing in the ice (default for IceModel and all derived classes except IceCompModel (to preserve refinement paths)), '-z_spacing equal' sets equal spacing. The option '-zb_spacing' sets bedrock vertical spacing (quadratic is the default here, too). Also, fixed a bug that caused equal spacing to be the default in the ice. * Documented -z_spacing and -zb_spacing and replaced -quadZ with -z_spacing everywhere.Constantine Khroulev12+83-143
2009-11-12 20:32* Fixed the pisms -i foo.nc -eisII ... initialization bug. * Created the object needed to read and access snow precipitation maps for climate forcing (in pGreenlandAtmosCoupler.cc). This mechanism is still broken, though.Constantine Khroulev8+130-76
2009-11-12 03:21Added Burgess et al. 2009 reference and corrected alphabetical order.Ed Bueler1+47-38
2009-11-12 00:27* Added the climate_forcing_buffer_size configuration variable. Current default is 60 (5 years of monthly records). * Replaced MonthlyDataMaps with an IceModelVec2T in the PISMGreenlandAtmosCoupler. Renamed -ppd_monthly_temps to -pdd_snow_temps. (Rationale: new behavior requires a new name.) * More testing. * Removed monthlyDataMaps.{cc,hh} because they are no longer used. * Documented IceModelVec2T.Constantine Khroulev10+166-396
2009-11-11 20:43More work on future climate forcing. Most code is written and tested now. Next step: modifying the climate coupler.Constantine Khroulev3+119-39
2009-11-10 20:47* Added a new variable (rank) to output files. * IceModelVec2T (a class for reading and accessing 2D climate inputs like monthly temperatures) is mostly done. Needs more testing and some optimization. Note that it *is not* used right now. * Implemented the (empty and inactive) PISMClimateCoupler::max_timestep(...) * Added the report_range boolean to LocalInterpCtx (sometimes it's nice to be able to turn it off) * Updated software tests.Constantine Khroulev16+536-8
2009-11-10 03:32Minor stuff in documentation. Minor stuff in reporting to stdout: -usr1 causes message that mentions flush of time-series.Ed Bueler5+19-19
2009-11-09 22:40In IceEnthalpyModel, fixed actual bug part of bug #13957. It makes almost no difference for Greenland runs, which also do not show 'wall' of Hmelt=0 at grounding lines on 3km grid calculations, but connection to choice of K in diffuseHmelt() is acknowledged. Also, trivial massage to stdout for regridding.Ed Bueler2+4-4
2009-11-06 08:22Precipitation for PISMGreenlandAtmosCoupler now has offset from formula at SeaRISE Model Initialization page. No further issues known with coupler, for SeaRISE-Greenland.Ed Bueler4+47-7
2009-11-06 00:19Changed arguments of the \ref doxygen command so that all references are resolved.Constantine Khroulev11+18-18
2009-11-05 03:031. Refactor of main derived class of PISMAtmosphereCoupler: - namely, PISMSnowModelAtmosCoupler is renamed PISMGreenlandAtmosCoupler - it is put into separate file src/coupler/pGreenlandAtmosCoupler.[hh|cc] - it has the SeaRISE-chosen Fausto et al (2009) parameterization for temperature, as before - EISGREENAtmosCoupler is still a derived class of it - monthly temperature inputs presumably broken (but that mechanism should change immediately to be SeaRISE future climate forcing case anyway)Ed Bueler11+736-628
2009-11-03 00:13* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bib* Added util/pism_config_editor.py. We'll see if it's of any use. * Documented -config_override * Minor fixes in ice_bib.bibConstantine Khroulev4+263-8
2009-11-02 20:36Added a ref, and corrected a ref.Ed Bueler1+24-12
2009-11-02 00:29Clean up. Put verification tests, and their tests, in src/verif/tests/, with a local Makefile. This means 'simple...' executables don't clutter up build/Makefile. Moved tryLCbd into src/earth, where it belongs.Ed Bueler34+51-19
2009-10-30 22:16Minor fix (I forgot to rename earth_gravity to standard_gravity in enthalpyConverter.cc).Constantine Khroulev1+1-1
2009-10-30 21:06Finished task #6774.Constantine Khroulev6+53-3
2009-10-30 19:56* Added doxybib.dbj, a batch file used to create doxybib.bst (so that we can change some aspects of reference formatting later). * ice_bib.bib: moved dois to separate bibTeX entries (doi = {...}). Moved urls to the "note" entry in all records that don't use "note" for something else (this way the URL shows up with bibTeX styles that don't support the "url" entry). * Renamed IceType to IceFlowLaw and PlasticBasalType to IceBasalResistancePlasticLaw (see task #6642). I don't think they need to have a common base class, though. * Removed DeformableEarthType and added lithosphere_density, lithosphere_flexural_rigidity and mantle_viscosity variables to pism_config.cdl.Constantine Khroulev22+3355-2835
2009-10-30 00:18Changed (slightly) the way references are formatted.Constantine Khroulev1+838-472
2009-10-29 22:19* Added -pc_type asm -sub_pc_type lu to options used by test/verifynow.py for test J. Will close non-bug #14629. * Switched to using standard gravity and ideal gas constants set in pism_config.cdl * Switched to M_PI defined in gsl/gsl_math.h instead of hard-wiring it in PISM. * Switched to using GSL_NAN instead of PISM_DOUBLE_NAN. * Removed IceModel::bootstrapLIC, because it is not used any more. * Some code clean-up.Constantine Khroulev27+97-100
2009-10-29 00:09* Made doc/config_doc.py sort configuration parameters. * Removed BedrockThermalType, SeaWaterType and FreshWaterType. * Moved a lot of hard-wired constants to pism_config.cdl. This should fix bug #14043 but might break EISMINT (the old value of beta_CC_grad is *gone*). * Added examples/pism_matlab.mConstantine Khroulev21+210-164
2009-10-28 03:11User Manual updates while waiting for jobs.Ed Bueler1+35-30
2009-10-27 20:57Added ice softness and Glen exponent to pism_config.cdl; fixed beta_CC_grad initialization in IceType::IceType().Constantine Khroulev2+10-4
2009-10-27 20:13* Some work on task #6642: Constructors of IceType and its derived classes have a const reference to NCConfigVariable as an argument, so one can call config.get("parameter") (and get_flag("flag")) in a constructor of an ice type. * Removed src/base/ssaJed/* * Removed the "-law" command-line option. * Minor fixes in test/vnreport_new.pyConstantine Khroulev24+170-2137
2009-10-26 20:28Added two Greenland precipitation references to bibliography. Clarified long name of vsurftemp in PISMAtmosphereCoupler.Ed Bueler2+27-3
2009-10-26 18:09Fixed bug #14355 in the trunk by making PISMClimateCoupler::findPISMInputFile() report whether we need to regrid. (I.e. if the file found was the -i file we can just read data straight.)Constantine Khroulev3+86-25
2009-10-25 06:36Work on User Man text.Ed Bueler3+75-77
2009-10-24 20:56User's Manual: new figures showing results of verification tests B and G.Constantine Khroulev10+54-23
2009-10-23 20:18More work on vnreport_new.py (should be usable by now), new test I & K figures for the User's Manual.Constantine Khroulev11+49-20
2009-10-23 17:55More work on the new vnreport; NetCDF error reporting in IceExactSSAModel.Constantine Khroulev3+152-32
2009-10-22 23:25* Added test/vnreport_new.py (work in progress). * Removed the -max_dt agrument from the test K command in verifynow.py. * Some fixed related to writing error reports to a NetCDF file.Constantine Khroulev3+169-8
2009-10-22 20:29* Added the -r (--report_file) option to test/verofynow.py * Added #include <algorithm> to NCVariable.cc. This should fix the r822 compilation error on pingo. * Added the "pism_" prefix to the "ssa_velocities_are_valid" global attribute.Constantine Khroulev5+14-7
2009-10-22 19:33* Implemented min,max reporting of forcing time-series (see task #6798). * IceCompModel writes error reports in NetCDF now (try 'pismv -test G -report_file foo.nc')Constantine Khroulev6+188-15
2009-10-21 18:23* Code clean-up and a helper-function: username_prefix() creates a string like "user@hostname 2009-10-21 10:10:44 AKDT: ". * Wrote NCGlobalAttributes, a class for reading and writing NetCDF global attributes. Note that this breaks reading initial guesses of SSA velocities from old PISM files: the ssa_velocities_are_valid global attribute should have the value of "yes", "true" or "on" for that to work now. * Moved projection parameters out of IceModel and into pism_config.nc. Note that PISM will not recognize the old variable we used to store them (polar_stereographic). Please rename "polar_stereographic" to "mapping" in your NetCDF files.Constantine Khroulev17+237-153
2009-10-20 18:00Added a pism_wait_for_gdb(), a debugging-helper function.Constantine Khroulev3+34-1
2009-10-19 22:07Some code clean-up and refactoring to make sure that PISM (trunk) closes all the files it opens (see bug #14541).Constantine Khroulev5+143-115
2009-10-17 02:24Fixed un-initialized allowAboveMelting flag bug in r816.Ed Bueler3+4-2
2009-10-17 00:59Typo fixed in reference. doc/Makefile cleans revision.tex with 'make clean'.Ed Bueler2+8-2
2009-10-17 00:35* Fixed bug #14152.Ed Bueler12+333-572
2009-10-17 00:21Added reference. Noted pcctest illustration of surface mass balance over time in EISMINT-Greenland, as good example; needs improved text.Ed Bueler2+17-1
2009-10-16 17:16Fixed bug #13904 in the trunk. After some more thinking I agree with Ed's formulation (used in IceModel::writeSSAsystemMatlab()); coordinates need not be symmetric with respect to x0 and y0, so going from x0-Lx to x0+Lx-dx and from y0-Ly to y0+Ly-dy is just fine.Constantine Khroulev3+48-7
2009-10-14 23:17Boring nothings: (1) Makefiles delete revision on 'make clean'; re-made by 'make', of course, (2) computational parameters msg from PISM runtime stdout clearer.Ed Bueler3+3-3
2009-10-14 17:29* Got rid of -view_big, replaced with -viewer_size [number]. All viewers are of the same size now. * Finished implementing and documenting viewers (including soundings, which were broken before now). * Rewrote IceModel::regrid(): -regrid_vars takes variable names now! (such as temp,age,...,enthalpy). Only model_state variables can be regridded. Removed IceEnthalpyModel::regrid(), because it is no longer necessary. * Updated software tests to use -regrid_vars correctly.Constantine Khroulev15+295-498
2009-10-12 19:40Re-implementedConstantine Khroulev13+289-167
2009-10-12 04:07* Added -usage to all drivers, because I couldn't remember how pcctest works. Example below. Note -usage causes driver to print message and quit. Option -help adds the usage message to all the mess you normally get. These usage messages DO NOT (and cannot credibly) give all PISM options, but should/do include options special to the executable. Task #6493 still applies.Ed Bueler9+265-113
2009-10-09 21:40* Some more work on the User's Manual, especially the list of available diagnostic quantities * Diagnostic viewers are back (almost). Please see the User's Manual (table 10) for the list of variable names; they match ones used in PISM output files. New interface: -view_map <list of variables> turns on map-plane viewers of 2D and 3D ice quantities. (This shows surface values of 3D quantities.) -view_slice <list of variables> turns on "horizontal" slice viewers (ice quantities, at a certain level above the base of ice). The option -slice_level specifies this level, in meters. Finally, -view_big <list of variables> specifies which viewers should be big. If the same variable shows up more than once (as in "-view_map temp -view_slice temp", you can use "-view_big temp_map" to make sure that the surface view (and not the slice) is big. Similarly, "-view_big temp_slice" tells PISM to make the slice viewer big. Not implemented so far: soundings, nuH viewers, diffusivity and the special strain heating pismv viewer, viewers with logarithmic color scales, viewers of bedrock quantities. Also, IceCompModel needs more work. * Renamed -save_at to -save_times, -save_to to -save_file and added deprecation warnings. (For consistency.)Constantine Khroulev15+307-102
2009-10-09 06:39Minor edits of User Man discussions of -save_to and -ts_file and -extra_file usage.Ed Bueler1+38-23
2009-10-09 01:56* Updated software test #9 (it needed the -Lbz option). * Removed most of the old code dealing with diagnostic viewers. Diagnostic viewers are officially broken now, but will probably come back tomorrow. * More work on new diagnostic viewers.Constantine Khroulev22+170-653
2009-10-09 00:06* updated verifynow.py to use quadratic spacing by default * PISM's default vertical spacing is quardatic now. Need to come up with command-line option names for turning it off (or selecting equal spacing). Should we also drop Chebyshev vertical spacing? * Some work on the User's Manual. * Some work on diagnostic viewers. * Fixed a minor bug: PISM should require -Lbz if -Mbz is set and Mbz > 1.Constantine Khroulev8+184-44
2009-10-07 19:43* Removed MATLAB variable output. (Except for the SSA system stuff.) * Major changes in the way vertical spacing is computed: now bedrock thickness and spacing is specified using both -Mbz and -Lbz. Lbz has to be zero or positive; Mbz >= 1. Default bedrock vertical spacing is quadratic; -no_quadZ_bed allows using equal instead.I.e. users have as much flexibility as with vertical spacing in the ice now. The mechanism computing fine equally-spaced vertical grids used in temperature and age calculations is better: it guarantees matching spacing at the ice/bedrock transition. (We allow the ice computational grid to be taller than the storage grid and extend into the air to get this matching.) Verification tests ABCDEFGHIJK report the same errors as before; test K shows near-quardatic (O(dz^2)) convergence with both equal vertical spacing and -quadZ. See task #5874. * Updated verifynow.py to use -no_quadZ_bed. * Some minor fixes.Constantine Khroulev14+193-602
2009-10-06 20:55Added/fixed references. Improved EISMINT-Greenland steady state script.Ed Bueler2+38-11
2009-10-05 18:33Revisions 796 through 799 fail software test #9 (regridding from files with different variable orders). I was happy to delete NCTool::transpose() and replace it with a different NetCDF library call to read and write IceModelVecs. Revision 800 passes test #9.Constantine Khroulev5+68-142
2009-10-03 19:54* Fixed some proposed standard names using suggestions from Jonathan Gregory * Added dt (mass continuity time-step) to time-series outputs. * Fixed floating ice area calculation.Constantine Khroulev12+54-24
2009-10-03 03:45Put default values, used when variables are missing in bootstrap file, into pism_config.nc. This simplifies derived class like IceGRNModel which alters the bootstrap defaults.Ed Bueler6+57-62
2009-10-02 02:10* Added grounded ice area (iareag) and floating ice area (iareaf) time-series. * Wrote timestamp(), a function that creates just that. It is used to build the history string. * IceModel::endOfTimeStepHook() flushes all the time-series buffers now, so that kill -USR1 pismr allows the user to look at time-series collected so far. * Changed the way bedrock spacing is set with command-line options: -Lbz turns on quadratic bedrock spacing, -Mbz turns on equal bedrock vertical spacing. -Lbz and -Mbz together are not allowed. The idea is that quadratic vertical spacing in the bedrock always goes with -Lbz and equal -- with -Mbz.Constantine Khroulev11+141-82
2009-10-01 02:05Significant: * changed setting of Hmelt in floating regions to be equal to maximum value (2.0 m), not 0.0; should fix wobble at grounding lines seen by Maria, and possibly help with upstream velocities * fixed setting of temperature within the bed for EISMINT II experiments, in the non-standard cases where bedrock is present; this was an initialization sequencing problemEd Bueler8+134-140
2009-10-01 01:43 - updated series.py: series.py -h/--help prints help screenAndy Aschwanden1+27-26
2009-09-30 21:15* Finished task #6743 and task #6643 (saving scalar time-series, including ice volume). Usage: pismr -i foo.nc -y 1000 -o bar.nc -ts_file time-series.nc -ts_times 0:10:1000 -ts_vars ivol,iarea This code uses linear interpolation to save exactly at times requested by the user. * -quadZ turns on quadratic vertical spacing in bedrock. -no_quadZ_bed overrides; -quadZ_bed turns it on without -quadZ.Constantine Khroulev10+206-74
2009-09-29 17:36* The units testing executable, pismtests, can actually be used to test IceModelVec3 and IceModelVec3Bedrock. * Fixed some typos in the Installation Manual. * Implemented uneven (quadratic) vertical bedrock spacing. It can be turned on by specifying -quadZ_bed and -Lbz. PISM computes the number of bedrock levels necessary to have matching spacing at the top of the bedrock and the bottom of the ice and desired Lbz. Note: this code needs testing! * Implemented another feature similar to snapshots: 'pismr -i foo.nc -y 1000 -extra_file bar.nc -extra_times 0:50:1000 -extra_vars csurf,cbase,cbar -o quux.nc' will save snapshots of csurf,cbase and cbar on the first time-step after requested times to bar.nc. One can also add -split_extras (similar to -split_snapshots). * Some more work on saving scalar time-series. * Minor code clean-up.Constantine Khroulev23+715-247
2009-09-24 18:29User Manual massage. Updated a ref.Ed Bueler2+20-25
2009-09-22 17:01Added IceModelVec2::create(...) that allows creating IceModelVec2s with a custom stencil width.Constantine Khroulev2+11-6
2009-09-22 00:45* Created a new derived class IceUnitModel, for unit tests. It is mostly empty and not ready to use now. Moved IceModel::testIceModelVec3... there. * Created a new executable for unit testing (pismtests). * Very minor code clean-up.Constantine Khroulev8+288-197
2009-09-18 00:06I got somewhat side-tracked, but still. These structural changes will make life easier down the road.Constantine Khroulev26+662-446
2009-09-15 19:29* moved IceModel::getMagnitudeOf2dVectorField(u,v) to IceModelVec2::set_to_magnitude(u,v) * cleaned up the snapshot-writing code * moved hasSuffix(...) out of IceModel and into pism_const.cc (as ends_with(...)) * got rid of IceInfoNeededByCoupler (replaced by a dictionary-based PISMVars). Lots of changes are related to this. * minor code clean-up and fixes in the User's Manual * some work on the code saving scalar timeseriesConstantine Khroulev42+780-477
2009-09-14 01:34Squashed commit of the following:Andy Aschwanden1+2-1
2009-09-09 19:07More code clean-up. In particular, removed HEINO (src/ismip/heino/*) and the thermalBedrock boolean flag (which was not used since some early revision).Constantine Khroulev32+287-1491
2009-09-08 20:10Code clean-up.Constantine Khroulev38+364-384
2009-09-08 02:48References used in Karthaus lectures.Ed Bueler1+37-1
2009-09-02 07:11Minor changes, including a ref.Ed Bueler2+15-2
2009-09-02 07:08Added references.Ed Bueler1+47-2
2009-08-28 23:48Added references.Ed Bueler1+35-10
2009-08-26 06:19Reference mods.Ed Bueler1+57-10
2009-08-24 18:46Squashed commit of the following:Andy Aschwanden2+16-8
2009-08-20 18:59Squashed commit of the following:Andy Aschwanden2+29-6
2009-08-19 16:58Squashed commit of the following:Andy Aschwanden17+76-25
2009-08-05 13:45Squashed commit of the following:Andy Aschwanden1+1-0
2009-08-03 16:29Squashed commit of the following:Andy Aschwanden10+33-33
2009-08-03 11:59Squashed commit of the following:Andy Aschwanden3+16-9
2009-08-02 11:22Squashed commit of the following:Andy Aschwanden2+3-3
2009-07-31 13:19A fix for bug #14043: overwrite ice->beta_CC_grad with value from pism_config.cdl However there are still discrepancies between pismr and penth -cold but right now I'm not able to tell whether these are numerical issues or if there's still something fundementally wrong. The fix looks the following:Andy Aschwanden1+20-1
2009-07-29 17:52Fiddles with docs about PISM use of standard names.Ed Bueler1+18-12
2009-07-17 21:05* Modified pore water pressure model in IceEnthalpyModel. Basal melt rate contributes, but this contribution is highest under thick ice and is reduced by a factor for thinner, presumably marginal, ice. That is, ice thickness is used as surrogate for closeness to the margin, which is presumed to control basal drainage. * Parameters to control above model are set by options, with these defaults -bmr_enhance FALSE -bmr_enhance_scale 0.1 m/a -thk_eff FALSE -margin_reduced 0.97 -margin_H_high 2000.0 m -margin_H_low 1000.0 m * Usage example: Using both the basal melt rate enhancement and the thickness effect, but keeping the default constants, looks like penth -bmr_enhance -thk_eff Just penth uses the older model which only connects pore water pressure to the time-integrated and capped water layer thickness bwat=Hmelt. * Added stdout report from IceEnthalpyModel::setFromOptions(), to show state of IceEnthalpyModel data members (ones not in IceModel).Ed Bueler2+115-21
2009-07-17 20:14Slight cleaning on bootstrapping messages to stdout.Ed Bueler1+3-6
2009-07-16 09:56Small docs fixes/additions.Ed Bueler2+14-2
2009-07-16 02:17The r762 change to IceModel::updateSurfaceElevationAndMask() omitted some code. This is the fix.Ed Bueler1+4-0
2009-07-16 02:15Port of r762:763 stable0.2 changes to trunk.Ed Bueler1+2-0
2009-07-14 05:59* A bit of cleaning of trunk. IceModel::updateYieldStressFromHmelt() was modifying the mask and this moved into IceModel::updateSurfaceElevationAndMask(), where it should be. Reduces mask communication and cleans up iMbasal.cc. * IceEnthalpyModel cleaned up to reflect better base class behavior, on same issue.Ed Bueler4+175-57
2009-07-14 02:16* Segregation of SIA sliding from good sliding. In particular, IceModel::basalVelocity() renamed to basalVelocitySIA(), and moved to iMSIA.cc. * Slight documentation improvements in iMbasal.cc. * README in src/verif/fortran cleaned up.Ed Bueler9+107-94
2009-07-12 18:31* EnthalpyConverter: + Changed signatures to reflect better error handling. Those that can produce errors return PetscErrorCode and their results are pass-by-ref. Others typically are functions. + Added isTemperate() which returns boolean (E > E_s(p)). + Added getWaterFractionLimited() and getEnthAtWaterFraction(). These are the right forms to use inside the drainage model, where excessive enthalpies (E > E_l) should be tolerated but should lead to high basal melt rates followed by resetting enthalpy to maxed-out temperate value: E = E_s(p) + 0.01 L * Changes to IceEnthalpyModel to use changed EnthalpyConverter. Also some error handling cleanup there. * Trivial stdout change to iMIO.cc.Ed Bueler4+206-138
2009-07-10 23:48I think I have dealt with the weird temperatures in bedrock in the E Greenland ice sheet fjords area. The underlying issue is that the model must deal stably with totally bizarre bed geometry *from the point of view of shallow ice theory*. (The particular case which triggered flaws involved a 500m change in bed topography in 10km horizontal, plus an upstream high basal meltrate with cold ice base downstream, but only in transient sense, as it was resolved later in the run *if* you didn't check for low bedrock temps.)Ed Bueler6+113-83
2009-07-10 10:07* Trivial editing in BOMBPROOF docs. * Changed error reporting in some non-PetscErrorCode reporting functions, to use PetscErrorPrintf(). * Added T > 0 check for absolute temperatures in enthalpyConverter.cc. * Draft of bedrock-only re-solve attempt to fix temperature highs in bedrock; currently commented out. * Returned to having conductivity in temperate ice at ice/bedrock interface.Ed Bueler7+337-88
2009-07-09 08:50I think that a "silent hang" occurring in IceEnthalpyModel is from two sources: (i) ice/bedrock interface physics which lead to high bedrock temps when high frictional heating and (ii) mis-handling of error messages.Ed Bueler7+171-21
2009-07-08 08:38* Added scaling factor to trivial rows of SSA system. Improves condition number of matrix, but no noticable change in accuracy of solutions. Possibly redundant with preconditioning (i.e. diagonal preconditioner might automatically scale appropriately). Zero cost change, however. * Fixed little-used code for -ssa_matlab output of A, x, b for Ax=b system in SSA.Ed Bueler2+20-15
2009-07-07 02:20* Cleaned metadata and short names for liquid fraction, basal melt rate, basal frictional heating, and strain heating. * Added diagnostic output of basal melt rate and frictional heating in IceEnthalpyModel. * columnSystemCtx will only write full matrix if system is small; otherwise writes by diagonals, as stored. * Trivial comment change in Timeseries.hh. * Working to resolve unreasonably high bedrock temps, close to ice, seen in Greenland runs. These come from very high frictional heating, I think, but there is a smallish jump in temps anywhere temperate ice contacts bedrock. The smallish jump has magnitude 0.01 * L / k_b, essentially, namely the temperature jump corresponds to the latent heat of the liquid fraction.Ed Bueler6+97-43
2009-07-06 06:21Made destructor virtual for PISMOceanCoupler.Ed Bueler1+1-1
2009-07-06 03:39Removed test program src/verif/gridL.c, which was clutter. (Sorry, Jed, that you spent a minute changing comments to C90 compliance. Thanks for that sweep through the code, though ...)Ed Bueler3+3-201
2009-07-05 23:27Only call KSPSetFromOptions once, not every solve.Jed Brown3+1-3
2009-07-05 23:00Clean up comments in C source.Jed Brown5+40-34
2009-07-05 22:39Update CMake build system. Put PISM_DEFAULT_CONFIG_FILE in a separate file like PISM_REVISION, so that all of PISM does not need to be rebuilt if it changes. This sort of hard-coding of paths does *not* work with installation. It should be replaced with a function that returns an appropriate config file, after consulting environment variables and the location (perhaps installed) of the PISM libraries in use. I have not done this at this point because I don't want to intrude, let me know if you want me to do it.Jed Brown7+39-15
2009-07-05 20:15Added wall clock time (i.e. date and time) to history line written by snapshot. Helpful for long runs killed by batch system.Ed Bueler1+14-3
2009-07-01 08:00* Factored out energy and age related statistics calculations, previously in spagetti code in IceModel::summary(). Now IceModel::energyAgeStats() does the job more clearly. * Documented summaryPrintLine() a bit better. * IceEnthalpyModel::energyAgeStats() overrides base class version to use "E >= E_s(p)" criterion for melted base. Closes task #6653.Ed Bueler4+199-72
2009-07-01 02:14Cleaned up appearance of report on computational box and grid cell dimensions at start.Ed Bueler1+15-18
2009-06-28 01:06* Trivial section heading changes in BOMBPROOF docs. * Added bootstrapFromFile() to IceEnthalpyModel. Sets enthalpy to cold ice value based on temperature field set by IceModel::bootstrapFromFile(). Tested on SeaRISE-Greenland. * Where there is no ice and where the mask is FLOATING, set z=0 level of enthalpy in ice to air value; above that it is already air value. * In ice-bedrock interface segment, always use cold ice value for conductivity in ice. Fixes weirdness seen in SeaRISE-Greenland runs. Thus conductivity is zero for temperate ice except in the 10m (typical) near bedrock, where we again conduct (= diffuse enthalpy).Ed Bueler4+72-55
2009-06-27 08:37A couple of references added to bibliography.Ed Bueler1+26-0
2009-06-27 01:37* Temperate case for ice/bedrock interface segment fixed, I think. * Documentation on the computation done by enthSystemCtx::solveThisColumn() is finalized. (Andy: you can print out the BOMBPROOF doc page and check it against the solveThisColumn() code. Nontrivial.) * Trivial rearranging and doc cleanup in EnthalpyConverter. * penth takes -cold option, which should then do the same as pismr except produce all the output fields usually produced by penth (and IceEnthalpyModel). So 'penth -cts' and 'penth -cts -cold' can be compared in most scripts. * Most debug clutter cleared out of IceEnthalpyModel etc.Ed Bueler6+280-122
2009-06-26 19:15Squashed commit of the following:Andy Aschwanden4+60-0
2009-06-26 02:46Got it: testK succeeds in IceEnthalpyModel. Docs still need some special cases.Ed Bueler4+151-124
2009-06-25 08:14Yet more incomplete work. But really close.Ed Bueler4+393-175
2009-06-24 23:48Ongoing, incomplete work on documenting-before-coding the enthalpy scheme.Ed Bueler5+540-7
2009-06-24 09:18* Cleaned up signatures in trivial ways to quash warnings in doxygen-erating. * Simplified enthalpy normalization in bedrock; not tested or expected to run well, but we know the path of righteousness to follow. * More work on BOMBPROOF-for-enthalpy docs. My plan is to fully document the ice/bedrock interface change before actually implementing it in enthColumnSystem.cc.Ed Bueler7+170-96
2009-06-23 09:26Got it. That is, I know how to proceed on the ice-bedrock interface, with respect to an enthalpy formulation there. Created new BOMBPROOF page for notes. Will write notes first and modify enthColumnSystem.cc and enthalpyConverter.cc when I like the notes.Ed Bueler4+435-1
2009-06-23 02:39Added tools to columnSystemCtx to help diagnose issues in IceEnthalpyModel. Use options '-view_sys -id N -jd M' to see system at map-plane (i,j)=(N,M).Ed Bueler5+141-13
2009-06-23 00:26Squashed commit of the following:Andy Aschwanden1+1-1
2009-06-22 17:34Squashed commit of the following:Andy Aschwanden2+7-3
2009-06-21 22:03* r731 didn't compile even for me. Sorry for the stupidity. Different method for quashing 'unused parameters' warnings; compiles for me at least ... * Changed LocalMassBalance init, to hand in a config without using MPI calls or an IceGrid.Ed Bueler8+35-44
2009-06-21 20:24Quash pedantic compiler warnings. (Now I have set PISM_USE_DEBUG=1, so that I see them on my development machine. The compiler defaults on ARSC machines are pedantic/extras, you see.)Ed Bueler6+17-17
2009-06-21 19:48Trivial change to user manual to get figure numbers right.Ed Bueler1+3-3
2009-06-21 00:19Added simpler diagnostic view methods to ColumnSystem and EnthalpyConverter, in an attempt to find error made obvious by Test K, in penth. (pismv, pismr do fine).Ed Bueler5+105-7
2009-06-20 21:14Added #include <algorithm> to Timeseries.cc to address bug #13731.Constantine Khroulev1+1-0
2009-06-20 02:19Completion of unfinished business in r725. Documented the NCO route to alternate config.Ed Bueler3+26-9
2009-06-20 01:08* Added documentation and cleaned up verbose output related to using and viewing alternate configuration (-config foo.nc). * Used dynamic_cast correctly to avoid constructing multiple IceModel derivatives in pisms.cc and pismd.cc. Note pismd is deprecated anyway.Ed Bueler6+70-41
2009-06-19 09:01* Fixed one substantive bug re ice-bedrock interface. * Nonethless still not satisfied with bedrock temp behavior. * I think we can use Test K to test IceEnthalpyModel, even fairly easily. * Some additional code hardening and comment corrections. * Parallel communication should be fine; report right away if processor- dependence for results of sample runs, lines at top of src/penth.cc.Ed Bueler6+107-90
2009-06-19 02:33More refactoring on IceEnthalpyModel and related tools. No sign of processor-dependence bug, but there is a problem with positive thickness bedrock layer. New version does conversion from temperature to enthalpy for bedrock outside of EnthColumnSystem.Ed Bueler10+427-340
2009-06-19 01:40* Removed src/coupler/forcing.*, because it is not used any more. * Implemented Timeseries::write() and the (new) DiagnosticTimeseries class. The latter makes it fairly easy to write time-series to NetCDF files (it does not write every time you add an entry, avoiding lot of I/O and does not store more than a given number of entries, reducing memory use). Note that this code is tested, but is not used (yet). Please see the DiagnosticTimeseries doc string or the class browser for details.Constantine Khroulev9+181-574
2009-06-18 05:39Replaced 'static const' with 'const'. Fixes penth bug on at least one machine.Ed Bueler1+33-17
2009-06-17 21:54Changed to new enthalpy normalization following Aschwanden-Bueler conversation 6/16. Many other clean-ups. Definitely IceEnthalpyModel has a parallel communication bug.Ed Bueler7+182-152
2009-06-16 23:24* Completed task #6641 * Fixed bug #13413, switched over to the new Timeseries object for climate forcing and IceROSSModel, changed eiscore.py to produce time-series suitable for linear interpolation * Minor code cleanup, typos, etcConstantine Khroulev14+136-118
2009-06-16 18:41Added ability to regrid enthalpy with 'y' flag: '-regrid_vars y'.Ed Bueler3+74-1
2009-06-16 08:35* Trivial appearance changes to penth.cc, pccoupler.cc. * Addition of -temp_pa, which writes pressure-adjusted temperature at end of run, in deg_C. * Cleaning, finishing enthalpy method. * I still need reassurance on functions in enthalpyHelper.hh. Does temperature of a chunk of cold ice change just as you change the pressure?Ed Bueler9+223-102
2009-06-15 02:49IceEnthalpyModel *much* betters. Runs without producing nonsense for anything. Still has many 'FIXME' in core routines. Andy: we can talk about these and finish up, I think.Ed Bueler4+115-137
2009-06-14 03:35Finally I have written all the code for the first time for IceEnthalpyModel. Does not work right yet, of course, but compiles produces some numbers.Ed Bueler6+233-108
2009-06-13 21:46Improvements in exposition of standard_names table. Removal of obscure comment in iMgeometry.cc; if we want to know about calving laws we should ask PIK.Ed Bueler2+12-45
2009-06-13 08:28Small changes to IceEnthalpyModel.Ed Bueler4+36-85
2009-06-12 09:21Yet more work on IceEnthalpyModel, mostly of code-hardening nature. Actual system of equations (enthColumnSystem) still not even initially implemented. Clean-up sweep of BOMBPROOF doc, but with no content/intent changes there.Ed Bueler8+358-323
2009-06-12 02:29add comment/questions regarding boundary conditionsAndy Aschwanden1+52-52
2009-06-12 01:13More work on IceEnthalpyModel, including an implemented drainage law.Ed Bueler9+1001-869
2009-06-11 07:29More work on IceEnthalpyModel. Worth talking about at this point. Split iceEnthalpyModel.cc into two parts, but won't grow much more in length. Starting to see benefits of conserving energy through keeping track of energy content (i.e. enthalpy) instead of temperature. Necessary to duplicate code for SIA and SSA uses of flow law, for now.Ed Bueler6+1089-144
2009-06-10 23:24* Trivial change in comment of pccoupler.hh. * More work on IceEnthalpyModel. Use of flow law now pretty clear.Ed Bueler4+181-77
2009-06-10 19:49* More work on IceEnthalpyModel. Compiles, but that's all ... * Addition of Lliboutry and Duval (1985) reference. * Comments changed in columnSystem.hh, materials.cc.Ed Bueler6+389-180
2009-06-10 00:29Work on IceEnthalpyModel. Compiles and runs but does exactly same thing as pismr (except for some debug msgs, and writing out an enthalpy field at end which doesn't mean anything ...).Ed Bueler6+295-69
2009-06-09 21:14* Fixed inclusion of STL vector and STL algorithm in iceCompModel.cc * Trivial change to browser. * Work on IceEnthalpyModel. Doesn't even compile.Ed Bueler5+61-11
2009-06-01 07:09Largely formal reorganization related to bed deformation code. See task #6630. The most substantial issue, of making a decent BedDeform base class and derived classes, including BedDeformLC which will manage the scatter to proc zero, is NOT yet done.Ed Bueler19+892-842
2009-05-31 22:22* Fixed bug #12916, re Test L. Actual issues were: (1) correct handling of points outside ice cap and (2) checking return values for status so that bug was localized. * Replaced embarassing heapsort code with proper STL mechanism (sort a vector of structs). Much nicer! * extrasGSL.{hh|cc} are mis-named. Currently only used in bed deformation code. Low priority to rename/reorganize. * Improved option handling for verifynow.py. * Suggestion: replace all options documentation in Appendix C of User's Manual (re Python scripts) with obvious 'foo.py --help' mechanism. Done for verifynow.py.Ed Bueler8+195-269
2009-05-31 00:36Added Ohmura 1987 ref. Fiddles with appearance of coupler and IceGRNModel stdout.Ed Bueler3+46-35
2009-05-20 18:04* Minor changes to the browser and the BOMBPROOF documentation. * Partially implemented the Timeseries class, which will replace Data1D and PISMClimateForcing. This will unify the way metadata and units are handled in PISM, and will allow saving time-series (once it is fully implemented). *As of now, this code is not used at all*. * Added a number of new methods to NCTool. * Refactored NCTool and NCVariable so that NCTool and NCConfigVariable do not require IceGrid (this closes task #6617). NCTool should be refactored some more by moving methods requiring IceGrid into a derived class.Constantine Khroulev14+585-156
2009-05-19 02:28Cleaning on EISMINT-Greenland example. Realized all code to delete Ellesmere Island is actually unnecessary (duh ...) because the fillmissing.py script puts it below sea level anyway. Bedrock thermal is included by default, and bed deformation for CCL3, closing task #6537. IceGRNModel is a model citizen.Ed Bueler9+180-211
2009-05-17 09:08Replacement of old couplers (PDDCoupler, MonthlyTempsCoupler, and the PDD in EISMINT-Greenland by the new system (SnowModelCoupler + LocalMassBalance and MonthlyDataMaps classes) largely complete. User's Manual not updated yet.Ed Bueler16+364-1479
2009-05-16 22:26* Completed implementation of Fausto et al (2009) scheme. Reasonable pcctest results from EISMINT- and SeaRISE-Greenland. * Monthly temps not yet tested. Also remains to remove PISMPDDCoupler and PISMMonthlyTempsCoupler, and rewrite EISMINT-Greenland to incorporate new scheme. * Renamed a couple of pism_config attributes to emphasize they are defaults.Ed Bueler8+178-50
2009-05-16 01:41PISMSnowModelAtmosCoupler with Fausto et al parameterization looks good. Tested on SeaRISE-Greenland data. Seems the pseudo-random version warmer with sigma=0 than it should be? It remains to implement formula (6) from Fausto et al (2009) and to revise EISMINT-Greenland and User's Manual appropriately.Ed Bueler4+97-36
2009-05-15 22:47* PISMSnowModelAtmosphereCoupler mostly implemented. Builds and runs but not right mass flux result. * snow_temperature now used as output standard_name for PISMSnowModelAtmosphereCoupler. * IceInfoNeededByCoupler struct replaces both IceInfoNeededByAtmosphereCoupler and IceInfoNeededByOceanCoupler. * PISMClimateCoupler::writeCouplingFieldsToFile() takes current time as argument so models can write snapshot at current time. * "snow_temp_d_ma" names of Fausto et al parameters replaced by "snow_temp_fausto_d_ma", etc. * pcctestsm.cc is temporary rewrite; will replace pcctest.cc when deprecated couplers PISMMonthlyTempsCoupler, PISMPDDCoupler are removed.Ed Bueler25+613-336
2009-05-15 02:15* Added PDD and snow temperature parameterization constants to pism_config.cdl; not done. * Trivial comments/includes changes in iceGRNModel.cc, iceModel.hh. * Config setup for LocalMassBalance. * Historical note from src/coupler/README integrated into pccoupler.cc at appropriate point. * Lots of work toward PISMSnowModelAtmosCoupler which will be simpler than, and supercede both PISMPDDCoupler and PISMMonthlyTempsAtmosCoupler.Ed Bueler8+555-42
2009-05-14 18:43Browser and bibliography maintenance.Ed Bueler2+22-26
2009-05-14 08:33Moved monthlyDataMaps.{hh|cc} to src/coupler/. Potentially they could be in base/, but no reason actually foreseen for that.Ed Bueler3+168-168
2009-05-14 08:18* Additional work on the temperature parameterization and PDD refactor: LocalMassBalance object now implements both Calov-Greve default PDD calculation and random calculation. * Plan: PISMSnowModelAtmosCoupler will replace both PISMMonthlyTempsCoupler and PISMPDDCoupler. Reduced inheritance desirable, and possible through creation of MonthlyDataMaps and LocalMassBalance objects. * Add Fausto et al to doc/ice_bib.bib. * Remove mention of Reference Manual in doc/installation.tex. Add graphviz dependency in same place.Ed Bueler7+243-71
2009-05-14 02:01Revision compiles but new code in localMassBalance.cc and monthlyDataMaps.cc is not called by anything. The goal is to deal with the unmanagable inheritance among PISMAtmosphereCoupler derived classes. Now the monthly temperature maps will be handled by a separate tool, and the PDD will be an instance of a local (i.e. no gridded information) model. Default implemention of PDD will be Fausto et al, but not ready yet.Ed Bueler6+451-5
2009-05-08 18:46* Documented most of flags and parameters in pism_config.cdl * Removed the -a option from the User's Manual * Minor changes to the PISM Class browser * Fixed a couple typos in ice_bib.bib * Removed NCTool::{read,write}_polar_stereographic(...)Constantine Khroulev18+132-156
2009-05-08 01:55Various little fiddles to improve browser main page, to add some useful links in that browser, and to clean up the doc directory (by moving deprecated summary).Ed Bueler10+75-100
2009-05-07 00:46* Removed comments.hh and doc_tempage.hh, because their contents are in doc/browser_{pages,bombproof}.txt now. * Fixed some typos in iceMISMIPModel.hh and iMtemp.ccConstantine Khroulev4+4-986
2009-05-07 00:43* Moved the standard names table, "PETSc for PISM users" and BOMBPROOF documentation to the HTML browser. * Removed the Reference Manual as redundant.Constantine Khroulev9+839-630
2009-05-06 08:18A small start on enthalpy method; executable 'penth' calls IceEnthalpyModel. At this point, just reads and writes enthalpy field.Ed Bueler4+269-2
2009-05-06 01:54pst.sh in examples/pst/ cleaned up to be easier to use for testing. docstring handling bug in doc/config_doc.py fixed.Ed Bueler2+21-19
2009-05-06 01:13* Fixed a little bug I introduced in the previous revision (PetscErrorCode basalstress(...) had a wrong signature). * Replaced \cite{...} commands with doxygen-style \ref almost everywhere.Constantine Khroulev15+38-34
2009-05-06 00:50* Added the reference list (http://www.pism-docs.org/dev/doxy/html/references.html) to the class browser and the script used to generate it. * Moved more constants to pism_config.{cdl,nc} * Added a page documenting configuration flags and parameters (http://www.pism-docs.org/dev/doxy/html/config.html) * Removed fullbib.tex * Some more error checking in nc_util.ccConstantine Khroulev22+1351-191
2009-05-02 01:15* Removed the -a (append) option. * Fixed EISMINT-Ross. * Added the "config file" src/pism_config.cdl and associated options -config and -dump_config. Moved some default constants and flags from iMdefaults.cc to pism_config.cdl * Changes to NCTool and NCVariable related to the above. In particular, created NCConfigVariable, a class for reading and writing scalar parameters and boolean flags stored as NetCDF attributes. Will use this to store projection (mapping) parameters and global attributes later.Constantine Khroulev18+561-198
2009-05-01 19:53Add -nc to wget in examples/eisross/quickstart.sh.Ed Bueler1+1-1
2009-05-01 08:33* Moved fine and equally-spaced grid parameterization procedures from iMtemp.cc to IceGrid.Ed Bueler7+575-365
2009-04-28 21:31Added fields accessible to IceInfoNeededByCoupler; mechanism is deprecated in trunk. Replacement will be more flexible.Ed Bueler3+16-4
2009-04-27 22:58Completed task #6597.Constantine Khroulev5+48-12
2009-04-27 03:05Completed task #6564. The temperature code now clearly identifies which column-independent and which column-dependent quantities go into the solution of the tridiagonal system in each column. No obvious loss of efficiency. Should be much easier to create a new enthalpy-based version.Ed Bueler6+373-520
2009-04-26 09:28Progress on task #6564. Fully implemented system context for ageStep(). Note that the file src/base/iMtemp_ctxdev.cc will disappear once implementation is completed. Also, moved bulge reporting to temperatureAgeStep(). Also, added a Joughin et al reference to doc/ice_bib.bib.Ed Bueler9+336-869
2009-04-26 02:38Start on task #6564 in new source file src/base/iMtemp_ctsdev.cc. Will create class for columnwise calc of age and temperature step. Then enthalpy version of this class will be easier to write.Ed Bueler1+968-0
2009-04-24 23:28Bug #13424 dealt with. midnight.arsc.edu also gives compiler warning "'double saving_after' might be used uninitialized in this function" for src/base/iMIO.cc line 707, so an unlikely value is used to initialize; please correct this if value matters in non-obvious way.Ed Bueler2+3-3
2009-04-24 19:50* Lots of changes related to switching to std::string in IceModelVec. * Fixed the "disappeared short_names" issue. * More metadata-handling methods in IceModelVec. * Improved error-checking in IceModelVec methods. * Added documenting comments to NCVariable.cc. * Fixed the "t"-variable-related bug (hopefully). * Added the call to oceanPCC->writeCouplingFieldsToFile() to IceModel::write_snapshot(), fixing bug #13421 (in the trunk).Constantine Khroulev17+308-216
2009-04-23 18:37Bug fix for PISMConstOceanCoupler, which should be writing sea level time series even if it doesn't write constant values for basal temp and mass flux. Bug fix for pcctest, which was going over the end of the time interval.Ed Bueler2+21-10
2009-04-23 17:33Slight semantic change to pcctest which make it go fast for long paleo evaluation of climate. Namely, PISMPDDCoupler is only asked for dt=1.0a length updates.Ed Bueler1+17-6
2009-04-22 04:02Added Greenland references to doc/ice_bib.bib. Very minor typo correction in error message in src/base/iMtemp.cc.Ed Bueler2+64-3
2009-04-19 20:23Update of standard names list in doc/. Change of vGhf=bheatflx to have no standard name; the proposed is lithosphere_upward_heat_flux.Ed Bueler2+25-13
2009-04-17 23:11* Major overhaul of IceModelVec: created NCVariable to allow having more than one variable per IceModelVec. This is a step in the dof > 1 direction. * Changed the required NetCDF version from 3.6.0 to 3.6.1 in the installation manual. (netcdf4-python requires 3.6.1, the rest of PISM works with 3.6.0). * Fixed lots of "unused parameter" warnings. * Switched from characted arrays to std::string in lots of places (mostly NCTool). * Turned off NetCDF variable pre-filling. This should make writing output files (and snapshots) a little bit faster. * _FillValue and missing_value attributes are ignored now (interpreting them according to the NetCDF User's Manual is a pain and is not really necessary).Constantine Khroulev23+1006-1117
2009-04-17 20:59Merge of changes on stable0.2, r658:659.Ed Bueler3+30-6
2009-04-17 09:01Merge of r655:657 bug fix from tags/stable0.2/; fixes to monthly temperature handling.Ed Bueler3+42-29
2009-04-15 21:03* Switched to using std::string to store the history attribute and the executable name. * Moved the code defining the NetCDF variable from IceModelVec{2,3} to the base class. * Removed the dummy IceModelVec::operator=() definition, because it was causing warnings. * Bug-fix: IceModel::ssa was not explicitly set to NULL in the constructor. This should go in stable0.2, too.Constantine Khroulev8+60-155
2009-04-14 01:32* Added the initialization sequence flow-chart to the PISM Class browser. * Removed IceModel::putLocalOnProcZero and getLocalFromProcZero, because they are not used any more.Constantine Khroulev5+33-59
2009-04-12 19:17Addition of references to doc/ice_bib.bib.Ed Bueler1+60-2
2009-04-06 20:54Code clean-up: removed calls to IceModelVec::destroy() from destroyVecs(). See task #6502. Also, IceModelVecs are passed to functions by reference (and not by value) now. A copy-constructor allows passing by reference and creating aliases such as "IceModelVec2 vHnew = vWork2d[0];" in IceModel::massContExplicitStep(). These copies are shallow, they do not allocate new Vecs and do not free the memory used by the original IceModelVec once destroyed.Constantine Khroulev9+108-118
2009-04-03 16:28* Moved utility functions processing command-line options out of IceModel. * Made SSAX, SSAKSP etc protected instead of private.Constantine Khroulev9+77-81
2009-04-02 19:55* User's Manual: Documented the grid extension mechanism. * Code clean-up, documenting comments, more error checks (mostly related to IceModelVec).Constantine Khroulev6+143-75
2009-04-01 21:21* Global find-and-replace in comments: "ignor " -> "ignore", "ignors" -> "ignores". * More documentation comments in IceModelVec3.cc * Moved IceModel initialization routines to iMinit.cc * Implemented the automatic vertical grid extension (task #4218). * Wrote the grid extension software test (test_11.sh). * Completely removed the IOwnDA flag from IceModelVecs.Constantine Khroulev13+729-477
2009-03-31 22:18I believe I have fixed bug #13922. Check occurs in computeMax2DSlidingSpeed() for whether point is in ice-free ocean, given the option-controlled semantics of the mask. Needs testing in Antarctic model because no verification etc is likely to be affected; i.e. good test results but that's expected since they don't have ice-free ocean. Also changed BPbulge reporting to stop reporting zero percent all the time.Ed Bueler3+21-6
2009-03-31 02:14Because of 'too many low temps' problem in fjord in SeaRISE-Greenland, and previous rare instances, added 15 K advection bulge limiter. Probably replaces 'too many low temps' entirely. Should be ported to stable0.2. Also, declared IOwnDA in iceModelVec.hh, and removed unuse ierr in pccoupler.cc.Ed Bueler3+39-9
2009-03-30 20:42Removed the IceModelVec{2,3}::createSameDA() mechanism as unnecessary. (Verification tests show that everything works just as before.)Constantine Khroulev4+5-73
2009-03-27 22:21User's Manual: Added a paragraph about restarting from a snapshots file and using ncks to extract a record sitting in the middle of a NetCDF file.Constantine Khroulev1+17-3
2009-03-27 21:26Fixed my silly mistake (in the trunk).Constantine Khroulev1+2-1
2009-03-27 20:26* Replaced PetscOptionsHasName with check_option(), a replacement that does not depend on the argument given with an option. * Replaced PISMOceanCoupler::reportSeaLevelElevation() with updateSeaLevelElevation(). Moved -dSLforcing there.Constantine Khroulev24+121-107
2009-03-27 08:26Fixed units for lat,lon: degrees -> degree. Added comments on proposed std names, and added html page of such tables in doc/.Ed Bueler3+335-7
2009-03-21 02:47* BUG found; my fault. "dt * secpera" appeared as argument to PISMClimateCoupler::updateSurf... where either "dt / secpera" was meant, or where initialization was occurring with no meaningful dt, so "0.0" was meant. Will need porting to stable0.2.Ed Bueler14+157-113
2009-03-20 20:11* Cleanup: removed util/matlab and doc/pism-install-* * Internal structured are allocated late now. See task #6517.Constantine Khroulev11+77-544
2009-03-20 17:28* Fixed processing -ys, -ye and -dt options in pcctest (this should happen before the PCC->initFromOptions() call) * Added NCTool::create_timeseries() and NCTool::append_timeseries(). These methods are used in pccoupler.cc to write time-series of surface temperature offsets and the sea level.Constantine Khroulev4+100-9
2009-03-19 20:56Snarky Bueler-generated comments in iceEISModel.cc removed. 'unreachable return' warnings squashed in pccoupler.cc.Ed Bueler2+2-16
2009-03-19 19:38* Minor fixes in test/software * Moved the code processing the -eisII option from IceEISModel::setFromOptions() to IceEISModel::set_grid_defaults(). This should fix the bug #13219.Constantine Khroulev5+41-30
2009-03-19 09:02Follow-up on r625. Allow customization of crit_temp.Andy Aschwanden1+1-0
2009-03-18 22:58* Rewrote a regular expression in run_tests.sh so that it works on pingo (hopefully). See task #6533. * Added the new figure: artm_timeseries.png and a paragraph about testing the surface temperature forcing with pcctest to the User's Manual. * Fixed a bug in pccoupler.cc * Removed util/nc_template.py from the trunk (I removed it from stable0.2 already)Constantine Khroulev5+24-196
2009-03-17 19:20User Man is now FIXME-free. Cleaned up bed deformation model usage and computational grid (quadZ and chebZ options) sections. Also killed unused var warning on pccoupler.cc.Ed Bueler2+56-48
2009-03-17 13:03Replace dropped line in r625.Jed Brown1+1-0
2009-03-17 13:01Allow customization of {A,Q}_{cold,warm} in Paterson-Budd.Jed Brown1+4-1
2009-03-17 00:53Small fixes. I was too quick to commit last time; sorry...Constantine Khroulev2+1-4
2009-03-17 00:50Minor fixes related to the task #6536 plus some code cleanup. This should probably go into stable0.2, too.Constantine Khroulev7+24-57
2009-03-17 00:24Moved the FIXME note at the beginning of the EIMINT-Greenland section to a footnote.Constantine Khroulev1+6-9
2009-03-17 00:16* Fixed examples/mismip/solverSM.py. See task #6518. * Documented (roughly) -quadZ and -chebZ command-line options.Constantine Khroulev2+22-41
2009-03-16 18:03Updated software tests; see the task #6533.Constantine Khroulev12+112-36
2009-03-15 19:42* build/Makefile: Style change on PISM_REVISION. * Deleted comment in src/base/iMsia.cc. * src/base/doc_tempage.hh: Improved doc of BOMBPROOF. * Rest: Fixed unused variables and unreachable default case warnings from pingo.Ed Bueler7+62-69
2009-03-14 08:38Merge 615:616 changes from stable0.2, with fix to bottom line on front page of User Man.Ed Bueler3+5-7
2009-03-12 08:11Updates to manual in trunk. Especially PDD, pcctest, and plastic ice streams usage sections. Also a few Appendix A edits.Ed Bueler2+115-119
2009-03-12 00:37Manual and pcctest update (in the trunk).Constantine Khroulev3+118-10
2009-03-10 20:42Scripts examples/eisgreen/*.sh maintained on trunk. See bug #13164.Ed Bueler3+6-6
2009-03-10 01:10Installation manual cleanup on trunk. Added python-dev recommendation for Debian.Ed Bueler1+18-11
2009-03-08 03:20Changed src/CMakeLists.txt according to Andy's concrete suggestions, bug #13133. Not tested ... Debian CMake too old.Ed Bueler2+8-11
2009-03-07 03:44Minor maintenance on trunk/doc/.Ed Bueler2+8-2
2009-03-07 02:49MISMIP mostly fixed up, but not really satisfactory. Runs, anyway. See task #6518 for one unresolved issue.Ed Bueler14+300-363
2009-03-06 21:06* Updated the Installation Manual * Fixed Makefile so that 'make pismr', etc work.Constantine Khroulev2+83-106
2009-03-06 19:08* Some more work on the Manual. * Changed -constant_hardness command line option to the -ice_type custom -ice_custom_hardness combination in examples/eisross/tune.py * Removed a and A diagnostic viewers. * Minor code cleanup.Constantine Khroulev10+85-159
2009-03-06 08:06Sweep through manual--sorry if I caused conflicts for you Constantine ... perhaps not. Verification figures redone for r589, which is fine for stable0.2.Ed Bueler9+293-157
2009-03-06 00:00* Fixed MISMIP initialization. Note that now -Mx sets the number of points in the flow direction. * Fixed bed deformation initialization. * Added an empty virtual destructor ~SSAStrengthExtension(). * More work on the Manual.Constantine Khroulev8+72-46
2009-03-05 23:17Jed's finite element implementation more hidden. Will not appear in stable0.2 docs, but still compiles and is present.Ed Bueler13+14-12
2009-03-05 22:57Updates to manual, especially verification and associated figures.Ed Bueler30+507-358
2009-03-05 21:08* Changed the ice age default to 0 years. This should fix the bug #13117. * Disabled regridding bedrock temperatures from files with only one bedrock layer.Constantine Khroulev3+8-6
2009-03-05 03:49Maintenance.Constantine Khroulev5+68-38
2009-03-04 22:37Minor fixes/cleanup.Constantine Khroulev5+11-11
2009-03-04 06:55Fiddles with IceType methods, as related to IceCompModel, all from very strange non-bug.Ed Bueler6+107-30
2009-03-03 23:53doc/ directory cleanup. browser mainpage cleanup.Ed Bueler7+73-87
2009-03-03 23:28* Added a new figure: climate_cartoon.{svg,pdf} * Replaced some figures with better versions. * NCTool::open_for_reading() stops if a file can't be opened now. * All the executables print out the PISM revision number in the first line of the output. * Switched to "CF-1.4" in the "Conventions" global attribute (IDV 2.6u2 still can't load these files, but a development version, 2.7a1, works with our output files just fine.) * Added the "source" global attribute with PISM's revision number. Also, PISM prints a warning if someone tries to use the -i option with a file that does not have the "source" attribute containing the string "PISM". (Which should warn someone trying to use as PISM input a file that is not a PISM output file.) * Minor fixes in the User's Manual.Constantine Khroulev30+443-197
2009-03-03 22:32Manual improvements.Ed Bueler3+184-193
2009-03-03 03:42Fixed pismv restart bug by adding init_coupler() to derived class.Ed Bueler2+143-112
2009-03-03 03:26Updated software tests.Constantine Khroulev8+18-16
2009-03-03 02:06* Everything builds. Next step: testing. * Added doc/initialization-sequence.{dot,png} -- a flow-chart illustrating PISM's initialization sequence.Constantine Khroulev13+357-223
2009-03-02 22:16pismv runs with -i. test_01.sh is pismr exact restartability, not pismv. cleanup.Ed Bueler8+39-45
2009-03-02 18:25Fix: IceModelVecs look at the units in the -i file now.Constantine Khroulev2+20-5
2009-03-02 17:53* We write age and the time variables in years (instead of seconds) now. * Should fix the bug #13078.Constantine Khroulev7+24-10
2009-03-01 23:55This should take care of the bug #13084.Constantine Khroulev10+11-11
2009-03-01 09:01Documentation cleanup, especially to section 3 of Users Manual.Ed Bueler4+99-64
2009-02-28 02:11A very minor fix in src/base/grid.cc.Constantine Khroulev1+6-3
2009-02-28 02:07Major reorganization of IceModel initalization sequence and the IceGrid class.Constantine Khroulev34+1443-1370
2009-02-27 17:35Fixed examples/eisross/eisross.py and /examples/eisross/rossplot.pyConstantine Khroulev2+20-20
2009-02-27 05:24Update scripts in examples/eis{ross|green} to use -i, -boot_from.Ed Bueler2+12-26
2009-02-27 01:04Update manual to reflect -ice_type and -shelfext_ options.Jed Brown3+46-41
2009-02-26 22:56Make Ross work again (bug #13072)Jed Brown4+21-14
2009-02-26 22:56Don't warn on usage for -gk and -gk_age since this is still the way to influence grain size computation, it also sets the defaults for ice type, although an explicit -ice_type can override it (but grain size computation would be done anyway).Jed Brown5+63-24
2009-02-26 18:28Add missing iceShelfExtension.{hh,cc}Jed Brown2+206-0
2009-02-26 18:07Added FindGit.cmakeJed Brown1+35-0
2009-02-26 18:07Cleanup in manual, initialization and option-handling robustness in materials.cc.Jed Brown4+30-20
2009-02-26 18:07Introduction of IceShelfExtensionJed Brown23+140-127
2009-02-26 11:34Move udunits.dat to pismudunits.dat. This way, if PISM is packaged, it won't conflict with udunits. Handle the uninstalled case correctly with CMake and give a more helpful error message.Jed Brown6+41-11
2009-02-25 09:04Minimally nice error message if dTforcing, dSLforcing file not found. Closes task #5884.Ed Bueler1+5-1
2009-02-25 08:51Non-const static data members generally removed from classes. In particular, all constants in materials.hh classes are initialized in constructors, a change from r4. Closes bug #13055.Ed Bueler5+165-185
2009-02-25 05:41Cleanup of initialization sequence. Will remove non-constant static data members of classes in materials.{hh|cc}, according to bug #13055.Ed Bueler11+97-85
2009-02-23 19:54Display -ssa_external in -help.Jed Brown1+5-1
2009-02-23 19:54Sanitize git revision stringJed Brown1+1-0
2009-02-23 19:54Better help for materials optionsJed Brown1+5-5
2009-02-23 19:54High-level SSA documentationJed Brown1+52-0
2009-02-23 19:54Don't pretend to use dynamically loaded libs for SSA.Jed Brown1+8-1
2009-02-23 19:54Remove logging option, if PETSc is built without logging, it will just do nothing (obviously).Jed Brown1+0-8
2009-02-23 19:54Implement various inhomogeneous Neumann conditions in SSAFE.Jed Brown7+232-52
2009-02-23 08:30Various small steps toward maintenance and integration of Jed's IceFactory into documented PISM. Removed environment variable controlling event logging; Jed is no doubt right there is no good reason to disable it; it is always on now.Ed Bueler15+119-84
2009-02-22 03:15Added IceFactory initialization to IceModel constructor (iceFactory.setType(ICE_PB)) so that executables pismr and pismd and pgrn at least start up.Ed Bueler2+5-2
2009-02-21 21:40Make flowTable and tryLCbd use new IceType interface.Jed Brown2+8-5
2009-02-21 19:32Maintenance on Getting Started section of Users Manual. Removed a bunch of text on semantics of viewers and on evolution-vs-diagnostic. Should those be kept or moved, instead of deleted?Ed Bueler2+116-67
2009-02-21 19:05Scaling so that initial residuals are O(1)Jed Brown1+4-4
2009-02-21 19:02Make verifynow.py refine in the x direction, because of r514Jed Brown1+3-3
2009-02-21 19:02Recognize -law -gk -gk_age and issue deprecation warnings.Jed Brown4+22-20
2009-02-21 19:02Check PETSc versions everywhere we need to.Jed Brown6+23-10
2009-02-21 19:02Fix memory errors.Jed Brown4+7-4
2009-02-21 19:02Makefile update for r517 and 526Jed Brown1+3-3
2009-02-21 19:02Start SSA implementation, use -ssa_external to access.Jed Brown8+1488-2
2009-02-21 19:02Huge refactoring of IceTypeJed Brown36+520-473
2009-02-21 19:02Inconsequential changes, mostly to quash warnings due to unused variables.Jed Brown11+23-27
2009-02-21 19:02grav -> earth_gravJed Brown10+18-21
2009-02-21 19:02Add IceType interface to factor out vertical integrals.Jed Brown2+23-21
2009-02-21 19:02Compiler warningsJed Brown3+5-3
2009-02-21 19:02The PETSc API can give us a string for convergenced reasons.Jed Brown1+2-5
2009-02-21 19:02Whether an ice type uses grain size should not be managed externally. Move it into IceType.Jed Brown5+10-8
2009-02-21 19:02Basic Mask functionality does not need to be part of IceModel.Jed Brown14+78-85
2009-02-21 19:02Move PISM_REVISION to it's own file so that less needs to be rebuilt.Jed Brown5+8-5
2009-02-21 19:02Give basal stress and viscosity derivatives, factor viscosity.Jed Brown2+83-13
2009-02-21 19:02Updates to CMake build, minor cleanup, PETSc 3.0.0Jed Brown10+63-35
2009-02-21 01:56Rollback to r514. Working on stable0.2 continues.Ed Bueler59+645-2260
2009-02-21 01:00Attempt to fix up so builds with GNU Make. Not successful because PISM_HAVE_PETSC3 flag is not checked consistently: src/base/ssa/ssa.cc:26: error: PetscLogEvent does not name a type. Would like to avoid having to explore Petsc 2.3 versus 3.0 semantics to make revisions work ...Ed Bueler6+11-19
2009-02-20 23:48Start SSA implementation, use -ssa_external to access.Jed Brown8+1488-2
2009-02-20 23:48Huge refactoring of IceTypeJed Brown36+520-473
2009-02-20 23:48Inconsequential changes, mostly to quash warnings due to unused variables.Jed Brown11+23-27
2009-02-20 23:48grav -> earth_gravJed Brown10+18-21
2009-02-20 23:48Add IceType interface to factor out vertical integrals.Jed Brown2+23-21
2009-02-20 23:48Compiler warningsJed Brown3+5-3
2009-02-20 23:48The PETSc API can give us a string for convergenced reasons.Jed Brown1+2-5
2009-02-20 23:48Whether an ice type uses grain size should not be managed externally. Move it into IceType.Jed Brown5+10-8
2009-02-20 23:48Basic Mask functionality does not need to be part of IceModel.Jed Brown14+78-85
2009-02-20 23:48Move PISM_REVISION to it's own file so that less needs to be rebuilt.Jed Brown5+8-5
2009-02-20 23:48Give basal stress and viscosity derivatives, factor viscosity.Jed Brown2+83-13
2009-02-20 23:48Updates to CMake build, minor cleanup, PETSc 3.0.0Jed Brown10+63-35
2009-02-20 06:23* More cleanup. * Changed meanings of x and y dimensions (and variables): 'x' and 'y' should always be consistent with projection x and y variables, correspondingly. This changes the order of variables used in PISM output files to t,y,x,z; note that this PISM revision can initialize from old PISM output files. The only drawback is that the output will be transposed (naturally). One solution is to use the following script to convert an "old" PISM file to a "new" one:Constantine Khroulev5+58-43
2009-02-19 20:03More cleanup.Constantine Khroulev6+71-168
2009-02-19 17:50More cleanup.Constantine Khroulev3+52-64
2009-02-19 06:29Input/output code cleanup.Constantine Khroulev6+79-78
2009-02-19 04:22Fixes to src/pcctest.cc to follow non-symmetric grid range modifications. Addition of diagnostic instantaneous snow temperature for PDD diagnosis.Ed Bueler5+61-61
2009-02-19 03:27* Added software test #10: comparing regridding with mpiexec -n {1,2,3,8,10}. * Cleanup in nc_util.ccConstantine Khroulev2+101-57
2009-02-19 02:35IceModelVec::regrid() reports default values in glaciological units (if set).Constantine Khroulev2+6-2
2009-02-19 02:12* Fixed a bug introduced in r502. Sorry.Constantine Khroulev1+1-1
2009-02-18 23:05* Maintenance and clean-up. * fill_missing.py removes "missing_value" and "_FillValue" attributes now.Constantine Khroulev5+20-19
2009-02-18 07:11Maintenance resulting from (in part) look at manual on EISMINT-Greenland and EISMINT-Ross. Nothing like a full sweep; just dabbling.Ed Bueler18+136-209
2009-02-17 01:05EISMINT-Greenland starts up correctly. Sorry about that.Ed Bueler5+13-3
2009-02-16 23:26* Removed debugging messages from nc_util.cc. * Added a test of 3D regridding from files with different variable orders.Constantine Khroulev2+42-5
2009-02-16 23:06* More comments in test/software/pism/test* * Added NCTool methods for creating start and count arrays in a more general manner and transposing data. The code can actually "transpose" 3D arrays, but this capability is never used (because 3D fields we regrid come from PISM and have the expected storage order). Note that we still need to rename x and y in the output so that x corresponds to projection_x_coordinate. * Wrote a test_05.sh, which tests the ability to bootstrap from files with different orders of variables. * Documenting comments and clean-up in LocalInterpCtx.ccConstantine Khroulev9+318-103
2009-02-16 09:02Bug in r500 fixed. Seg fault was a stack overflow, not pointer deref. Onward to sweep through manual for stable0.2 release ...Ed Bueler2+7-9
2009-02-16 05:45Cleaned up PISMPDDCoupler. Documentation and maintenance. Builds but does not run; do pisms -eisII A -verbose 5 to see seg fault caused by something in PISMOceanCoupler::updateShelfBaseTempAndProvide().Ed Bueler12+532-899
2009-02-15 05:17PISMClimateForcing mostly put into PISMAtmosphereCoupler and PISMOceanCoupler. Sea level (elevation) now the responsibility of PISMOceanCoupler, as it should be. THIS REVISION COMPILES BUT WILL NOT RUN FORCING CORRECTLY; need to actually use PISMClimateForcing::updateFromClimateForcingData() in PISM...Coupler::update...(). Also need to svn delete src/base/iMforcing.cc.Ed Bueler23+717-1178
2009-02-14 20:10Added documentation on PISMClimateCoupler, and other maintenance. forcing.cc still not integrated into PISMClimateCoupler.Ed Bueler12+1294-1100
2009-02-13 20:59Added util/nccmp.py, a tool for comparing two NetCDF files using the infinity norm and test/software, a collection of 8 software tests.Constantine Khroulev13+456-0
2009-02-12 22:48pismv now uses PISMClimateCoupler. Remaining: lots of testing as part of stable0.2 release, and integration of forcing.cc etc into PISMClimateCoupler.Ed Bueler9+187-126
2009-02-12 18:22Added more Vec size checks to methods of IceModelVec.Constantine Khroulev1+21-0
2009-02-12 10:05pismd now working. Maintenance on EISMINT-Ross.Ed Bueler10+399-383
2009-02-12 08:41IceGRNModel seems to be back to functioning, but with PISMPDDCoupler. Still need to work on pismd and pismv executables, which are currently turned off.Ed Bueler13+193-119
2009-02-12 03:44IceGRNModel illustrates that initialization sequence needs fixing (as we plan to ...). pgrn not working now.Ed Bueler12+160-113
2009-02-09 08:58More PISMClimateCoupler. Needed: pgrn testing, verification use, docs.Ed Bueler22+318-554
2009-02-08 22:32More PISMClimateCoupler: implemented trivial ocean coupler. Needs work: back to PDD, check pgrn, docs.Ed Bueler16+327-114
2009-02-08 09:27PISMClimateCoupler is (initially) attached to IceModel. Needs work. IceGRNModel, pgrn, pismd, and pcctest all turned off for the moment, and PDD itself doesn't work.Ed Bueler20+389-185
2009-02-07 01:03Cleanup and maintenance.Ed Bueler7+37-121
2009-02-06 23:23Added util/check_polarstereo.pyConstantine Khroulev1+209-0
2009-02-06 19:27* PISM can bootstrap from a file with non-symmatric x and/or y-range. This closes task #5802.Constantine Khroulev9+98-47
2009-02-05 20:22Smoothing of observed velocities added to inversion mechanism; under development ...Ed Bueler2+64-2
2009-02-04 23:58Bug-fixes: * Regridding (using -regrid_from) from a file with no vertical dimensions is handled correctly now. * Fixed bootstrapping from a file with a non-trivial time dimension (such as a PISM output file).Constantine Khroulev2+40-14
2009-02-03 21:54* Renamed -if to -i, -bif to -boot_from and -regrid to -regrid_from. Updated the manual. Closed task #6340. * NCTool::get_dim_limits() does not assume that a dimensional variable is ascending any more. (I.e. min and max are found by looking at all the values. Not a big deal, just a step towards being able to bootstrap from less PISM-specific files.)Constantine Khroulev23+299-145
2009-02-03 03:13More work on inverse model. Very clean results for case in examples/pst/convert_for_inv.sh, with no regularization. Ready to go on Joughin Greenland.Ed Bueler5+197-187
2009-02-02 08:48Work on inverse model, esp unregularized. Ready for Joughin data?Ed Bueler6+222-148
2009-02-01 22:25Maintenance; esp. doc of PETSC_OPTIONS and .petscrc in Users Manual.Ed Bueler4+51-8
2009-02-01 07:50Put the functionality of computeTillPhi.py into PISM. Option -topg_to_phi does this; see source. Option -surf_vel_to_phi read as alternative to also initialize till phi. Also: trivial maintenance of verification stdout.Ed Bueler9+164-48
2009-01-30 21:12* Renamed libudunits.{a,so} to libpismudunits.{a,so} and UDUNITS_PATH environment variable to PISMUDUNITS_PATH so that PISM *always* uses our custom version of UDUNITS. * Added amsmath to extra packages in Doxyfile_html. * Minor fixes in iceROSSModel.cc and eis_ross.py (related to unit conversion). * NCTool::get_att_text() 'returns' the length of the allocated string now. NCTool::get_att_double() only succeeds if the attribute has the length we want. * Rewrote and cleaned up all the code reading NetCDF attributes to avoid potential buffer overflows. Also, cleaned up the code reading and writing the history string. * Added the new IceModelVec method IceModelVec::is_valid(PetscScalar a), which returns 'true' if *a* is in the valid range and 'false' otherwise. * Other minor fixes/updates.Constantine Khroulev13+319-352
2009-01-29 20:43* Stripped down UDUNITS and moved it to src/udunits. Updated build/Makefile to build UDUNITS with PISM; we build a shared library by default and a static one if PISM_STATIC=1. * New NCTool methods for reading text and scalar attributes from a NetCDF file. * Preliminary work on IceModelVec::read_valid_range(). * Set valid_min = 0.0 for all the variables corresponding to magnitudes of a vector field. * Minor changes in build/Makefile.Constantine Khroulev90+6106-36148
2009-01-28 08:10Inverse model work: cleaning and clarifying. Other maintenance.Ed Bueler14+181-140
2009-01-27 01:21build/Makefile fix. Should compile and work now; compiles but *does not* work on marmaduke. Might be a yacc(bison) problem there.Constantine Khroulev1+9-5
2009-01-27 00:32* Added the UDUNITS library (see task #5802). Note that we might want to change the install location.Constantine Khroulev88+37769-154
2009-01-26 03:56Work on inverse model. Builds. Remains to respect observed mask when applying SSA operator. Also: fixed pgrn, pcctest (and other included stuff) dependency on libpism.so.Ed Bueler4+240-209
2009-01-25 19:55Work on inverse method. Does not compile.Ed Bueler3+383-288
2009-01-23 20:17r470 bad. Fixed; pism_intent=climate_steady case does not use GUNITS now.Ed Bueler1+2-2
2009-01-23 20:00Added GUNIT to pism_intent attribute of IceModelVec. Semantics used to want 'diagnostic' in that attribute to cause write in glaciological units; now it is 'GUNIT'. Fixed inactive name clash on IceModelVec::sqrt(); now it is IceModelVec::squareroot(). Cleaned up and completed IceModelVec::range() and norm() so they treat local Vecs appropriately.Ed Bueler7+137-59
2009-01-20 03:34Work on PISMClimateCoupler and PDD implementation. Seems to work when tested with executable pcctest. Other maintenance.Ed Bueler6+163-73
2009-01-19 07:49More PISMClimateCoupler: Initial implementation of PISMPDDCoupler done. Also made 'pcctest' executable that allows testing PISMClimateCouplers on fixed geometry ice sheets w/o involving IceModel.Ed Bueler7+426-159
2009-01-18 07:59More work on PISMClimateCoupler.Ed Bueler8+625-254
2009-01-16 22:19Maintenance.Ed Bueler8+24-34
2009-01-16 20:39Maintenance: C++ standard compatibility and makefile improvements based on Don Bahls advice from pingo.arsc.edu.Ed Bueler3+12-2
2009-01-16 08:10More work on PISMPDDCoupler. Added was_created() method to IceModelVec; checkAllocated() terminates if not allocated, while I just want to know if create() was called.Ed Bueler5+289-13
2009-01-16 03:21More work on PISMClimateCoupler. Creating more transparent interpretation of 'acab' in input, which currently is interpreted as net surface mass balance generally, except re-interpreted as snowfall accumulation if using a PDD.Ed Bueler10+351-100
2009-01-15 22:44Maintenance re stable0.2: -lstdc++ added to flags, CXX used for dependency gen, -Wall warnings from src/num/cubature.c killed (but not tested; i.e. if the uninitialized value is actually used it will cause a runtime error).Ed Bueler2+17-3
2009-01-15 03:16Work on PISMClimateCoupler. I want initial capability in stable0.2 so existing users can actually work with stable0.2.Ed Bueler4+108-20
2009-01-14 23:49Empty draft PISMClimateCoupler.Ed Bueler8+359-15
2009-01-13 20:51Maintenance. (All related to pingo.arsc.edu and to stable0.2 prep.)Ed Bueler2+9-3
2009-01-13 20:31Maintenance.Ed Bueler3+9-4
2009-01-13 19:37Maintenance.Ed Bueler4+64-55
2009-01-11 01:29Messages associated to regrid etc are more visually consistent. Fiddled with verbosity for some messages. Build uses '-g' in compilation if PISM_USE_DEBUG=1.Ed Bueler7+52-42
2009-01-10 23:21Added the explicit initialization of regrid_2d_only and no_regrid_bedrock flags.Constantine Khroulev1+6-1
2009-01-10 22:49Added grid_info::print() (for debugging).Constantine Khroulev2+41-6
2009-01-10 06:29* iMssaSNES.cc is a partially functional implementation of the SSA using PETSc SNES. Turned on with -ssa_snes. There is no Jacobian or preconditioner routine and it must be used with -snes_mf. Only works for Test J for now. Not clear why I can't make the non-Newtonian and basal resistance mechanisms work; they work for old SSA implementation; there is a bug as well as conceptual confusion I think.Ed Bueler7+202-155
2009-01-10 00:24A (very) minor change in iMIO.cc.Constantine Khroulev1+7-7
2009-01-07 00:52* Added and documented the -split_snapshots option. * Added the IceModel::write_extra_fields() method.Constantine Khroulev4+55-16
2009-01-03 23:32Test I shows nontrivialness of using SNES for SSA. Working on an ad hoc scheme combining a Picard stage and a Newton-Krylov stage.Ed Bueler3+194-114
2008-12-29 05:24Working on SNES use for SSA. Test J already works.Ed Bueler9+462-42
2008-12-28 06:01* build/Makefile much improved. "install" target gone; "make" or "make all" puts libraries in lib/, executables in bin/Ed Bueler7+184-272
2008-12-27 05:30Inverse model works. I.e. on PST experiment the tillphi map (from tauc map) can be recovered enough to give a pretty good match (1 in 100?) of surface velocity. I short-circuited the regularization scheme which solves a poisson-like problem; need to get back to that.Ed Bueler2+141-64
2008-12-26 05:45* PETSc 3.0.0 came out on Dec. 19. This revision should build under either 2.3.3 or 3.0.0; note environment variable PISM_HAVE_PETSC3 in build/Makefile.Ed Bueler11+751-207
2008-12-21 22:08Moved examples/trypetsc/ to obvious location src/trypetsc/. Added src/trypetsc/bugVecMax.c to clarify behavior of VecMax() etc. on DA-created local Vecs.Ed Bueler7+211-100
2008-12-20 05:31* Cleaner python library install instructions in Installation Manual. * Documented new EISMINT-Greenland and -f3d options in User's Manual. * I think there is a bug with VecMax,VecMin; use of PetscGlobalMax/Min necessary to get the right numbers. I.e. ranges on bootstrap vars. * Added 1 in 10^6 slop for LocalInterpCtx to allow slightly off grid subset check.Ed Bueler6+78-53
2008-12-19 09:14Use of pism_intent=diagnostic to determine if glaciological units used. More glaciologial units added.Ed Bueler4+45-21
2008-12-19 01:28pgrn should work with most Greenland setups. Fixed warning related to virtual destructor in materials.hh. New and better semantics for -f3d. More work on iMinverse.cc; still incomplete.Ed Bueler7+163-82
2008-12-17 21:11Fixed up poisson.c to be like IceModel in i,j meaning (i.e. re 'fundamental transpose').Ed Bueler3+100-73
2008-12-17 03:28* More progress on inverse modeling. Code all there. Will debug by trying to reconstruct phi from PST experiment P1.Ed Bueler10+318-147
2008-12-15 23:40Progress on inverse modeling. I promise.Ed Bueler7+307-241
2008-12-15 08:54Clearly headed toward results with iMinverse.cc stuff. See examples/trypetsc/poisson.c, which does the job needed for getting a regularized version of the tau_c from the basal shear stress, the sliding velocity, and a pseudo-plastic till model. Note iMinverse.cc got big so I split it.Ed Bueler10+806-340
2008-12-13 02:58Incremental fixes to iMinverse.cc.Ed Bueler2+88-58
2008-12-12 02:29* Changes to iMinverse.cc which put it on a good path. Now fofv is not read but computed by principled way compatible with -super model. * New method thicknessTooLarge() added, re task #4218. Almost redundant check should stay in ageStep().Ed Bueler5+321-176
2008-12-11 05:28Incomplete work on "inverse methods" and on calving front stress boundary condition. Both incomplete.Ed Bueler5+293-163
2008-12-11 03:23* Modified IceModel::bootstrapFromFile() and NCTool::regrid_global_var() to allow absent 't' dimension (and variable) in the bootstrapping file. Also, 'z' and 'zb' dimensions are completely ignored at the bootstrapping stage now. See task #5802. * Added the code reading valid_range, valid_min, valid_max, _FillValue and missing_value attributes and warning the user if a field in the bootstrapping file has values outside the valid range. * Code clean-up in nc_util.cc, iMbootstrap.cc and LocalInterpCtx.cc * LocalIntertpCtx(grid_info, double *, double *, IceGrid) should be called as LocalInterpCtx(grid_info, NULL, NULL, IceGrid) for 2D-only regridding. The following code from iceROSSModel.cc is an example. +verbatim+ grid_info g; ierr = nc.get_grid_info_2d(g); CHKERRQ(ierr); ierr = nc.close(); CHKERRQ(ierr);Constantine Khroulev24+636-571
2008-12-09 03:13* Sweep through iMinverse.cc methods. No intended change in semantics other than addition of ability to write out inverse model fields (basal velocity and basal shear stress). This code is ready for testing on Joughin surface velocities using essentially (or exactly) the IceGRNModel derived class. Run would look like pgrn -bif green.nc -Mx .. -My .. -Mz .. -quadZ -Lz 4000 -OTHER OPTIONS \ -y LONGISH -o foo.nc pgrn -if foo.nc -surf_vel_to_tfa green.nc -y 10 -o bar.nc assuming green.nc contained basic bootstrap fields (thk, topg, acab) and that green.nc *also* contains surface velocity in variables us,vs.Ed Bueler9+287-273
2008-12-08 03:48Initial creation of class IceCalvBCModel, derived from IceExactSSAModel. Already has finite difference scheme which computes outward normal to calving front.Ed Bueler6+431-14
2008-12-06 01:52Small change to semantics of FLOATING_OCEAN0: if -ocean_kill is set when -bif is done, will use OCEAN0. Otherwise -bif sets ice-free ocean to FLOATING. Note -if runs never set OCEAN0, but use it if -ocean_kill is set. Also: cleaned up standard out from iceModelVec::regrid_from_netcdf(); changes to EISMINT-Greenland stdout text to match.Ed Bueler6+95-58
2008-12-04 23:23Moved LocalInterpCtx to src/base/LocalInterpCtx.[cc|hh].Constantine Khroulev5+367-315
2008-12-04 22:37* Fixed the EISMINT-Ross example. (I broke it in r426.) * Added IceModelVec::checkAllocated() calls to various mathods of IceModelVec.Constantine Khroulev3+40-4
2008-12-04 20:42Updated the User's Manual.Constantine Khroulev1+107-122
2008-12-04 08:20Fixed test M. Now a credible way to test coming calving front implementation. Other cosmetic changes. EISMINT-Ross example seems to be broken?Ed Bueler6+74-40
2008-12-04 01:10* All the input/output command-line options expect a full file name (no '.nc' appending now). Specifying an output file name without the '.nc' suffix produces a warning. * Replaced all the SETERRQ(...) macros corresponding to runtime errors with {PetscPrintf(...); PetscEnd();}. All the runtime errors (such as invalid command-line arguments, etc) should produce short messages without a trace-back now. See task #5884. * Got rid of dumpToFile_diagnostic_netCDF(), used C++ default argument values to simplify initFromOptions() and writeFiles(). * Replaced 'new' with 'New' in exactTestM.c to avoid the C++ keyword name clash and 'old' with 'Old' -- for consistency. ;-)Constantine Khroulev32+343-304
2008-12-03 23:32More work on test M: matching flux at grounding line.Ed Bueler3+55-38
2008-12-03 08:44Update to verifynow.py to run test M. Everything works fine of what I implemented in r422; prev commit message wrong. Still need to check on test M setting of velocity in grounded part.Ed Bueler1+3-0
2008-12-03 08:31Initial implementation of test M. Something is wrong with r422 for these, but under r421 'pismv -test M -eo -Mx 201 -My 201 -Mz 11' gave high resolution version of exact solution. Will fix tomorrow.Ed Bueler8+257-71
2008-12-03 02:19Looked at CF standard name table and modified PISM to match. Search for 'PROPOSED standard_name' to see that.Ed Bueler3+55-37
2008-12-02 23:06* Updated src/base/iMinverse.cc and moved it to its original place. * Implemented the snapshot-saving feature, see task #6039. Now it's possible to save model state snapshots with -save_to snapshots.nc -save_at 1e3:1e3:1e4 (every 1000 years from 1000 to 10000) or -save_at 100,400,1000,5000 to save at these times. It does not affect the time-stepping mechanism, so snapshots are saved on the first time-step after the time requested. See the manual for more details. Note that saving a subset of variables in a snapshot is not supported at this point. * Documented -save_to, -save_at, -dontreadSSAvels and -a options. * Changed the default behavior when the file specified by the -o option already exists: now we print a warning, move it to filename~ (with the tilde added). It is also possible to append to the output file by giving the -a option. Updated test/verifynow.py to match this behavior. * A little bit of code clean-up in iMIO.cc * Renamed varname to short_name in IceModelVec. Also, added some more documenting comments.Constantine Khroulev18+1112-774
2008-12-02 04:15Improvements to ice shelf base thermo boundary condition in preparation for use of new PISMClimateCoupler. Note vTnew3 etc. must be 'protected', not 'private', for derived class temperatureStep() to work. Should *any* IceModel members be private?Ed Bueler4+66-43
2008-11-30 21:33More work on test M. Note build dir is a pain re creation of executables from simple?.c.Ed Bueler7+140-146
2008-11-30 08:29More test M preliminary work.Ed Bueler4+70-183
2008-11-30 03:40Preliminary work on test M, an 'exact' solution is a uniform thickness annular ice shelf with prescribed velocity accross grounding line (Dirichlet) and a calving front. Requires ODE solve like test L, thus not completely exact. src/verif/exactM.py is only meaningful so far. Also deleted unmaintained aix and xd1 config clutter.Ed Bueler7+299-369
2008-11-29 01:11Modified src/base/iMbootstrap.cc to fix the bug #12645. (We need to read all the variables *before* calling setMaskSurfaceElevation_bootstrap().) Please check if it works now and leave a comment or close the bug.Constantine Khroulev1+14-17
2008-11-28 01:59Added comments to temperatureStep() to indicate where PISMClimateCoupler will be called. Changed ice shelf base temperature boundary condition to 273.15K for now; PISMClimateCoupler will fix that. Sought the source of bug #12599; not sure if bug is still present; needs PIK help to reproduce.Ed Bueler3+57-50
2008-11-27 19:48Added ref to doc/ice_bib.bib. Trivial comment changes in src/verif.Ed Bueler2+11-1
2008-11-27 02:46* Finished implementing the IceModelVec abstraction. See task #5762. (The list of members of IceModelVec can be changed if we need more VecXXX functionality.) * Updated and extended NCTool to search for variables using standard_name. See task #5756 and task #5802. Also, there is a change in the way PISM writes its state to a file: instead of overwriting the output file we try to append along the t dimension and overwrite if it is not possible. (Note that we should append instead of overwriting the history string of the output file.) * Removed src/netcdf (we don't need that code any more). Moved src/netcdf/pism_state.cdl to doc/. * Moved src/base/iMinverse.cc to base/old. It will come back next time I commit. * Reorganized bootstrapFromFile_netCDF() to use IceModelVec::regrid(). Moved bootstrapping variable range reporting to IceModelVec::regrid(). * Major changes to iMIO.cc: dumpToFile_netCDF() and dumpToFile_diagnostic_netCDF() use IceModelVec::write() and initFromFile_netCDF() uses IceModelVec::read() * Fixed a DAVecGetArray/DAVecRestoreArray pairing issue someplace and included a bit of code to check this pairing (only compiled with CFLAGS+=-DPISM_DEBUG). * Modified test/verifynow.py to remove verify.nc before running PISM. (This relates to the new output-appending behavior. On the other hand, this should not matter; feel free to change it back.)Constantine Khroulev49+3344-4302
2008-11-26 07:52Default is now changed back to Mahaffy scheme for surface gradient. The eta=H^8/3 transformation is available with -eta; -no_eta is no longer an option. verifynow.py changed accordingly. vHexactLCreated flag handled correctly in iceCompModel.cc.Ed Bueler6+55-48
2008-11-18 03:39Fixed copy-and-paste error associated to communicating vHmelt, caused by Bueler in r407.Ed Bueler3+23-17
2008-11-17 23:05Documentation maintenance.Ed Bueler5+103-110
2008-11-15 02:20Added a couple of references and modified some comments.Ed Bueler3+25-6
2008-11-14 10:00Update CMake build, require 2.6.2 due to HINTS semanticsJed Brown8+428-169
2008-11-14 04:32Moved diffuseHmelt() from iMtemp.cc to iMbasal.cc. Removed all mention of capBasalMeltRate(). Trivial reference fix.Ed Bueler7+57-64
2008-11-12 19:44Derived classes should not have to rebuild the LocalInterpCtx for the -bif file. Now they don't: they can used IceModel::bootstrapLIC (after checking it is not == to PETSC_NULL). Also added references to ice_bib.bib.Ed Bueler4+48-18
2008-11-10 20:11* More work on the IceModelVec abstraction. * Updated and added on to the NCTool class. * Changes all over the place to match changes in NCTool and IceModelVec. * Added a comment about grid.da2 to grid.hh.Constantine Khroulev14+843-442
2008-11-08 03:25* Deleted old and unmaintained sources in src/base/old; can be recovered from gna.org.Ed Bueler35+661-3999
2008-11-06 05:41* Ice shelf extension mechanism is better. There is still an extension with constant membrane strength, but there is no change of the geometry. This means we avoid stupid mass conservation errors and stupid extra communication. This is still no substitute for a good force boundary condition, but it is cleaner. Performs well (still) on EISMINT-Ross.Ed Bueler14+175-165
2008-11-05 23:09* A better fix in iMIO.cc (there was a sizeof(int) != sizeof(size_t) issue).Constantine Khroulev1+4-2
2008-11-05 22:37* Some work on the IceModelVec abstraction. * A minor fix (PISM should work on marmaduke now).Constantine Khroulev3+87-30
2008-11-04 23:39* The history string can be arbitrarily long now. * Renamed HISTORY_STRING_LENGTH to TEMPORARY_STRING_LENGTH.Constantine Khroulev6+57-34
2008-11-04 18:45capBasalMeltRate() in src/base/iMvelocity.cc is an artifact of our old concern with SIA+SSA by horizontal gluing. Trying PST P1 with it removed.Ed Bueler3+7-2
2008-11-03 22:46Add CMake build system.Jed Brown8+452-3
2008-11-03 22:46avoid const castJed Brown1+11-9
2008-11-03 22:46Zap trailing semicolonsJed Brown2+4-8
2008-11-03 22:46Quash compiler warningsJed Brown1+31-31
2008-11-03 22:46Fix % escape characterJed Brown1+2-2
2008-11-03 22:46LOG_PISM_EVENTS -> PISM_LOG_EVENTSJed Brown5+14-14
2008-11-03 22:46Namespace WITH_FFTW -> PISM_HAVE_FFTWJed Brown6+14-14
2008-11-03 01:58Fixed bug #12555. The two critical parameters controlling velocities are the vertically averaged hardness, which is constant in EISMINT-Ross, and the thickness of the ice shelf extension, which controls force at the calving front. (Options -constant_hardness and -ssa_min_thk, respectively.) Now EISMINT-Ross, MISMIP, and test I and J all work as expected (at least all the basic runs in User's Man do).Ed Bueler14+79-84
2008-10-31 23:55* Moved IceModelVec3Bedrock to iceModelVec3.cc * Moved IceModelVec2 stuff to iceModelVec2.cc * Updated build/Makefile.Constantine Khroulev4+379-325
2008-10-31 17:41* Fixed a bug that I introduced in rev. 388. (Sorry.) * Modified pism_state.cdl and such (vubarSSA and vvbarSSA should not have standard names).Constantine Khroulev4+2-13
2008-10-29 22:25Implemented reading and writing vubarSSA and vvbarSSA.Constantine Khroulev6+126-33
2008-10-29 17:57* Switched to writing in double precision. * Preliminary work on saving (and reading) vubarSSA and vvbarSSA.Constantine Khroulev11+134-156
2008-10-27 02:15Added a couple of basic refs to doc/ice_bib.bib.Ed Bueler1+22-0
2008-10-24 23:48* Modified build/Makefile so that it's possible to build PISM on a system like Cray XT5 by running $ export PISM_STATIC=1 $ make $ make install without modifying any makefiles. * Updated the Installation Manual.Constantine Khroulev2+33-14
2008-10-22 17:27* Very minor changes to doc/installation.tex and doc/manual.tex * Added doc/gripDeltaT.py, the script used to generate doc/figs/gripDeltaT.png * Added examples/trypetsc/petsc-copyright.html, just to be on the safe side.Constantine Khroulev4+78-6
2008-10-21 23:11Fiddles with new installation manual. Sorry if I stepped on your toes Constantine ...Ed Bueler4+99-102
2008-10-20 22:28Added a PETSc tutorial example (src/snes/examples/tutorials/ex5.c; the Bratu example). This way we can say to users with build problems 'try building this one to see if it is a PETSc problem'. Note that some users really don't have the full PETSc source already.Ed Bueler3+644-0
2008-10-20 02:34Slight changes to BombProofSacrifice and CFL violation reporting at different verbosity levels.Ed Bueler2+20-9
2008-10-19 04:06O.k. The changes of several nights ago are now commited correctly. Sorry to spam everyone.Ed Bueler1+33-24
2008-10-19 03:25Still trying to fix my screw-up of manual update.Ed Bueler1+68-77
2008-10-19 03:08Updates to MISMIP part of User Manual; thank god for .svn/tmp/ files which saved me from losing an evening of work.Ed Bueler1+92-30
2008-10-19 02:42Some changes to manual.Ed Bueler1+135-188
2008-10-18 23:09* Replaced doc/figs/gripDeltaT.png with a prettier picture. * Removed the "Installation" section from the User's Manual, fixed a couple typos in the EISMINT-Greenland section. * Added the svn:executable property on examples/eisgreen/preprocess.shConstantine Khroulev3+6-202
2008-10-18 20:13* Created the Installation Manual, doc/installation.tex. * Updated doc/Makefile. * Fixed a typo in src/base/comments.hh * A minor fix in util/nc_template.pyConstantine Khroulev5+614-5
2008-10-15 01:55Grainsize revised. Pseudoage mechanism eliminated. Actual age mechanism implemented; seems to work for 200ka to 300ka EISMINT II A run. New options: -gk, -gk 3, -gk_age; all turn on G-K with first using constant 1mm grains, second is 3mm grains, third is real age.Ed Bueler7+83-49
2008-10-14 17:44Various cleanup but grain size with pseudo age is still broken; see bug #11242.Ed Bueler7+29-25
2008-10-14 05:04I believe I have found and fixed a bug in examples/eisgreen/eisgreen.py which reverses the time axis in grip_dT.nc with the result that the option -dTforcing has no effect. (I want to get our of the business of modeling climate, actually ...) Note there are still debugging commands in src/base/forcing.cc which should be removed. I also split examples/eisgreen/bootstrap.sh into two: preprocess.sh and bootstrap.sh. Improved ssl2.sh and ccl3.sh. Reworked EISMINT-Greenland chapter in User's Manual.Ed Bueler11+328-264
2008-10-10 12:57* Fixed some (very minor) typos in ice_bib.bib * Modified nc_template.py to a) be able to include scalar variables; b) add polar_stereographic so that Unidata IDV can (potentially) display PISM input files for Antarctica.Constantine Khroulev2+10-5
2008-10-10 01:38Various maintenance: comments in build/Makefile give static linking suggestions; added math-y references in ice_bib.bib; text in User Man on verification; pst script maintenance.Ed Bueler5+81-39
2008-10-03 17:27Small mods to stdout during -bif bootstrap.Ed Bueler3+18-17
2008-09-27 07:51More MISMIP, naturally, including establishing new fiddle option which allows a bit of dragging on bottom of shelves. Lots of documentation fiddles.Ed Bueler18+160-131
2008-09-26 02:34MISMIP is making me work for it. Note code with beta_for_floating in iMssa.cc needs to be made official if it should be kept; WARNING that it is there. Renamed taub to taud because it stands for driving stress, of course.Ed Bueler9+284-167
2008-09-24 06:56More MISMIP, especially documentation of flaws. Add option -ssa_min_thk to control calving front force in some subtle manne. Use of -regrid to initialize thickness from result of Schoof asymptotic matching result; works.Ed Bueler11+170-63
2008-09-23 09:13createDA_done in IceGrid (grid.cc) clearly a relic. Sorry. And I had already created a relic like that in IceMISMIPModel, so I deleted it too.Ed Bueler4+5-14
2008-09-23 08:55More MISMIP: solverSM.py can write NetCDF with thickness, readable by 'pisms -mismip' for initialization to fix extreme thinning in experiment 3.Ed Bueler4+340-180
2008-09-23 07:56Consistent function names for warning, see bug #12346.Jed Brown1+2-2
2008-09-23 02:57Fixed bug which wiped out boundary conditions on EISMINT-Ross. More work on MISMIP.Ed Bueler18+344-804
2008-09-19 01:09Work to finalize first MISMIP submission, and to document in User's Manual.Ed Bueler9+231-113
2008-09-18 02:56Added -Woverloaded-virtual to CFLAGS compile options, and associated 'using' directives in .hh files; this is generally helpful but does not address task #6212, which is not a problem of hiding. (One can still change the base class signature and loss functionality of the derived class version, which never runs because now the base class version does.) Improved MISMIP ascii output to make nice figures; this is cheap on top of required ascii output.Ed Bueler16+214-136
2008-09-17 17:50* Added code in IceModel::velocity() to zero out velocities if computeSIAVelocities is false. Changed adaptive time stepping to only apply diffusive time step restriction if computeSIAVelocities is true. * Changed IceModel::determineSpacingTypeFromOptions() to only change grid spacing type if asked. IceModel::setFromOptions() now doesn't change anything if there are no options set. * Removed MISMIP_PLAY compilation environment variable.Ed Bueler20+397-248
2008-09-14 21:05quick backup/transfer of MISMIP stuffEd Bueler2+20-4
2008-09-14 17:39A bit more work on MISMIP, including reattaching IceMISMIPModel::basalDragx(). Still have very irregular flux near grounding line, but steady state is achievable, at least with MISMIP_PLAY=1 smoothing of C at grounding line, and perhaps without.Ed Bueler5+110-58
2008-09-14 06:17Additional work on MISMIP. Found cross-flow asymmetry bug; now x direction is correctly periodized. And IceGrid::rescale_using_zlevels() allows TruelyPeriodic in one direction but not the other. Added MISMIP run script.Ed Bueler14+218-265
2008-09-11 22:32Small changes: (i) added references and reformat for manual; (ii) changed initialization details for PST experiments.Ed Bueler5+67-22
2008-08-15 00:21Replaced pycdf with netcdf4-python everywhere. See http://code.google.com/p/netcdf4-python/ To install netcdf4-python to use with NetCDF 3.6.x, untar and doConstantine Khroulev17+209-229
2008-08-13 08:11* -skip replaces -tempskip; new version also skips SSA updates. Seems to test well for PST experiments.Ed Bueler18+228-197
2008-08-11 01:07Preliminary change to using -tempskip for skipping SSA calculuations. Need to test using PST experiments. Also added references in doc/ice_bib.bib.Ed Bueler6+85-51
2008-08-04 20:55util/series.py now writes UNLIMITED as size of t dimension, so NCO tools using record dimensions work, for example. Updated (and perhaps finalized) scripts for IcePSTexModel examples in examples/pst/.Ed Bueler4+342-43
2008-07-30 20:31Cleanup of documentation, especially comments and references for exact solutions.Ed Bueler12+45-53
2008-07-27 06:58Bug fix: changed basal vertical velocity to zero (or basal melt rate), instead of a function of bed slope (which is appropriate for grounded ice). Also: improved diagnostics cbase, csurf to only show speed where there is ice, and not show artifacts from SSA calculation.Ed Bueler3+37-8
2008-07-22 19:57Maintenance:Ed Bueler5+38-18
2008-07-17 23:52Maintenance on EISMINT-Greenland example. Note pgrn no longer requires -ocean_kill, which is built in; roughly 'pgrn' includes 'pismr -ocean_kill -e 3 -tempskip 20'.Ed Bueler6+108-115
2008-07-13 08:37Maintenance on EISMINT-Greenland tutorial example.Ed Bueler4+59-86
2008-07-13 03:11Bug fix: TsOffset needed to be set to zero in IceGRNModel::updateTs() so that the -dTforcing temperature offset would actually be added. We will see substantial effects in the CCL3 run.Ed Bueler12+112-27
2008-07-09 01:51Trivial changes, to synchronize versions.Ed Bueler2+10-3
2008-07-07 22:45* Fixed a couple of typos in doc/manual.tex. * Modified the svn:ignore property on doc. * Some more changes to Makefile: 'make update' finally works the way it should; 'make clean' cleans both build/ and doc/, plus it is possible to build userman, refman, browser, summary and fullbib from the top-level directory now.Constantine Khroulev2+10-3
2008-07-07 19:43Added option -float_kill, which is calving at the grounding line according to the floatation criterion. Note that -ocean_kill is calving at the location of the original calving front. Also added a post-processing script for the PST experiments, which uses several good tools.Ed Bueler7+292-10
2008-07-03 09:36I believe I have fixed -dSLforcing option so it adjusts sea level correctly. Also made -tempskip 20 automatic for pgrn. pgrn -ccl3 should run without forcing, e.g. for climate warming scenario at positive time.Ed Bueler9+65-112
2008-06-26 23:37* Modified test/vnreport.py to use matplotlib instead of Gnuplot.py. Updated doc/manual.tex to reflect the change. * Regenerated doc/figs/*errs*.pdf using the new test/vnreport.py * Added the svn:ignore property to build/, lib/, {src,doc}/revision. Updated svn:ignore on bin/ * util/nc_template.py does not include 3D mode_state variables by default now. * Minor changes to doc/Makefile. * Finished documenting nc_template.py and nc2mat.py. Described changes to [build/]Makefile.Constantine Khroulev10+83-55
2008-06-26 19:28* Fixed build/Makefile: made it work on marmaduke (which has GNU Make 3.80 instead 3.81 on my machine)Constantine Khroulev2+7-5
2008-06-26 18:59* Improved Makefile and build/Makefile; updated INSTALL; moved SConstruct to build/. It should work better (namely, give less warnings) now. * Edited build/Makefile and src/base/iMutil.cc to add revision stamping. * Note: 'make update' should be usable by now.Constantine Khroulev6+182-158
2008-06-26 01:13Lots of changes to Makefile; added the build/ subdirectory to reduce the clutter... I hope this works. (Do blame me if it doesn't. :-) )Constantine Khroulev3+223-199
2008-06-25 00:47tillphi was not being read under -if; fixed. Also -regrid_vars now only takes single character flags bBehHLT, for the model state variables. There is no regridding of climate.Ed Bueler7+50-64
2008-06-25 00:46* Modified doc/Makefile to use vfnow_r327_L5_23june08.txt and regenerated figures (i.e. ran 'make vfigs').Constantine Khroulev8+23-2
2008-06-24 08:06Fixed another asymmetry, I think. This time kind of trivial: average speed reporting for IcePSTexModel.Ed Bueler1+6-3
2008-06-24 07:33Noticed PST experiments had asymmetry. Diagnosed as failure to communicate Hmelt before explicit diffusion step in diffuseHmelt(). Fixed. Trivial fiddling with velocity().Ed Bueler2+14-33
2008-06-24 02:09Intended to be final version of PST experiments ... we'll see ...Ed Bueler2+83-44
2008-06-23 23:40Default for IceModel::useConstantTillPhi should be FALSE. ALso completed level five (rev 327) verification on 8 core machine for BCELGIJK in only 15 hours or so, so a copy is put doc/.Ed Bueler3+278-4
2008-06-23 23:21* Changes to the User's Manual: started documenting nc2mat.py and nc_template.py (more work on this later). * Modified pism_state.cdl: removed the pism_intent attribute from x,y,z and zb, removed last signs of ncgen.rb * Renamed pism_netcdf_template.py to nc_template.py. Made this script include model_state, mapping and climate variables only. * Fixed a bug in nc2mat.py * Added examples/ant/ant08.pyConstantine Khroulev6+344-82
2008-06-23 07:54Realized that the velocity combination -super did not have quite the right effect on the SIA terms in the mass continuity explicit stepping; fixed it. Gave massContExplicitStep() its proper name.Ed Bueler5+70-28
2008-06-21 18:38Attempt to speed SIA by making SIA-sliding a special case; it is bad so we should generally assume the SIA value for basal sliding is zero. Also cleaned up comments in temperatureStep() and realized verification tests A B C D E L all did better with option -no_eta.Ed Bueler5+102-91
2008-06-20 01:56The diagnostic driving stress variable in the output file is now correctly named 'taud' not 'taub'.Ed Bueler4+29-29
2008-06-17 20:53Fixed reportBIFVarFoundMinMax() bug; min and max are now properly globalized. Updates to bibliography for manuals.Ed Bueler2+17-9
2008-06-17 02:07* Noted that assembleSSARhs() did not need to be called repeatedly while iterating the SSA; took it out of the loop. * Now believe that eta=H^{8/3} transformation for computing surface gradient is almost always beneficial. Made it the default. Replaced option -grad_from_eta by -no_eta. * Maintenance on -bif default bootstrapping. Now will read tillphi if present. Also gives better default behavior and standard out messages. * Removed prototype for cryptic flags at start of run; the flags themselves are really useful and remain. * Makefile now generates dependencies without calling linker and getting stupid error messages.Ed Bueler12+171-124
2008-06-15 19:16Playing with greater use of eta=H^{8/3} transformation to regularize computation of surface gradient. In particular, it is now used in assembleSSArhs() to compute driving term in SSA solution, and diagnostically when computing taub, the driving stress. PST experiments will be redone with this choice. Additional small changes (Glen exponent, bibliography).Ed Bueler11+153-58
2008-06-13 08:19Reorganized test/ subdirectory: 1) test/ only contains scripts for testing PISM, for now only verification scripts; 2) util/ contains the documented utility scripts in Appendix D of the User's Manual; util/ should be on PATH; 3) examples/ contains subdirectories with examples.Ed Bueler39+243-395
2008-06-11 21:59* Changed eis_green.py to use 'missing_value' outside the valid range, as recommended by the NUG. (The value I picked is -9999.0 -- for the lack of a better choice). Tried switching to '_FillValue', but it breaks mysteriously if I do that. * Fixed 'missing_value' interpretation in fill_missing.pyConstantine Khroulev2+10-5
2008-06-11 02:28* Finished working on the new fill_missing.py: - added the -i (--initial_guess) option - it uses the SOR with Chebyshev's acceleration now, closely following the algorithm described in 'Numerical Recipes' - it follows the CF Metadata conventions (regarding valid_range, valid_min, valid_max, _FillValue and missing_value interpretation) - it can be used both as a command-line script and a Python module - it uses a temporary file to avoid overwriting the output file with junk in case of failure * Added nc2mat.py -- a general-purpose script converting from NetCDF to MATLAB's .mat binary data file format (with the help from pycdf and scipy) * Added pism_netcdf_template.py -- a helper script creating an empty PISM-readable file with given grid dimensions by reading pism_state.cdl and running ncgen. This script can also be used as a Python module. * Updated the User's Manual to reflect the changes in fill_missing.py and finish the transition from ross_plot.m to ross_plot.py * Replaced MATLAB-generated pictures (rossquiver.png and rossscatter.png) with Python-generated ones. * Fixed a typo in pism_state.cdl.Constantine Khroulev7+487-156
2008-06-09 23:12More work on PST experiments: smooth till phi transitions. Coming soon: Test M is rotated and 2d (as opposed to 1d) version of Test I from Schoof.Ed Bueler5+218-178
2008-06-06 08:42More work on IcePSTexModel: creating smoother till phi maps. (Not done.) Also clean pst.sh and SSA progress reporting.Ed Bueler4+88-96
2008-06-04 21:41script test/pst.sh has better approximation of intended experimentsEd Bueler1+56-61
2008-06-04 09:46Most work done on reconfiguring plastic till stream experiments, I think. Need to try it out on faster machine.Ed Bueler3+233-152
2008-06-03 23:46Restart on plastic till ice stream experiments. Now each run has three ice streams so that parameter studies are easier.Ed Bueler8+585-503
2008-06-01 20:31Various fiddles: (1) fixed standard out from IceEISplModel to agree with others. (2) Made options on python scripts agree more, and documented more. (3) Improved documentation on installation.Ed Bueler10+185-163
2008-05-31 02:12Finalized removal of Ruby dependecy (ncgen.rb). Converted Sigma to J/(s m^3) units; before was K/s. New units compatible with simple formula Sigma = D_ij tau_ij, and with new paper. (Old units appeared in Bueler et al 2007 J. Glaciol.)Ed Bueler12+62-172
2008-05-30 22:43Forgot to turn on new version of correctSigma(). Also edited IceType comments, etc.Ed Bueler3+24-22
2008-05-30 22:28Made test/vnreport.py produce PNG by default or PDF as option; documented in User Manual.Ed Bueler20+403-354
2008-05-29 04:21Attempt to fix vertical advection at ice base bug in temperatureStep(). Removed correctSigma() from velocity(), so Sigma is just SIA value; note frictional heating still occurs in sliding. Fiddle with error messages.Ed Bueler3+28-12
2008-05-28 06:42Back to work on plastic till experiment. Velocities now combined by convex combination. -superpure option removed; -super remains. Minor fiddles with manual.tex, ross_plot.py, test/ross/README.Ed Bueler10+169-152
2008-05-28 02:37Set svn:executable on eis_ross.py and ross_plot.pyConstantine Khroulev2+0-0
2008-05-28 01:25Edited test/ross/ross_plot.py: added more about installing scikits.delaunay and its dependencies in general.Constantine Khroulev1+18-9
2008-05-25 17:22Doc cleanup: delete (*large*) old figs/*.eps. Additions to bibliography related to NASA ROSES proposal.Ed Bueler11+5-169119
2008-05-24 00:46Edited (fixed and improved) test/ross/ross_plot.py.Constantine Khroulev1+60-31
2008-05-23 02:20* Added test/ross/ross_plot.py -- an alternative to ross_plot.m which uses open-source libraries only. :-) * Fixed two typos in doc/manual.tex.Constantine Khroulev2+91-3
2008-05-21 17:43* Added amsmath to "extra packages" in doc/Doxygen_html (it is needed for the align* environment in doc_tempage.hh, iMinverse.cc, iMssa.cc and nc_util.cc in src/base). * Edited doc/Makefile and test/vnreport.py to generate figures in PDF instead of creating .eps figures and converting them to .png (There's one less thing to break this way, and we get scalable figures, too.)Constantine Khroulev18+28-47
2008-05-16 23:47Added the pism-state-ncgen.py (a replacement for ncgen.rb). The next step is to verify that it works and remove Ruby from the list of PISM dependencies in the User Manual.Constantine Khroulev1+141-0
2008-05-16 19:29Fixed a typo in the user manual (section 3.1, 'Getting Started').Constantine Khroulev1+1-1
2008-05-15 00:37Commented out 'include config/macosx_macports' in the Makefile, since it should be that way by default. (Sorry.)Constantine Khroulev1+1-1
2008-05-15 00:25* Added config/macosx_macports * Added some notes about installing PISM on Mac OS X 10.5 (doc/pism-install-macosx-notes.txt) * Modified Makefile to allow redefining switches needed to link shared libraries * Modified src/base/iMIO.cc to use strlen instead of strnlen, since strnlen is a Linux extensionConstantine Khroulev4+124-4
2008-05-13 23:10Bibliography and install notes updates.Ed Bueler2+101-53
2008-04-26 23:23Debian install of PETSc now works for me; see doc/pism-install-deb-note.txt. Makefile default is to use Jed's sed trick of mpicc-->mpicxx in linking.Ed Bueler4+191-5
2008-04-20 19:40Manual update to put hierarchy on options documented in Appendix A.Ed Bueler4+226-140
2008-04-20 05:18Various maintenance: expansion of doc/ice_bib.bib; addition of -tempskip 10 to test/eis2plastic.sh; tucked away ryan/ and embayment/ in less notable locations.Ed Bueler17+257-15
2008-04-02 03:31Added a diffusion for Hmelt. The Hmelt values under the margin of the ice are not as pretty with BOMBPROOF as they were with the more explicit previous scheme. Should not be an adaptive time step restriction. More work on iMinverse.cc.Ed Bueler3+85-14
2008-04-01 20:30Found and fixed initialization bug. More work related to inverse modeling.Ed Bueler11+97-49
2008-04-01 05:33Initial implementation of iMinverse routines completed. Note that there is viewer for till friction angle with '-d I'. To test ...Ed Bueler8+253-156
2008-03-31 07:23Progress toward inverse modeling as part of mainstream PISM, according to options -cbar_to_till or -csurf_to_till.Ed Bueler7+313-110
2008-03-31 00:44Task #5879 done. Continuing work on minimal inverse method using pseudo-plastic till.Ed Bueler18+204-157
2008-03-30 03:22PISM model state files now a bit bigger; four more 2D fields always written. Most important is 'tillphi', the till friction angle; in some cases this is a model state variable which will be read in for continuation. Others are diagnostic: sliding speed 'cbase', driving shear stress at the base of the ice 'taub', yield stress 'tauc' (applicable for plastic or arbitrary pseudo-plastic law; can even be part of linear sliding).Ed Bueler6+172-10
2008-03-28 01:29Start on new version of inverse modeling job: read velocities (either cbar or csurf), remove vertical shear from SIA, apply MacAyeal nonlinear operator, use pseudo-plastic law to extract (pseudo-)till yield stress, use modeled basal water etc to extract till friction angle, treat that as time-invariant. Also dealt with task #5688, so nicer behavior when reading input files.Ed Bueler13+2139-1670
2008-03-25 20:35Bug fixes having to do with test H and test I,J.Ed Bueler8+179-157
2008-03-25 05:23Cleanup of BOMBPROOF documentation. Addition of pseudo-plastic till law, with new options -pseudo_plastic, -pseudo_plastic_q, and -pseudo_plastic_uthreshold. Has resonable effect on Test I; now repeating plastic stream experiment P0.Ed Bueler9+289-92
2008-03-23 08:44Doc cleanup associated to BOMBPROOF implementation. Verification and EISMINT II look good for that scheme.Ed Bueler8+108-244
2008-03-23 07:19Initial, successful implementation of BOMBPROOF, a scheme for vertical advection which has no vertical CFL but has maximum principle stability. Good verification restults; great consequences for speedup with -tempskip.Ed Bueler9+175-452
2008-03-22 20:42More doc maintenance.Ed Bueler5+140-53
2008-03-22 04:30Documentation maintenance. The new temperature scheme is not ready.Ed Bueler5+221-31
2008-03-21 20:59Working on a robust but faster advection scheme for temperature and age. The current one involves a vertical CFL; my goal is to remove this and gain accuracy too by doing things more implicitly in the vertical.Ed Bueler7+691-85
2008-03-20 01:41Small change to semantics of MASK_FLOATING_OCEAN0 to make it hydrophobic even if land rises above sea level.Ed Bueler4+35-11
2008-03-19 01:46Bug found with 'pisms -eisII' and bed deformation model: bug fixed. The bedrock related part of task #5787 has been done; IceModel now has a BedThermalType member and a DeformableEarthType member. Index added to the User's Manual.Ed Bueler13+218-200
2008-03-16 06:21Caught bugs in climate forcing mechanism.Ed Bueler6+102-56
2008-03-16 00:59Slight change to pism history string. Source code cleanup.Ed Bueler3+65-66
2008-03-14 07:10Caught bug with temperature at top of bedrock under ocean. Added separate source file for tests; should become a derived class, perhaps.Ed Bueler6+194-17
2008-03-13 08:03Various bits of cleanup: HEINO out of the way; iMadaptive.cc contains its stuff, citations in doxygen.Ed Bueler17+609-421
2008-03-11 07:35Added ability to read monthly temps from file, so that PDD model can use data for its yearly temperature cycle; option is -pdd_monthly_temps. Moved the -dTforcing and -dSLforcing options, and associated instances of IceSheetForcing, into base class IceModel. EISMINT-Greenland derived class IceGRNModel very minimal now. Apparently grain size update is broken; changed flow law to GK to (temporarily) always have 1mm grain size; see iMsia.cc.Ed Bueler20+519-238
2008-03-09 19:32Fixing stupid matching '...' behavior in python.Ed Bueler6+8-7
2008-03-09 18:56Trivial formatting stuff, mostly. Substantive change to vertical spacing in bedrock: now equally spaced with spacing equal to dzMIN within the ice.Ed Bueler4+114-76
2008-03-08 06:29Little refinements re making pgrn work with Tolly's data.Ed Bueler8+196-124
2008-03-07 01:02Changes to bootstrapping to make it more flexible; see task #5802 for some remaining flexibility issues.Ed Bueler1+317-220
2008-03-06 01:58MISMIP essentially implemented. Saves correct files; see comments at head of iceMISMIPModel.cc. FUNDAMENTAL PROBLEM remains: PISM is not doing a good job of flux continuity at grounding line.Ed Bueler8+350-139
2008-03-03 07:41Playing with MISMIP. Small change to updateViewers() to avoid wasting time when viewers not needed.Ed Bueler5+240-168
2008-02-14 20:23Cleaning up EISMINT II plastic-till-version-of-experI. Switching to use non-equally spaced vertical grid.Ed Bueler3+12-43
2008-02-14 00:16Added executive summary of PISM (needed that anyway). Changed the plastic till version of eisII I to use unequally-spaced vertical. Also improved the script itself.Ed Bueler5+1522-69
2008-02-12 07:37Fix up docs in trunk.Ed Bueler3+25-3
2008-02-12 05:17Maintenance: improved documentation of IceGrid, options in manual for unequal spaced grid, transform to eta for surface gradient ...Ed Bueler11+148-52
2008-02-11 03:15Ongoing general maintenance. Less memory is used by refactoring the grain size scheme; the new lighter weight scheme breaks grid symmetry when using pseudo age but not when using real age. (It only makes any difference at all when using Goldsby-Kohlstedt.)Ed Bueler49+458-511
2008-02-09 19:59Improved reference manual introduction. Default values are *all* not members of IceModel anymore; this will clean up initialization.Ed Bueler16+326-257
2008-02-08 02:04Reorganizing reference manual a bit; User's is shorter and Reference is longer. Reference is more readable because it starts with text written by a human.Ed Bueler3+254-182
2008-02-08 00:03Additional essential fix to regrid_global_var(). EISMINT-Ross now completely functional and documented. On any grid. In parallel. Now regrid_global_var() is used for: regridding, bootstrapping (both EISMINT-Greenland and EISMINT-ROss run the *same* code for default bootstrapping), and reading boundary conditions in EISMINT-Ross.Ed Bueler17+544-479
2008-02-06 01:36Bootstrapping now really works via regrid_local_var(). The format of a -bif file is now clear: it is a subformat of pism_state.cdl. EISMINT-Greenland works again; still needs a full go-through-and-try-each-line-in-Users-Manual. EISMINT-Ross is broken for now: fix scripts to produce proper format. Also need to document unequal-spaced vertical grid in Users Manual. Cruisin' toward first stable release.Ed Bueler16+262-161
2008-02-05 07:41Miscellaneous updates.Ed Bueler5+55-56
2008-02-05 04:47Eliminated NCTool::var_to_da_vec() and associated VecScatter to processor zero and integer indexing scheme. All functionality replaced by that in regrid_local|global_var(), which now works for arbitrary spacing in the vertical. Note regrid_global_var() now has ability to regrid an integer mask using a MaskInterp tool. Also: attempted to assure ANSI C compatibility for all .c source files; attempted, by adding CXXLINKER to Makefile, to help fix build errors associated to PETSc linker-choice bug and derived PISM bug #9371.Ed Bueler20+480-488
2008-01-30 01:55Regridding implemented for non-equally spaced vertical grids. __PLEASE TRY TO BREAK IT AND TELL ME ABOUT IT___ It remains to remove var_to_da_vec() from bootstrapping, replacing with regrid_local_var().Ed Bueler5+416-243
2008-01-29 18:44Comments regarding parallel regridding and irregular grid.Jed Brown1+157-47
2008-01-26 06:47Bug fix revealed by test E: communication of vWork2d[] arrays had to happen sooner, within velocity(), for basalSIA().Ed Bueler5+19-17
2008-01-20 23:29I think unequal spaced grid works well (i.e. no better than lots of equal spaced levels, but not much worse, and with some speedup), except: regridding and Test K results. IceParam *p removed from IceGrid; removes 'p->' from everywhere; remember PetscBag? iMlegacy.cc no longer even compiled.Ed Bueler39+813-790
2008-01-19 18:08Unequally-spaced vertical implementation still not stable. Verification in Test G pretty good. Init -if and regrid still need work. Now removing legacy code as too hard to maintain.Ed Bueler23+435-352
2008-01-17 04:03Builds. Test K and Test F results not great but not silly.Ed Bueler4+148-29
2008-01-16 18:50Option -cheb now implemented; produces vertical grid spaced like the [-1,0] portion of the Chebyshev extreme points. Aspects of temperature solution, with and without bedrock, still need addressing. Test K verification not so good yet. Note bootstrap stuff separated into src/iMbootstrap.cc. Note src/iMIOnetcdf.cc moved into src/iMIO.cc. Regridding for unequal spaced grids not yet implemented. THIS REVISION DOES NOT LINK right now because IceModelVec3Bedrock::getValColumn() and setValColumn() are not yet implemented.Ed Bueler24+1600-1440
2008-01-14 01:34Ongoing structural cleanup and documentation related to non-equally-spaced vertical grid and Vec abstraction. Of the numerical models, only the temperature temperatureStep() uses an internally equally spaced grid at this point; it then linear interpolates to the storage grid. This can be changed later if desired; verification and profiling should determine what happens. Verification shows that the linear interpolation is doing no harm if the storage grid actually has equal spacing.Ed Bueler21+442-285
2008-01-12 19:26Bit more work on removing equal spacing dependence. Revised and documented grain size model.Ed Bueler12+176-102
2008-01-11 18:04Abstraction of 3d Vecs now in good shape. Still equally spaced. Side effects have been: (1) some slow-down from lots of linear interpolation code and (2) some multiplexing of communication because each 3d Vec has its own DA and there is no conflict if we try to begin several and then end several LocalToLocals.Ed Bueler20+541-419
2008-01-10 19:54Critical fix to make -Mz option work. Note Test G is not performing as well as in pre 235 revisions.Ed Bueler1+3-0
2008-01-10 19:25Abstraction of 3D Vecs mostly completed, including former vWork3d.Ed Bueler25+1056-534
2008-01-08 18:47First implementation of abstraction of 3d Vecs now works. Some slowdown with current fine equally-spaced vertical grid, of course; no actual implementation of non-equal grid yet. verifynow.py essentially same output.Ed Bueler16+548-330
2008-01-06 22:17Major revision mostly done; won't build yet. 3d Vecs are abstracted, in preparation for changing to unequally-spaced grid. Still need to complete in these ways: (1) iMvelocity.cc, (2) all uses of vWork3d[], (3) NetCDF read and write, (4) communication simultaneous with copying.Ed Bueler21+861-425
2008-01-05 21:35Added horizontal ice flux to PISM output NetCDF files. A bit more work on MISMIP. A start on abstracting da3-based Vecs.Ed Bueler9+600-32
2008-01-03 23:01Initial implementation of MISMIP: not to be trusted until more seriously tried. Needs explicit treatment of calving front. Needs text output routines. Needs management scripts.Ed Bueler7+519-168
2008-01-02 16:59PDD model probably done; see trials in comment at end of src/base/iMpdd.cc. *Monte carlo method produces results in close agreement with (Calov and Greve 2005). Note Calov-Greve clearly needs more than monthly evaluations for quadrature; at this point we use weekly. PDD re-documented in manual. Maintenance. Superficial start on MISMIP; not functional at all.Ed Bueler11+349-139
2008-01-01 18:59Debugging associated to Calov-Greve PDD model. Maintenance.Ed Bueler20+352-371
2007-12-29 00:23Modifications to summary line: step moved to flag line. Initial implementation of (Calov and Greve 2005) positive degree day scheme; not trustable without additional testing.Ed Bueler10+335-178
2007-12-22 18:28Ongoing fiddles to plastic till mod of EISMINT II exper I. Work on manual.Ed Bueler4+136-96
2007-12-20 04:46Fixed and simplified and improved docs on regrid.Ed Bueler9+96-118
2007-12-19 21:31Fiddlesticks.Ed Bueler1+2-2
2007-12-19 21:26Update to plastic till experiment script: add bedrock thermal.Ed Bueler1+9-9
2007-12-19 21:13Update to plastic till experiment script.Ed Bueler2+108-81
2007-12-18 23:56I believe I have completed the conversion to CISM names for NetCDF variables. Units still do not match (in general). Modifications of the plastic till SSA EISMINT II experiment I code.Ed Bueler8+220-151
2007-12-16 07:50Switched to GLIMMER names for variables in PISM NetCDF files. Units not changed because we conform closer to CF-1.0 than GLIMMER, I think. Also changes to tools and manual. Need to modify scripts for EISMINT-Greenland and EISMINT-Ross to use new variable names.Ed Bueler17+774-524
2007-12-05 04:59Small fix to test/eis2Iplastic.sh.Ed Bueler1+7-7
2007-12-05 04:53Improved reporting and time series extraction. Manual needs revision accordingly ...Ed Bueler10+171-101
2007-12-04 20:54Script modifications: (i) fix of fill_missing.py to fix for old pycdf(?); (ii) clarification of eis2Iplastic.sh.Ed Bueler2+61-30
2007-11-30 07:22Cleaned up viewer sizing and standard out summary. Simplified src tree structure slightly; it is actually useful to have some structure, but I had overdone it.Ed Bueler39+323-288
2007-11-26 10:05Theme and variations on EISMINT II experiment I with plastic till modifications preliminarily set up.Ed Bueler4+134-43
2007-11-26 01:02Various fiddles related to plastic till modification of EISMINT II experiment I.Ed Bueler10+40-28
2007-11-24 00:50Cleaning up. IceModel now allocates about 30 percent less memory, incidentally. More comments.Ed Bueler7+85-113
2007-11-23 21:02Refactoring. (Need to eliminate pernicious influence of SIA basal sliding, though it is still possible, and pernicious influence of ISMIP-HEINO in particular.) New mass continuity scheme should be strict improvement. A few more comments. Working Vecs made private in IceModel.Ed Bueler10+295-272
2007-11-21 22:17added script for verifying upwinded method for mass continuityEd Bueler2+46-0
2007-11-21 08:11Testing of upwind method for mass continuity for SIA; apparently a new explicit method for SIA.Ed Bueler6+140-18
2007-11-20 23:49Furtherwork on plastic till modification of EISMINT II exper I. Added lake and ocean possibilities. Moved code into derived class IceEISplModel.cc which is derived class of IceEISModel (derived from IceModel).Ed Bueler7+226-125
2007-11-18 21:05Fixed missing communication in iMssa.cc, I hope.Ed Bueler2+22-13
2007-11-18 08:39Addition of option -ssa_dt, which allows the SSA to only run every few steps. Should speed up plastic till superposed runs, in particular. Looked at EISMINT-Ross again; mostly in good shape. Tests I and J still do fine.Ed Bueler15+159-184
2007-11-14 23:39Yet more work on EIS II exper I and superposed plastic till SSA.Ed Bueler3+26-28
2007-11-14 22:46Continued work on mod of EIS II exper I with plastic SSA addition.Ed Bueler6+51-25
2007-11-09 07:41Fiddling with reporting. Now the flags for which model has run and the summary quantities (vol,area,melt fraction,...) are on separate lines. Additions to bibliography and manual.Ed Bueler13+210-79
2007-11-05 01:07EISMINT-Greenland more-or-less fully documented. In any case, I'm done with it for now ...Ed Bueler6+22-7
2007-11-01 19:15Additions to manual. Minor simplification of directory structure in src.Ed Bueler16+1014-992
2007-10-31 07:14Little stuff.Ed Bueler3+14-5
2007-10-30 22:46Work on EISMINT-Greenland tutorial example. Manual work on that.Ed Bueler6+79-15
2007-10-29 19:37Improvements to tools for EISMINT-Greenland tutorial example.Ed Bueler4+150-97
2007-10-26 22:29Woops .. EISMINT-Greenland has no sliding.Ed Bueler2+11-10
2007-10-26 21:36Maintenance on EISMINT-Greenland. Improvements to that part of manual. Not yet done ...Ed Bueler9+186-124
2007-10-23 20:47Improvements to manual and source browser. Cleaned up trough topography code in IceEISModel. Made options -low_temp and -max_low_temps, for counting 'too low' violations in temp eqn.Ed Bueler8+115-51
2007-10-19 07:28Work on plastic till experiment based on EISMINT II experiment I. Additions to manual along these lines.Ed Bueler6+88-101
2007-10-16 07:25More work on the EISMINT II experiment I stuff.Ed Bueler5+38-7
2007-10-16 06:18Work on a modification of EISMINT II experiment I. The only change to the data of the problem is to assume that the till friction angle in the trough region is 5deg while elsewhere it is 20deg. Then Shoof's model is used, but determining the basal yield stress through the stored basal water. Note that there is a new viewer for basal driving stress: -d Q. Not that IceModel has another 2D Vec: vtillphi = the till friction angle.Ed Bueler17+305-148
2007-10-12 22:38A little bit more work on the manual.Ed Bueler14+202-125
2007-10-10 10:10I have now automated the process of verifying PISM and reporting the results graphically. Scripts verifynow.py and vnreport.py are used in sequence. This is now incorporated into the user's manual; see pism/doc/Makefile.Ed Bueler11+237-161
2007-10-09 22:44New script test/vnreport.py processes verification outputs from test/verifynow.py. Creates postscript files from gnuplot. Resulting plots will go in user manual.Ed Bueler4+195-30
2007-10-05 22:40More temperature-verification-related changes. First, it turns out that the geothermal flux boundary condition at the base of the bedrock was of lower order; see section 2.13 of Morton and Mayers. Once this was fixed I got O(dt^1,dz^2) convergence from Test K, which is optimal. I submitted preprint 'An exact solution to the temperature equation in a column of ice and bedrock' to arXiv.org. This preprint discusses Test K and verification; I am reminded that though verification documents are boring, the process is quite exciting because you have to get it perfect. Then I fixed all basal thermal flux boundary conditions in temperatureStep(), including at the base of the ice when there is no bedrock and at the base of a floating ice shelf.Ed Bueler14+117-91
2007-10-04 22:15Lots of improvements to verification: (i) Removal of run() in IceCompModel, and replacement by additionalAt[Start|End]TimeStep(). This helps maintainability. (ii) Replacement of updateTest*() by compSourcesTest*() and fillSolnTest*(); ditto re maintainability. (iii) Fixed up Test K. (iv) Fixed unreported bug in Tests A and E. Also fixed setFromOptions() to correctly handle boolean flags.Ed Bueler12+556-438
2007-10-01 19:18Exact solution 'Test K' ready to go. It is documented in a separate preprint, which I will post at arXiv.org. Test K is for the pure conduction problem in both ice and bedrock. It should help to debug the ice/rock interface part of the temperature solver (IceModel::temperatureStep()). In fact, we see convergence but only at a linear (O(dz)) rate; see verifynow.py output below. I believe it *should* be converging at O(dz^2). So will look into this.Ed Bueler7+124-26
2007-09-30 08:00Most work done on Test K. Basically works; try 'pismv -test K -d HTt -Mx 6 -My 6 -Mz 401 -Mbz 101 -y 135000 -maxdt 100.0' and note that the exact solution first has z=0 ice hit pressure-melting at about 133465 years. Still needs proper error reporting and also documentation in User's Manual.Ed Bueler8+435-40
2007-09-26 20:45Added a tool (pism/test/series.py) for turning PISM standard output into time series in a NetCDF file.Ed Bueler5+137-29
2007-09-26 05:51Several improvements: 1. Improved transfer of SIA velocities from staggered to regular grid. 2. Viewers for basal velocity (both -d and -matv). 3. Clarified save of ubar,vbar in preparation for SSA calculations. 4. Cleaned up IceRYANModel.Ed Bueler16+288-283
2007-09-25 06:02Fixed -constant_nu option so it should work.Ed Bueler1+6-3
2007-09-24 22:13Bug fix: dbdt term removed from incompressibility; it is incorrect when there is no mass balance. For now it is removed; it must be fixed up for long term.Ed Bueler4+17-16
2007-09-21 20:27Fixed up initialization in legacy bootstrap. Removed initSounding() as unnecessary.Ed Bueler7+94-56
2007-09-21 06:28I CAUGHT A BIG BUG. Or at least a long-lived bug. It was in putTempAtDepth() in iMIOnetcdf.cc and it meant that when using thermal bedrock in realistic circumstances we were making big errors (in the whole thermocoupled flow). It is now clear that much of what we diagnosed as surface roughness was this error.Ed Bueler4+80-48
2007-09-20 18:50Fixing up -bif_legacy. Also split long iMutil.cc; now iMreport.cc contains methods related to runtime outputs to standard out, esp. summary().Ed Bueler9+642-557
2007-09-10 08:13Cleaned up and completed the Matlab-readable views; nearly totally matches runtime graphical viewers user interface.Ed Bueler13+498-269
2007-09-09 07:27Renovation of diagnostic viewers code. Added IceModel::tn[], an array of names and titles. Add -matv option. Now the diagnostic viewers and the Matlab output more or less match in the user interface.Ed Bueler17+612-1075
2007-09-07 21:27Work on documentation. Work on variable names (when written/read from NetCDF or matlab) and short titles (e.g. on viewers). Should greatly ease conversion to CISM/GLIMMER naming conventions.Ed Bueler25+949-436
2007-09-05 01:34Working on documentation (i.e. comments which doxygen turns into pism/doc/refman.pdf and a source browser). Need to back up now as I am working outside and around kids ...Ed Bueler15+343-170
2007-09-03 07:49EISMINT II experiments I,J,K,L in better shape. Experiment I should be a basis for plastic till demonstration but still not getting it quite right; why? At least all signs of asymmetry are gone.Ed Bueler10+130-51
2007-09-02 01:11Test J now converges at O(dx^2) ON THE DOT (power is 2.0002 experimentally). THERE WAS NO FLAW IN JED'S macayeal equations code. Rather, test J truely uses a periodic grid and thus rescale() was changed to optionally accomodate that. Also L changed to 300 km in test J.Ed Bueler11+87-51
2007-08-28 23:00Preliminary completion of test J. Test J is a linearized periodic non-constant thickness ice shelf. (By 'linearized' and 'non-constant thickness' I mean that nu(x,y) H(x,y) = nu_0 H_0 even though H(x,y) is not constant.) The real significance is that the SSA differential operator (i.e. the left side of the 'MacAyeal' equations) is so well understood that one can know its exact eigenvalues, and thus apply iterative linear algebra with confidence.Ed Bueler8+149-132
2007-08-28 04:26Finally rewrote the vertical integral that gets vertical velocity from incompressibility. Better documentation of same. Now back to test J.Ed Bueler19+603-390
2007-08-26 02:28Work on implementing exact solution test J, which is a Fourier series solution to the SSA equations for an ice shelf in a linearized (and constant coefficient) case with periodic boundary conditions.Ed Bueler19+679-657
2007-08-24 00:31Forgot to add src/base/iMsia.cc. Also fiddled with Matlab output from SSA solution. Working on bug #9836 by building new nontrivial exact ice shelf solution (found by Fourier series soln). Fiddles with Makefile; no significant changes though.Ed Bueler7+622-136
2007-08-23 09:14In search of bug #9836. The pervasive 'MacAyeal equations' language has been replaced by 'shallow shelf approximation' (SSA) which is Weis-Greve-Hutter language. I have added comments to iMssa.cc (formerly iMmacayeal.cc) to indicate what continuum equations correspond to what routines.Ed Bueler21+1135-1480
2007-08-21 19:45In writing subsection 3.3 of the User's Manual a asymmetry of the numerical solution of the MacAyeal equations (actually Schoof plastic till case) was found. See bug #9836. Note all of EISMINT II is now in pism, including unpublished experiments.Ed Bueler9+115-48
2007-08-19 19:26Fixed a serious bug: EISMINT II experiments ABCDEF were all sliding on the base because of the renaming basal() --> basalVelocity(). This is fixed; in particular the main example in section 2 (Getting Started) now produces the usual spokes and not more extreme ones. Also: trivial changes to document build and on-going User's Manual improvements.Ed Bueler9+115-110
2007-08-14 19:50Start on using doxygen to automatically generate continuum model and numerical methods documentation (PDF); will also generate a class structure browser in (HTML). Read doc/README. Added EISMINT II heterodox experiments I,J,K,L; note experiment I will be used to document setting of mask and use of -mv and -plastic options. Renamed test/5k to test/green5k; these are utilities for reading NSIDC and PARCA Greenland data into NetCDF.Ed Bueler17+952-75
2007-08-08 04:36Added some directory structure to the test/ directory. Completed the redo of EISMINT-Ross: it is now parallel and at arbitrary resolution. See section 8.2 of the manual as it is pretty cool. Additional modifications to the manual.Ed Bueler17+265-462
2007-08-07 18:00So I did it: Directory structure added to src/ directory. (scons is hosed again; sorry) Also pismd -ross can now compute chi^2 against RIGGS data, and a hardness tuning script is included.Ed Bueler80+7017-6915
2007-08-06 19:17Further work on pismd -ross and documentation thereof. Generalized the forcing.[hh|cc] mechanism; added class Data1D of which IceSheetForcing is now a derived class. Modified constants DEFAULT_MAX_HMELT and a coefficient in excessToFromBasalMeltLayer() to represent a more conservative, slower production-of-basal-melt-water. (How should this mechanism be constrained, anyway?) Additional revisions to manual.Ed Bueler21+1147-715
2007-08-05 21:39More work on pismd -ross and related matters. Documentation thereof.Ed Bueler17+294-217
2007-08-04 00:33More work on EISMINT-Ross and the general mechanism for diagnostic computations of velocity. Basically works under pismd; see comment at start of src/pismd.cc.Ed Bueler14+351-722
2007-08-02 23:021. New PISM executable "pismd"; "d" for diagnostic. Purpose is to compute and save full velocity field from given geometry and temperature fields. May be used either with -if or -bif. Saves state with full 3D velocity field; see #3 below.Ed Bueler21+1665-166
2007-08-02 22:22fixed forcing bug with "size_t" data type. added accumulation to regridding (since it was forgotten). Other minor changesNathan Shemonski11+777-63
2007-07-27 20:48Oops, minor bug fix.Jed Brown1+1-1
2007-07-27 20:29Slight improvement in choosing C++ compiler if availableJed Brown1+11-3
2007-07-27 18:47Updated SCons build. It should be as rubust as the Makefile now plus install functionality. It autodetects availability of fftw3.Jed Brown5+90-53
2007-07-27 18:04Previous fix to initialization for derived classes broke pisms -eisII; this is fixed. Improved summary line for -tempskip. Cleaning on manual.Ed Bueler5+134-127
2007-07-24 22:18Fixed bug in IceGRNModel which overwrote regridding; note addition of doHook flag to overloaded IceModel::initFromOptions(). Now derived classes can init by first calling IceModel::initFromOptions, then doing their own init, then calling IceModel::afterInitHook(). (I hope I have not screwed up other derived class inits.) Work on manual.Ed Bueler4+52-95
2007-07-24 20:01Removed deprecated .pb regridding facilty.Ed Bueler6+35-301
2007-07-22 22:59Both -if and -bif read polar stereographic if present. Polar stereo info written at end of run. Moved legacy bootstrapping into separate source file; deprecated. Note commit to dogbert.gi.alaska.edu/home/jbrown/svn also: Antarctica close to -bif-ing same as Greenland.Ed Bueler8+658-480
2007-07-22 01:00Improvements to verifynow.py and additional verification documentation in manual.Ed Bueler3+265-156
2007-07-20 22:42Fixed bug in plastic till initialization. Playing with till material properties.Ed Bueler6+15-46
2007-07-20 20:23Added -plastic option, and associated calculation of till yield stress from basal melt water, to base class. (Previously only in pant.) This plastic stuff continues to need testing.Ed Bueler11+118-97
2007-07-20 02:14Climate forcing moved to its own class, namely IceSheetForcing in src/forcing.[hh|cc]. Only derived class IceGRNModel uses it at this point, but it could easily be moved to base class. Note eis_core.py writes different .nc files. iceGRNModel.cc is *a lot* shorter. Also, bootstrap bug found and fixed.Ed Bueler8+576-549
2007-07-18 23:09Bug fix for viewers in IceCompModel. Bug fix for creating Seq Vecs for -forcing. Makefile update for nelchina.Ed Bueler3+109-67
2007-07-18 20:25Changed the way bed elevation is changed for climate forcing. instead of saving the original bed, a bedDiff variable is saved.Nathan Shemonski7+60-44
2007-07-18 20:00Fixed bug in iMpdd.cc which was causing catastrophic ablation. (Greenland now grows a bit at steady state.) Removed .pb (PETSc binary) read/write code. Minor updates to manual and references.Ed Bueler7+135-241
2007-07-17 23:26Fixed bug with reading and (now) communicating time from .nc file by -if. Little additions to manual.Ed Bueler6+13-13
2007-07-17 18:30Revised and compile/linkable derived class IceRYANModel.Ed Bueler8+84-196
2007-07-14 21:01Added diagnostic maps to written .nc file (surface speed, surface vertical velocity of ice, dHdt, vertically-integrated horizontal ice speed; as diagnostics I put these in m/a units---wise?). Added section to manual on signals.Ed Bueler11+295-123
2007-07-13 23:26Significant refactoring and renovation of bed deformation code. The LC bed def model (-bed_def_lc) now works reliably on multiple processors on nosquare domains. Note new test program for the LC model: src/exact/tryLCbd.cc.Ed Bueler16+1184-628
2007-07-13 18:01added script "fill_missing.py"Nathan Shemonski5+185-31
2007-07-11 22:12Added command line arguments to eis_green.py; Began documenting EISMINT-GREENLAND tests in Realistic Ice Sheet Modelling section; changed run_until_done.py to ssl2_exp.py; eis_core.py now relys on specmap.017 from EISMINT website; Finished writing EISMINT-GREENLAND tests: ssl2, ssl3, ccl3, and gwl3 in pgrnNathan Shemonski8+431-211
2007-07-11 00:28I think I have rationalized the use of IceModel::startYear, IceModel::endYear and IceModel::grid.p->year. Also am preparing IceRYANModel.cc for Woodard; it is hidden away in subdirectories but may be more useful generally. Added table of -regrid_var options to manual. Fixed makefile library load order, I think.Ed Bueler16+812-99
2007-07-09 20:28We have a positive degree day model in the base class. See updated manual (section 3.4 and appendix A). Note some refinements to Makefile and to test L.Ed Bueler17+615-328
2007-07-08 02:031. Added "test L". It uses numerical integration (from gsl_odeiv.h) of a scalar ODE to compute the exact solution. The test has a nonflat bed but is steady and isothermal. I hope to use it to help identify slope computation issues; may relate to margin tricks but I don't know yet. 2. To speed the computation of test L I added heapsort to extrasGSL. The point is to do only one radial ODE solve for the whole grid (to initialize with and compare to the exact solution). 3. Because iceCompModel.cc was getting a little long, I split out thermo parts (tests F & G) into iCMthermo.cc. 4. I am learning on the Makefile; it is yet a little cleaner and shorter. Why were the implicit rules cancelled? Seems to not affect function either way. 5. We are now officially REQUIRING gsl. It is used in bed deformation model, in IceGRNModel, and in test L. All mentions of 'WITH_GSL' flag removed. Install instructions in manual adjusted accordingly. 6. nc_check() moved into IceModel (to avoid duplication) 7. clarified naming scheme for simple examples of exact solution evaluation (so there are 'simpleABCD', 'simpleE', 'simpleFG', 'simpleH', 'simpleI', 'simpleL' executables, and 'gridL' also) 8. changes to printed messages in bootstrapFromFile()Ed Bueler26+1651-830
2007-07-07 05:16Climate Forcing fully implemented.Nathan Shemonski5+78-125
2007-07-03 23:59nc_get_att changed to nc_get_att_text (again).Nathan Shemonski2+89-89
2007-07-03 23:37nc_get_att changed to nc_get_att_text. Also climate forcing is written but not tested.Nathan Shemonski5+145-60
2007-07-03 15:33Only use the C interface to netCDF.Jed Brown1+1-1
2007-07-03 07:25Added `pgrn' to SCons build system. Removed `iMregrid_netCDF.cc' from SCons since it was removed in r130.Jed Brown1+10-3
2007-07-02 22:05Temporarily fixed the bug with writing wrong accumulation data for Greenland Derived class. Also added in the python script that will run the steady state test (run_until_done.py).Nathan Shemonski5+120-4
2007-07-02 18:43Fairly widespread changes: 0. Makefile changed. Now "make" behaves correctly w.r.t. modified files, and "make install" puts libraries in lib/ and executables in bin/ and cleans up .o and .d files. Manual reflects this. 1. Change to pism_state.cdl: Made climate variables also depend on t (task #5200). Made t(t) a double (because we see effects of single precision all the time in long runs). Added pism_intent attribute where known; values are model_state, climate_steady, climate_snapshot, diagnostic. See comment near start of pism_state.cdl. 2. Modified nc_util etc to correctly pass around a double t, including for regrid. 3. iMregrid.cc now contains netcdf version (the default) and the unsupported (and probably not functional) old .pb version. iMregrid_netcdf.cc deleted. SHOULD WE GO THROUGH AND DELETE RELIC .pb STUFF? 4. bootstrapFromFile_netcdf() has different reporting on defaults. It should be fairly robust w.r.t. different kinds of -bif files. 5. small possible temperature bug changed: use ice material properties in k=0 segment of column if Mbz==1 6. setSoundingFromOptions() put in afterInitHook(); drivers changed to remove explicit call 7. improved the viewer code; note addition of utility functions getSurfaceValueOf3D() and getSliceOf3D(); note ideas in task #4989 still probably worth adding 8. Added better error reporting of pismv -test F|G: surface velocities and Sigma; does task #4184. THIS REVEALS test G DOES NOT CONVERGE FOR SURFACE VELS OR FOR Sigma, AS SUSPECTED FROM EISMINT II TESTS. This better reporting should help with implementing margin tricks (task #5244). 9. Small edits to iceGRNModel.cc. Changed set from options logic re enhancement factor.Ed Bueler28+736-462
2007-07-02 18:02More work done on Greenland derived class. GRIP data is in the process of being implemented. EISMINT data python scripts changed to use pycdf.Nathan Shemonski8+332-47
2007-06-30 01:01Fixes to EISMINT II tests: basal melt is supposed to be diagnostic only but PISM was including it in conservation of energy.Ed Bueler3+52-27
2007-06-28 18:57Mechanism for improved surface slope evaluation and map-plane flux evaluation esp. near margin. Is based on greater regularity of eta = H^{8/3} compared to H. (It is motivated by the Saito et al (2007) paper but is not the f.d. computation described there. To activate set env vars MARGIN_TRICK=1 and MARGIN_TRICK_TWO=1. Still under test. Seems to have mixed effects on tests B,C,G and only positive effects on EISMINT II, but even this conclusion is preliminary.Ed Bueler4+102-2
2007-06-28 18:01More work done on Greenland Derived class. Ablation Parameterisation added. Beginning of climate forcing added.Nathan Shemonski7+268-54
2007-06-27 20:48Added viewers for surface values of velocity. 0=surface hor speed, 1=surface u, 2=surface v, 3=surface w. (Note xyzXYZuvwUVW were already in use\!) Updated manual accordingly.Ed Bueler3+123-7
2007-06-22 22:55Fixed setting temperatures in Greenland derived class. removed manual.pdf from version control.Nathan Shemonski5+19-4
2007-06-22 20:27Bootstrapping in base class is finished. The Greenland derived iceModel class has been created with specific surface temperatures and removing of Ellesmere Island and Iceland. `make pgrn` has been added to the Makefile in order to make the executable to run the Greenland derived class.Nathan Shemonski11+670-46
2007-06-22 10:54SCons renovation. Now has an `install' target that respects prefix (defaults to pism build directory). There are also `install-bin' and `install-lib' targets.Jed Brown2+27-10
2007-06-22 02:07Changes to manual to note need for developer versions of libraries with header files. Also: (i) I think there is no need for copyright notices in scripts; (ii) do we want -v2 for ncgen?Ed Bueler5+18-28
2007-06-20 22:56Change of source file names to match executable names for pismr,pisms,pismv.Ed Bueler4+8-7
2007-06-18 17:27Starting Realistic ice sheet modelling Section. Just some beginning documentationNathan Shemonski3+202-0
2007-06-15 22:53Changes to manual based on Art's and Nathan's comments. Makefile now puts library files in pism/lib directory (amazingly enough).Ed Bueler4+783-770
2007-06-12 21:39Minor changes. Removed mention of environment variable WITH_NETCDF from code and Makefile; NetCDF is obligatory and it is an error to not have it. Cleaned up figures and eqns3D.tex so that both the manual (manual.pdf) and the continuum model notes (eqns3D.pdf) are hyperlinked PDF.Ed Bueler36+420-8142
2007-06-11 18:21Changes to accommodate petsc-2.3.3-p2. manual and Makefile updates for same purpose. Irritating that PETSc would make trivial changes to calling sequence or to names of procedures.Ed Bueler8+83-65
2007-06-09 21:57Many changes.Ed Bueler28+2742-914
2007-04-16 23:32Improvements to pant re. writing beta and tau, and computing tau_c.Ed Bueler3+39-23
2007-04-14 16:58Added regrid ability for Hmelt (-regrid_vars L). Summary has more digits for time step.Ed Bueler8+56-42
2007-04-12 07:15Fixed bug which caused Tb[i][j][Mbz-1]=0 (i.e. absolute zero). Note that it caused errors when *vertical* grid was regridded.Ed Bueler1+6-1
2007-04-12 03:22Fixed a bug with `-legacy_nc_Mbz'. It should work properly now.Jed Brown1+4-2
2007-04-12 02:27Bug #8710 is fixed. The communication of history was being done improperly, causing a hang.Jed Brown5+13-25
2007-04-10 06:51Test I works; note call to afterInitHook() and taucSet() must be in that order in IceExactStreamModel::initFromOptions(). get_drag.cc removed and SConscript and config/../Makefiles updated to built pant. plastic till version of pant seems to work.Ed Bueler9+35-423
2007-04-10 03:41Moved initialization of creating BasalType object in IceModel.cc. This fixes a memory bug that was introduced in r105.Jed Brown1+5-5
2007-04-10 01:47Bug #8859: test I works again.Jed Brown8+143-71
2007-04-09 20:28pant masking scheme adds control on betax/betay ratio. -super option implements 'glaciological superposition to low order', necessary for reasonable results under Schoof schemeEd Bueler7+111-116
2007-04-09 01:13fixing masking scheme in pant; scalar and vector beta methods functionalEd Bueler2+73-148
2007-04-08 19:39pant in better shape, but hard to tuneEd Bueler6+431-181
2007-04-05 10:09Start on pant.cc: computes and runs with nonconstant beta from balance velocities. Write NetCDF file containing beta, betax, betay. Note initBasalFields() called in afterInitHook(). Viewers for beta (-d B) and tau_c (-d C) added.Ed Bueler7+823-12
2007-04-03 23:51Added scons-local. If you don't have scons installed, unpack the tarball in config and run `config/scons.py' from the pism base.Jed Brown3+26-25
2007-04-03 18:21Bug #8742: The bedrock temperature now includes temperature at the ice/rock interface. At this point, it is always the same as the ice temperature at the bed, but need not be, particularly if the ice is floating. This fixes a very messy problem when regridding bedrock temperature with refinement in the vertical.Jed Brown27+317-206
2007-03-31 10:33Updates to Makefiles for ARSC machines. Small mods to manual.Ed Bueler4+171-181
2007-03-26 23:05Implemented signal handling. This resolves task #3985. To get a snapshot of a running pism executable without stopping it, send SIGUSR1 to *all* processes in the group. It will produce a file named `pism-YY.YYYY.nc' where `YY.YYYY' is the floating point representation of grid.p->year. Alternatively, you may just abort a run early by sending SIGTERM to all processes of a pism group. This will exit at the end of the current time step, presumably writing the model state as specified in the driver program. If you don't send a signal to every process in a group, PISM may hang.Jed Brown8+101-4
2007-03-25 22:03Updates to manual including how to do Ross validation and a discussion of the adaptive time-stepping scheme and its flags.Ed Bueler4+215-51
2007-03-25 18:54Bug #8351 fixed. dt_force issue in bug #8436 addressed. CFL for mass conservation upwinding in MacAyeal regions added; see task #4219. modifications to antNC.sh.Ed Bueler7+119-94
2007-03-16 18:19Posting test/antNC.shEd Bueler1+93-0
2007-03-15 00:21More twiddles. Schoof style regularization better documented and more controllable.Ed Bueler9+52-1329
2007-03-12 21:52More work on manual. It looks pretty good now. Minor twiddles with verbosity, viewers.Ed Bueler24+248-186
2007-03-12 02:17Further improvements to manual. Verbosity level fully implemented.Ed Bueler17+345-213
2007-03-11 16:50Various twiddles. doc/manual.tex updated but not ready for prime-time; now uses pdflatex. src/iMIO.cc write a bit more Matlab output. src/iceEISModel.cc allows user to set all named parameters for surface temp and accumulation in EISMINT II.Ed Bueler7+345-212
2007-03-08 02:48Added netCDF regridding. Now we can do things likeJed Brown12+800-369
2007-03-06 01:17History now includes number of processes. Fixed bug in reading zero bedrock state from netCDF files. This bug was causing temperature problems for `pismv -test G -if foo.nc' type runs.Jed Brown3+39-20
2007-03-03 22:34Removed excessive verbosity. Fixed bug in history string.Jed Brown1+8-5
2007-03-03 22:24We read and write netCDF files now. For bootstrapping, use `-bif' instead of `-if'.Jed Brown8+426-187
2007-03-02 04:19We can now write (almost) CF-1.0 compliant netCDF files; just include the character `n' in the `-of' option of any pism executable. It works for me in parallel using MPICH.Jed Brown42+1033-300
2007-02-16 21:42Modified regrid to use DAGlobalToNatural and cousins instead of the hacked together code implementing the same functionality. Fixed computation of dvoldt which was summing the *local* vector rather than the *global* vector. VecSum() should probably never be applied to a local vector in Pism.Jed Brown3+68-55
2007-02-16 01:42new Makefile which actually works on iceberg.arsc.edu (an AIX machine)Ed Bueler1+166-158
2007-02-14 05:05EISMINT II experiments G and H added. Note that the 'Experiment G' paragraph on page 230 of (Payne et al 2000) probably contains a typo, and should end with '... that of basal PRESSURE'.Ed Bueler4+59-25
2007-02-14 01:45petscfix.h included in grid.ccEd Bueler1+1-1
2007-02-14 01:04Minor fixes to HEINO constants. Note HEINO should be run *without* thermal bedrock.Ed Bueler1+3-1
2007-02-12 06:41Final fixes on Ross Ice Shelf validation. With \bar B = 2.22e8 we seem to do slightly better relative to the RIGGS data than any of the models in the (MacAyeal et al 1996) validation.Ed Bueler5+71-43
2007-02-11 20:41Comments added to IceROSSModel to clarify use. Tools added to test/ to use RIGGS data from MacAyeal.Ed Bueler8+418-15
2007-02-08 22:40Attempt to improve Makefiles.Ed Bueler3+10-6
2007-02-07 04:42test/mpiverifynow.sh runs successful; test/README.mpiverifynow addedEd Bueler1+162-0
2007-02-04 00:32test/verifynow.sh and test/mpiverifynow.sh revised to indicate -ksp_rtol, which needs to be higher for multiprocessors than for single; see PETSc FAQ. Waiting for mpiverifynow.sh output on marmaduke.gi.alaska.edu.Ed Bueler3+41-37
2007-02-03 22:23Stopping criteria for MacAyeal when using Schoof regularization fixed.Ed Bueler1+13-11
2007-02-02 02:43Updates to verify scripts.Ed Bueler2+45-11
2007-02-01 02:25Two scripts added to pism/test/: 'renames.sh' helps in renaming multiple files (useful for ISMPI-HEINO) and 'ant.sh' is a script to model the current state of the Antarctic ice sheet.Ed Bueler2+84-0
2007-01-28 00:00Small improvements to get_drag.cc: comments removed, Matlab output cleanerEd Bueler1+5-8
2007-01-27 17:37Slight fix of Matlab output in iMIO.cc.Ed Bueler1+2-3
2007-01-27 01:36Fixes to plotting from Matlab output (i.e. '-o foo -of m'). Debug messages in iMtemp.cc to try to understand bug#8351.Ed Bueler3+99-1
2006-12-19 22:21Small revisions. README file added to record sample results from test/verifynow.sh, including Schoof's test I.Ed Bueler5+95-4
2006-12-19 00:14Test I *is COOL*. It definitely works, and because it is essentially one-dimensional, one can go to a high level of refinement. See the new verifynow.sh. Therefore *Schoof's plastic base* definitely works. This means we can have real ice streams for real reasons.Ed Bueler9+159-43
2006-12-17 19:25Capability to differential surface inward added to MacAyeal; note assemble split into matrix part and RHS part. Fixes to Makefiles.Ed Bueler6+88-33
2006-12-16 00:25Initial implementation of Test I, the Schoof plastic till exact solution to the ice stream equations. Need -eo (exact only) evolution to look at solution, and need reportError(). Currently Dirichlet all around boundary; would like to use periodic conditions, which would require (e.g.) option 'gradOfSlopeInwards' in velocityMacAyeal() to compute RHS of MacAyeal equations.Ed Bueler6+153-108
2006-12-14 00:52Further work to make HEINO deliver as specified. (Preliminary indications that with -tempskip 50 and -Mz 201 there are no heinrich events, and there are some CFL violations; interesting...) Start on implementing exact ice stream solution from Schoof 2006 'A variational approach to ice streams'.Ed Bueler5+492-112
2006-12-13 09:13Bit more of HEINO code-writing: now allows '-time1 156789.07' kind of option, and writes 2D plan form stuff at those times. Not complete: MUST WRITE ONLY FROM RANK ZERO processor to get nodes in right order. OR: perhaps just try 'VecView'?Ed Bueler2+175-4
2006-12-13 00:23'-tempskip' mechanism now fixed. Default is -tempskip 0 so one temperature step per mass balance. Basically it is only wise to use positive tempskip if other mechanism is guaranteed to produce a pretty short temperature time step, like '-maxdt 2.0 -tempskip 5', etc.Ed Bueler10+51-35
2006-12-12 01:43Changes to velocitySIA() which are SIGNIFICANT. When there is not a 'updateAtDepth', e.g. for a temperature step, then OLD VALUES OF DIFFUSIVITY ARE USED IN MASS BALANCE. This makes some steps very fast in (e.g.) HEINO. New staggered grid Vecs vDf[2] added. Note that '-isoflux' is now broken; why?Ed Bueler5+168-129
2006-12-11 04:11Minor changes.Ed Bueler2+6-5
2006-12-11 03:56Changed MAX_TEMP_SKIP to 10.Ed Bueler1+1-1
2006-12-10 20:11Realized power of '-tempskip' mechanism can be exploited! HEINO should be a good bit faster; X2 or X3?Ed Bueler11+96-84
2006-12-09 18:40Ongoing updates of build system. Need a better way than to maintain a bunch of 'Makefile's.Ed Bueler2+17-13
2006-12-09 18:05Fixes to dumb stuff (e.g. C style comments).Ed Bueler2+3-3
2006-12-09 18:03Fixes to dumb stuff.Ed Bueler2+2-2
2006-12-09 17:58Update to some Makefiles; attempting to build on iceberg.Ed Bueler2+96-47
2006-12-09 00:32Added source sub-directory for exact tests.Ed Bueler13+8-8
2006-12-09 00:12Initial implementation of exact solution '-test I', namely ice stream with plastic bed from Schoof. Use for verification (i.e. additions to IceCompModel) not yet started.Ed Bueler6+191-5
2006-12-08 21:52A few more refinements of IceROSSModel. Now 'pisms -ross -tune 1.9e8,0.1e8,2.5e8' will work. Clarification of regularization as Schoof-form regularization in IceModel::computeEffectiveViscosity().Ed Bueler10+174-94
2006-12-08 00:13Further work on EISMINT ROSS. Continues to look good. See 'pisms -ross -tune'; should get a mean vertically-integrated hardness which works well for ice shelves around Antarctica.Ed Bueler9+219-152
2006-12-07 05:33Further work on EISIMINT Ross. Looks good. Need to figure out RIGGS.dat in order to compute chi^2. Note: low mv_rtol produces smaller peak velocity. Reasonable to consider two parameters for viscosity (barB and eps in computing effective strain rate) and ask for best fit.Ed Bueler5+280-110
2006-12-05 23:55Ross mostly done! Need to compute X^2 statistic, and max mag a little high (1576), but in good shape.Ed Bueler10+250-69
2006-12-05 05:18Some serious work on IceRossModel. Will run as pisms -ross. Reads primary data file successfully; try obj/pisms -ross -d muvHbra -pause 20 -verbose. Bit of work toward actual verbosity level; see head of iMutil.cc.Ed Bueler12+528-16
2006-11-29 19:26Correct calculation of Sigma in stream/shelf now in iMmacayeal.cc. effectiveViscosity() versions added to flow laws in materials.cc. Decision to re-include basal melt rate in mass continuity and vertical velocity for HEINO, following fix of bug re geo heat flux at base of bedrock. Minor change to documentation.Ed Bueler8+134-83
2006-11-26 21:04Further work on temperature in column: (i) doc updated with temp in ice shelf and refines to melting; (ii) similar for iMtemp.cc; (iii) found geothermal flux bug in iMtemp.cc. Fixed; all HEINO runs must redo.Ed Bueler10+4174-211
2006-11-25 22:41Small mods to Makefile and to HEINO.Ed Bueler4+53-30
2006-11-24 19:13Refinements of basal melt mechanism.Ed Bueler3+126-92
2006-11-22 19:19Structure of EISMINT II and ISMIP-HEINO derived classes fixed. Is 'multiple inheritance' technique in simplify.cc most efficient?Ed Bueler10+1044-728
2006-11-20 04:57Update of test/verifynow.sh.Ed Bueler3+13-8
2006-11-19 19:32Work done on temperatureStep(). Advection and strain heating only included in conservation of energy if neighbors are not too thin. Limitation on contribution of basal melt to mass continuity and to vertical velocity. Note new flag 'includeBMRinContinuity'.Ed Bueler9+177-102
2006-11-17 19:00Major revisions to iMtemp.cc. Substantial revisions to iMvelocity.cc. In both cases moved communication to top-level function (temperatureAgeStep() and velocity()). Separated temperature and age time steps. *In temperatureStep(), redid conservation of energy in a column of ice based on recently-added section 5 of doc/eqns3D.tex.* Note addition of basal melt water thickness vHmelt used in conservation of energy for ice/rock interface segment. Note basal temp b.c. for ice shelves now based on a +2 deg C ocean temperature. Testing underway.Ed Bueler15+509-321
2006-11-14 18:07moved temp and age communication to temperatureAgeStep()Ed Bueler2+20-18
2006-11-14 17:31separation of temperatureStep() and ageStep()Ed Bueler4+142-94
2006-11-13 20:36made old corrections to 3D notesEd Bueler1+14-10
2006-11-13 20:13small changes to 3D notes; IGS bib style file needed and addedEd Bueler2+1335-35
2006-11-12 19:59Also changes to bibliography for documentation ...Ed Bueler1+99-56
2006-11-12 19:55Significant modifications to model notes (doc/eqns3D.tex). A long section on temperature in a column has been added. These notes and the user's manual still need lots of revision to match current code. The temperature routine in the code needs to be updated to match the new section in the notes.Ed Bueler9+10142-186
2006-11-06 21:25minor changes to scripts (for verify and ISMIP-HEINO) and to tools for Cray XD1 (i.e. Nelchina)Ed Bueler4+83-90
2006-11-02 21:48Fixed get_drag.cc to not use C++ calls to netCDF. The get_drag executable is not built by default.Jed Brown5+38-28
2006-11-01 00:04minor massaging of temperatureStep(); need finite-volume-thinking-through of temperature in column problem, including basal frictional heating issuesEd Bueler5+93-71
2006-10-31 00:06added test/showheino.m to show time series from ISMIP-HEINOEd Bueler2+114-3
2006-10-30 20:56further work on ISMIP-HEINO; esp timesteppingEd Bueler5+30-21
2006-10-30 05:25updated time stepping in IceCompModelEd Bueler5+31-23
2006-10-29 20:21revised time-stepping with better reporting and allowing derived classes to override; note IceCompModel::run() needs corresponding updateEd Bueler4+112-39
2006-10-29 01:55raised ceiling to 6km for ISMIP-HEINOEd Bueler2+1-63
2006-10-27 23:25added missing file simpleISO.cEd Bueler1+110-0
2006-10-27 23:18ISMIP-HEINO time series deliverables done; need to writeu 2D plan form deliverables and document HEINO in manualEd Bueler1+63-11
2006-10-27 19:36further progress on ISMIP-HEINO including much of deliverables; move regrid stuff into iMregrid.ccEd Bueler7+644-346
2006-10-26 18:58improvements to diffusivity reporting and adaptive time stepping; changes to appearance of error report in PISMV; changes to basic run appearance of all PISM executablesEd Bueler11+129-122
2006-10-25 23:37reportErrors() in IceCompModel improved, including for test E; storage and viewing of basal velocities partly implemented; other small changesEd Bueler14+372-209
2006-10-25 01:51tests A and E nearly completed; need report_error for EEd Bueler6+191-94
2006-10-24 23:50addition of tests A and E from 2005 J. Glaciol. paper; put test H into separate file; work on ISMIP-HEINO; note pismv does not yet do A or EEd Bueler14+510-366
2006-10-24 04:28ISMIP-HEINO initial implementation: only standard run, no reportingEd Bueler5+74-28
2006-10-23 21:48start on ISMIP-HEINO; note basal() now allows dependence on horizontal location; more massage of Makefile (and other small changes ...)Ed Bueler9+130-71
2006-10-23 17:59revised Makefile to put executables in obj, named as in SCons buildEd Bueler1+57-32
2006-10-19 19:50Fixed another bug in adaptive time stepping, in IceModel::computeMaxVelocities(). Now both CFL and diffusivity time steps *should* be correct on multi-processor runs.Ed Bueler1+1-1
2006-10-19 17:51Removed remaining refs to c++ version of NetCDF. Add build tools for Cray XD1 (nelchina).Ed Bueler5+164-3
2006-10-15 22:43moved exact tests to C (from C++) for compatibility; updates to MakefileEd Bueler14+688-516
2006-10-03 23:14Executables are pismr, pisms, pismv and are linked to libpism.soJed Brown4+12-12
2006-10-03 22:51Removed C++ dependency for netcdf.Jed Brown6+178-141
2006-10-03 20:42fix of scons for bueler; remove debug print in runEd Bueler2+3-2
2006-10-03 19:17title in manual fixedEd Bueler1+2-2
2006-09-29 22:05Initial import.Jed Brown68+178376-0
2006-09-28 06:58 "default layout"Jed Brown0+0-0