tchanges-05-06.rst - pism - [fork] customized build of PISM, the parallel ice sheet model (tillflux branch) (HTM) git clone git://src.adamsgaard.dk/pism (DIR) Log (DIR) Files (DIR) Refs (DIR) LICENSE --- tchanges-05-06.rst (3942B) --- 1 Changes from 0.5 (May 2012) to 0.6 2 ================================== 3 4 Basal strength and basal hydrology 5 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 6 7 - A significant change to the model physics for till: effective 8 pressure in till is now computed by empirically-based (and 9 exponential) formula from [``Tulaczyketal2000``\ ] relating amount in 10 till to effective pressure. 11 - New mass-conserving subglacial hydrology model. 12 13 Marine ice sheet modeling 14 ~~~~~~~~~~~~~~~~~~~~~~~~~ 15 16 - Implement a parameterization of the melange back pressure as a part 17 of the stress boundary condition at the ice shelf front. 18 - Implement fracture density model and fracture-induced ice softening 19 (see [``AlbrechtLevermann2012``\ ]). 20 - Parameterization of the sub-grid position of the grounding line and 21 related improvements (see 22 [``Feldmannetal2014``,\ ``Gladstoneetal2012``]). 23 - MISMIP3D example (see [``MISMIP3d2013``\ ]), grounding line 24 reversibility. 25 - Fixes related to the handling of the grounding line (i.e. driving 26 stress, etc) 27 - Use an implementation of a serial two-scan connected-component 28 algorithm to identify icebergs. 29 - Report cumulative discharge (calving) flux as a 2D field. 30 31 Climate inputs and ocean inputs 32 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 33 34 - ``PISMSurfaceModel`` returns climatic mass balance in 35 ``[kg m-2 s-1]`` 36 - Remove ``pclimate``; add the ``-test_climate_models`` option. 37 - ``-part_grid`` extended to grounded marine termini. 38 - Require time bounds in time-dependent climate forcing. 39 - Improve the PDD scheme. (Keeps track of the snow depth during the 40 year to choose PDD factors but does not model multi-year snow cover.) 41 - Implement the 3-equation sub-shelf melt parameterization (see 42 [``Hellmer1998``\ ]). 43 - Implement caching surface and ocean models ``-ocean ...,cache`` and 44 ``-surface ...,cache``. (Update boundary inputs every X years but 45 include interannual variability.) 46 - Remove ``EISMINT-Greenland``. 47 48 Inverse modeling 49 ~~~~~~~~~~~~~~~~ 50 51 - Inverse modeling tools *are* a part of the release now. 52 53 Energy 54 ~~~~~~ 55 56 - New bootstrapping heuristic filling ice temperature at depth from 57 surface mass balance (if available). 58 - Allow "regridding" enthalpy from files containing ice temperature and 59 liquid water fraction. 60 - Allow cold flow laws in the enthalpy mode and GPBLD in the cold mode 61 (same as Paterson-Budd). 62 - Corrected basal boundary condition in the enthalpy system. 63 64 Usability 65 ~~~~~~~~~ 66 67 - Implement poor man's parallel I/O, with compression 68 - Ensure "continuity" in time of reported cumulative diagnostic fields. 69 - Let the user precisely specify the dates corresponding to the run 70 (``-time_file``). 71 - Many more diagnostic quantities; use ``-list_diagnostics`` to see. 72 - PISM keeps track of options and parameters that were set but were not 73 used. 74 - Use projection info to compute latitude and longitude bounds. 75 (Reduces post-processing needed to work with PISM's output.) 76 77 Miscellaneous 78 ~~~~~~~~~~~~~ 79 80 - Improve mass conservation and mass transport. 81 - New validation example based on laboratory experiment with xanthan 82 gum. Uses millimeter grid spacing and shows configurability of PISM. 83 - Implement a "constant 2D velocity" stress balance object (for testing 84 and prescribing sliding) 85 86 Under the hood 87 ~~~~~~~~~~~~~~ 88 89 - Improve the build system. 90 - Switch to PETSC 3.3 or 3.4, stop supporting 3.2. 91 - Switch to 92 `UDUNITS-2 <http://www.unidata.ucar.edu/software/udunits/>`__. 93 - Require `FFTW-3 <http://www.fftw.org>`__ to build PISM. 94 - Use `CalCalcs <http://meteora.ucsd.edu/~pierce/calcalcs/>`__ for 95 proper calendar support. 96 - Remove the Debian [meta-] package. 97 - Clean up command-line options selecting sub-models (calving, stress 98 balance, energy, basal yield stress). 99 - Better ``SSAFD`` recovery logic; save model state on failure. 100 - Use non-zero initial guess in the SSAFD KSP solver. 101 - Improved basal yield stress code. 102