[Dart-dev] [6397] DART/trunk/doc/html: updated the diffs document, and filled in the ' non-backwards
nancy at ucar.edu
nancy at ucar.edu
Mon Aug 12 16:55:09 MDT 2013
Revision: 6397
Author: nancy
Date: 2013-08-12 16:55:09 -0600 (Mon, 12 Aug 2013)
Log Message:
-----------
updated the diffs document, and filled in the 'non-backwards
compatible changes' section of the release notes. still working
on the 'new' parts of the release notes.
Modified Paths:
--------------
DART/trunk/doc/html/Lanai_diffs_from_Kodiak.html
DART/trunk/doc/html/Lanai_release.html
-------------- next part --------------
Modified: DART/trunk/doc/html/Lanai_diffs_from_Kodiak.html
===================================================================
--- DART/trunk/doc/html/Lanai_diffs_from_Kodiak.html 2013-08-12 17:51:26 UTC (rev 6396)
+++ DART/trunk/doc/html/Lanai_diffs_from_Kodiak.html 2013-08-12 22:55:09 UTC (rev 6397)
@@ -36,7 +36,23 @@
<a href="#Legalese">Terms of Use</a>
<!--==================================================================-->
+<A NAME="Overview"></A>
+<H2>Overview</H2>
+<P>
+This document includes an overview of the changes in the DART
+system since the Kodiak release. For further details on any of
+these items look at the HTML documentation for that specific part
+of the system.
+</P>
+<P>
+There is a longer companion document for this release, the
+<a href="Lanai_release.html">Lanai Release Notes</a>, which
+include installation instructions, a walk-through of running
+one of the low-order models, the diagnostics, and a description
+of non-backward compatible changes.
+</P>
+<!--==================================================================-->
<br />
<br />
<A NAME="NewModels"></A>
@@ -49,34 +65,73 @@
</P>
<UL>
+<LI>Support for components under the CESM framework:
+<UL>
+<LI>Added support for the Community Land Model (CLM).</LI>
+<LI>Added support to run the Community Atmospheric Model (CAM) under the
+CESM framework.</LI>
+<LI>Added support for the CESM 1.1.1 release for CAM, POP, CLM: includes
+experiment setup scripts and assimilation scripts.</LI>
+<LI>CAM, POP, and/or CLM can be assimilated either individually or
+in combination while running under the CESM framework.
+If assimilating into multiple components,
+they are assimilated sequentially with observations only affecting
+a single component directly. Other components are indirectly affected
+through interactions with the coupler.</LI>
+<LI>Setup scripts are provided to configure a CESM experiment using
+the multi-instance feature of CESM to support ensembles for assimilation.</LI>
+<LI>POP state vector contains potential temperature; observations from the
+World Ocean Database are in-situ or sensible temperature. The model_mod
+now corrects for this.
+<UL>
+<LI>The state vector has all along contained potential temperature and not
+in-situ (sensible) temperature. The observations from the World Ocean
+Database are of sensible temperature. Changed the specific kind in the
+model_mod to be <tt>KIND_POTENTIAL_TEMPERATURE</tt> and added new code to
+convert from potential to in-situ temperature. Differences for even the
+deeper obs (4-5km) is still small ( ~ 0.2 degree). (in-situ or sensible
+temperature is what you measure with a regular thermometer.)</LI>
+</UL>
+</LI>
+<LI>Support for the SE core (HOMME) of CAM has been developed but
+<strong>is not</strong> part of the current release. Contact the
+DART group if you have an interest in running this configuration of CAM.</LI>
+</UL>
+</LI>
<LI>Changes to the WRF model_mod:
<UL>
-<LI>allow advanced microphysics schemes (needed interpolation for 7 new kinds)</LI>
-<LI>option to interpolate in the vertical in log(p) instead of linear in p (log(p) is the default now)</LI>
-<LI>added support in the namelist to avoid writing updated fields back into the wrf netcdf files. The fields are still updated during the assimilation but the updated data is not written back to the wrfinput file during the dart_to_wrf step.</LI>
-<LI>fixed an obscure bug in the vertical convert routine of the wrf model_mod that would occasionally fail to convert an obs. This would make tiny differences in the output as the number of mpi tasks change. No quantitative differences in the output but not bitwise compatible before and is again now.</LI>
+<LI>Allow advanced microphysics schemes (needed interpolation for 7 new kinds)</LI>
+<LI>Interpolation in the vertical is done in log(p) instead of linear
+pressure space. log(p) is the default, but a compile-time variable can
+restore the linear interpolation.</LI>
+<LI>Added support in the namelist to avoid writing updated fields back into
+the wrf netcdf files. The fields are still updated during the assimilation
+but the updated data is not written back to the wrfinput file during the
+dart_to_wrf step.</LI>
+<LI>Fixed an obscure bug in the vertical convert routine of the wrf model_mod
+that would occasionally fail to convert an obs. This would make tiny
+differences in the output as the number of mpi tasks change. No quantitative
+differences in the results but they were not bitwise compatible before and they
+are again now.</LI>
</UL>
</LI>
<LI>Added support for the MPAS_ATM and MPAS_OCN models.
<UL>
-<LI>added code to the mpas_atm model to interpolate specific humidity and pressure, so we can assimilate GPS obs now.</LI>
+<LI>Added interpolation routines for the voroni-tesselation grid (roughly hexagonal)
+<LI>Includes vertical conversion routines for vertical localization.
+<LI>Added code to the mpas_atm model to interpolate specific humidity and
+pressure, so we can assimilate GPS obs now.</LI>
</UL>
</LI>
-<LI>Added support for the CLM land model.</LI>
-<LI>Added support for the CESM 1.1.1 release for CAM, POP, CLM: includes experiment setup scripts and assimilation scripts.</LI>
<LI>Added support for the 'SQG' uniform PV two-surface QC+1 spectral model.</LI>
<LI>Added support for a flux-transport solar dynamo model.</LI>
<LI>Added support for the NOAH land model.</LI>
<LI>Added support for the NAAPS model.</LI>
<LI>Added model_mod interface code for the NOGAPS model to the SVN repository.</LI>
-<LI>Fix temperature vs potential temp bug in POP model_mod.
-<UL>
-<LI>state vector has all along contained potential temperature and not in-situ (sensible) temperature. The observations from the World Ocean Database are of sensible temperature. Changed the specific kind in the model_mod to be <pre>KIND_POTENTIAL_TEMPERATURE</pre> and added new code to convert from potential to in-situ temperature. Differences for even the deeper obs (4-5km) is still small ( ~ 0.2 degree). (in-situ or sensible temperature is what you measure with a regular thermometer.)</LI>
-</UL>
-</LI>
<LI>Simple advection model:
<UL>
-<LI>Fix where the random number seed is set in the models/simple_advection model_mod - it needed to be sooner than it was being called.</LI>
+<LI>Fix where the random number seed is set in the models/simple_advection
+model_mod - it needed to be sooner than it was being called.</LI>
</UL>
</LI>
</UL>
@@ -92,9 +147,18 @@
</P>
<UL>
-<LI>Added a completely new random number generator based on the Mersenne Twister algorithm from the GNU scientific library. It seems to have better behavior if reseeded frequently, which is a possible usage pattern if perfect_model_obs is run for only single steps and the model is advanced in an external script. As part of this code update all random number code was moved into the random_seq_mod and random_nr_mod is deprecated.</LI>
-<LI>perfect_model_obs calls a seed routine in the time manager now that generates a consistent seed based on the current time of the state. This makes subsequent runs give consistent results and yet separate runs don't get identical error values.</LI>
-<LI>Added random number generator seeds in several routines to try to get consistent results no matter how many MPI tasks the code was run with. This includes:
+<LI>Added a completely new random number generator based on the Mersenne
+Twister algorithm from the GNU scientific library. It seems to have better
+behavior if reseeded frequently, which is a possible usage pattern if
+perfect_model_obs is run for only single steps and the model is advanced in
+an external script. As part of this code update all random number code was
+moved into the random_seq_mod and random_nr_mod is deprecated.</LI>
+<LI>Perfect_model_obs calls a seed routine in the time manager now that
+generates a consistent seed based on the current time of the state. This
+makes subsequent runs give consistent results and yet separate runs don't get
+identical error values.</LI>
+<LI>Added random number generator seeds in several routines to try to get
+consistent results no matter how many MPI tasks the code was run with. This includes:
<UL>
<LI>cam model_mod.f90, pert_model_state()</LI>
<LI>assim_tools_mod.f90, filter_assim(), filter kinds 2, 3, and 5</LI>
@@ -102,14 +166,39 @@
<LI>adaptive_inflate_mod.f90, adaptive_inflate_init(), non-deterministic inf</LI>
</UL>
</LI>
+<LI>There is a new &filter_nml namelist item: enable_special_outlier_code.
+If .true. the DART quality control code will call a separate subroutine at
+the end of filter.f90 to evaluate the outlier threshold. The user can add
+code to that routine to change the threshold based on observation type or
+values as they wish. If .false. the default filter outlier threshold code
+will be called and the user routine ignored.</LI>
+<LI>Filter will call the end_model() subroutine in the model_mod for
+the first time. It should have been called all along, but was not.</LI>
<LI>Added a time sort routine in the time_manager_mod.</LI>
-<LI>Avoid a pair of all-to-all transposes when setting the inflation mean and sd from the namelist. The new code finds the task which has the two copies and sets them directly without a transpose. The log messages were also moved to the end of the routine - if you read in the mean/sd values from a restart file the log messages that printed out the min/max values needed to be after the read from the file.</LI>
-<LI>Reordered the send/receive loops in the all-to-all transposes to scale better on yellowstone.</LI>
-<LI>Remove a state-vector size array from the stack in read_ensemble_restart(). The array is allocated only if needed and then deallocated. The ensemble write routine was changed before the Kodiak release but the same code in read was apparently not changed simply as an oversight.</LI>
-<LI>If the ensemble mean is selected to be written out in dart restart file format, the date might not have been updated correctly. The code was fixed to ensure the ensemble mean date in the file was correct.</LI>
+<LI>Avoid a pair of all-to-all transposes when setting the inflation mean and
+sd from the namelist. The new code finds the task which has the two copies
+and sets them directly without a transpose. The log messages were also moved
+to the end of the routine - if you read in the mean/sd values from a restart
+file the log messages that printed out the min/max values needed to be after
+the read from the file.</LI>
+<LI>Reordered the send/receive loops in the all-to-all transposes to scale
+better on yellowstone.</LI>
+<LI>Remove a state-vector size array from the stack in
+read_ensemble_restart(). The array is now allocated only if needed and then
+deallocated. The ensemble write routine was changed before the Kodiak release
+but the same code in read was apparently not changed simply as an
+oversight.</LI>
+<LI>If the ensemble mean is selected to be written out in dart restart file
+format, the date might not have been updated correctly. The code was fixed to
+ensure the ensemble mean date in the file was correct.</LI>
<LI>filter writes the ensemble size into the log file.</LI>
-<LI>Reorganized the code in the section of obs_model_mod that prints out the time windows, with and without verbose details. Should be clearer if the next observation is in or out of the current assimilation window, and if the model needs to advance or not.</LI>
-<LI>Added a fill_inflation_restart utility which can write a file with a fixed mean and sd, so the first step of a long assimilation run can use the same 'start_from_restart_file' as subsequent steps.</LI>
+<LI>Reorganized the code in the section of obs_model_mod that prints out the
+time windows, with and without verbose details. Should be clearer if the next
+observation is in or out of the current assimilation window, and if the model
+needs to advance or not.</LI>
+<LI>Added a fill_inflation_restart utility which can write a file with a
+fixed mean and sd, so the first step of a long assimilation run can use the
+same 'start_from_restart_file' as subsequent steps.</LI>
<LI>Added new location module options:
<UL>
<LI>Channel coordinate system</LI>
@@ -180,10 +269,20 @@
<LI>plus slots 151-250 reserved for Chemistry (specifically WRF-Chem) kinds</LI>
</UL>
</LI>
-<LI>Added a forward operator for total precipitable water. It loops over model levels so it can be used as an example of how to handle this without having to hardcode the number of levels into the operator.</LI>
-<LI>Added a forward operator (and obs_seq file converter) for COSMOS ground moisture observations.</LI>
-<LI>Added a forward operator (and obs_seq file converter) for MIDAS observations of Total Electron Count.</LI>
-<LI>Added a 'set_1d_integral()' routine to the obs_def_1d_state_mod.f90 forward operator for the low order models. This subroutine isn't used by filter but it would be needed if someone wanted to write a standalone program to generate obs of this type. We use this file as an example of how to write an obs type that has metadata, but we need to give an example of how to set the metadata if you aren't using create_obs_sequence interactively (e.g. your data is in netcdf and you have a separate converter program.)</LI>
+<LI>Added a forward operator for total precipitable water. It loops over
+model levels so it can be used as an example of how to handle this without
+having to hardcode the number of levels into the operator.</LI>
+<LI>Added a forward operator (and obs_seq file converter) for COSMOS ground
+moisture observations.</LI>
+<LI>Added a forward operator (and obs_seq file converter) for MIDAS
+observations of Total Electron Count.</LI>
+<LI>Added a 'set_1d_integral()' routine to the obs_def_1d_state_mod.f90
+forward operator for the low order models. This subroutine isn't used by
+filter but it would be needed if someone wanted to write a standalone program
+to generate obs of this type. We use this file as an example of how to write
+an obs type that has metadata, but we need to give an example of how to set
+the metadata if you aren't using create_obs_sequence interactively (e.g. your
+data is in netcdf and you have a separate converter program.)</LI>
</UL>
@@ -201,19 +300,35 @@
<UL>
<LI>Added an obs_sequence converter for wind profiler data from MADIS.</LI>
-<LI>Added an obs_sequence converter for Ameriflux land observations(latent heat flux, sensible heat flux, net ecosystem production).</LI>
+<LI>Added an obs_sequence converter for Ameriflux land observations(latent
+heat flux, sensible heat flux, net ecosystem production).</LI>
<LI>Added an obs_sequence converter for MODIS snow coverage measurements.</LI>
<LI>Added an obs_sequence converter for COSMOS ground moisture observations.</LI>
<LI>Added an obs_sequence converter for MIDAS observations of Total Electron Count.</LI>
-<LI>Updated scripts for the GPS converter; added options to convert data from multiple satellites.</LI>
-<LI>More scripting support in the MADIS obs converters; more error checks added to the rawin converter.</LI>
-<LI>Added processing for wind profiler observation to the wrf_dart_obs_preprocess program.</LI>
-<LI>Fix BUG in airs converter - the humidity obs are accumulated acrossthe layers and so the best location for them is the layer midpoint and not on the edges (levels) as the temperature obs are. Also fixed off-by-one error where the converter would make one more obs above the requested top level.</LI>
-<LI>Made gts_to_dart converter create separate obs types for surface dewpoint vs obs aloft because they have different vertical coordinates.</LI>
-<LI>Converted mss commands to hpss commands for a couple observation converter shell scripts (inc AIRS).</LI>
-<LI>New matlab code to generate evenly spaced observations on the surface of a sphere (e.g. the globe).</LI>
-<LI>Added obs_loop.f90 example file in obs_sequence directory; example template for how to construct special purpose obs_sequence tools.</LI>
-<LI>Change the default in the script for the prepbufr converter so it will swap bytes, since all machines except ibms will need this now.</LI>
+<LI>Updated scripts for the GPS converter; added options to convert data from
+multiple satellites.</LI>
+<LI>More scripting support in the MADIS obs converters; more error checks
+added to the rawin converter.</LI>
+<LI>Added processing for wind profiler observation to the
+wrf_dart_obs_preprocess program.</LI>
+<LI>Fix BUG in airs converter - the humidity obs are accumulated across the
+layers and so the best location for them is the layer midpoint and not on the
+edges (levels) as the temperature obs are. Also fixed off-by-one error where
+the converter would make one more obs above the requested top level.</LI>
+<LI>Made gts_to_dart converter create separate obs types for surface dewpoint
+vs obs aloft because they have different vertical coordinates.</LI>
+<LI>Converted mss commands to hpss commands for a couple observation
+converter shell scripts (inc AIRS).</LI>
+<LI>New matlab code to generate evenly spaced observations on the surface of
+a sphere (e.g. the globe).</LI>
+<LI>Added obs_loop.f90 example file in obs_sequence directory; example
+template for how to construct special purpose obs_sequence tools.</LI>
+<LI>Change the default in the script for the prepbufr converter so it will
+swap bytes, since all machines except ibms will need this now.</LI>
+<LI>The 'wrf_dart_obs_preprocess' program now refuses to superob observations
+that include the pole, since the simple averaging of latitude and longitude
+that works everyplace else won't work there. Also treats observations near
+the prime meridian more correctly.</LI>
</UL>
@@ -229,19 +344,39 @@
<UL>
<LI>Handle empty epochs in the obs_seq_to_netcdf converter.</LI>
-<LI>Added a matlab utility to show the output of a 'hop' test (running a model for a continuous period vs. stopping and restarting a run).</LI>
-<LI>The obs_common_subset program can select common observations from up to 4 observation sequence files at a time.</LI>
-<LI>Add code in obs_seq_verify to ensure that the ensemble members are in the same order in all netcdf files.</LI>
+<LI>Added a matlab utility to show the output of a 'hop' test (running a
+model for a continuous period vs. stopping and restarting a run).</LI>
+<LI>Improved the routine that computes axes tick values in plots with
+multiple values plotted on the same plot.</LI>
+<LI>The obs_common_subset program can select common observations from up to 4
+observation sequence files at a time.</LI>
+<LI>Add code in obs_seq_verify to ensure that the ensemble members are in the
+same order in all netcdf files.</LI>
<LI>Added support for the unstructured grids of mpas to our matlab diagnostics.</LI>
<LI>Fix to writing of ReportTime in obs_seq_coverage.</LI>
<LI>Fixed logic in obs_seq_verify when determining the forecast lat.</LI>
-<LI>Fixed loops inside obs_seq_coverage which were using the wrong limits on the loops. Fixed writing of 'ntimes' in output netcdf variable.</LI>
-<LI>Rewrote the algorithm in the obs_selection tool so it had better scaling with large numbers of obs.</LI>
-<LI>Added preliminary support for a list of 'trusted obs' in the obs_diag program. Can disable the rank histogram generation with a namelist item. can define height_edges or heights in the namelist, but not both.</LI>
-<LI>Extend obs_seq_verify so it can be used for forecasts from a single member. minor changes to obs_selection, obs_seq_coverage and obs_seq_verify to support a single member.</LI>
-<LI>Added Matlab script to read/print timestamps from binary dart restart/ic files.</LI>
-<LI>Default for obs_seq_to_netcdf in all the namelists is now 'one big time bin' so you don't have to know the exact timespan of an obs_seq.final file before converting to netCDF.</LI>
+<LI>Fixed loops inside obs_seq_coverage which were using the wrong limits on
+the loops. Fixed writing of 'ntimes' in output netcdf variable.</LI>
+<LI>Rewrote the algorithm in the obs_selection tool so it had better scaling
+with large numbers of obs.</LI>
+<LI>Several improvements to the 'obs_diag' program:
+<UL>
+<LI>Added preliminary support for a list of 'trusted obs' in the obs_diag
+program.</LI>
+<LI>Can disable the rank histogram generation with a namelist item.</LI>
+<LI>Can define height_edges or heights in the namelist, but not both.</LI>
+<LI>The 'rat_cri' namelist item (critical ratio) has been deprecated.</LI>
</UL>
+</LI>
+<LI>Extend obs_seq_verify so it can be used for forecasts from a single
+member. minor changes to obs_selection, obs_seq_coverage and obs_seq_verify
+to support a single member.</LI>
+<LI>Added Matlab script to read/print timestamps from binary dart restart/ic
+files.</LI>
+<LI>Default for obs_seq_to_netcdf in all the namelists is now 'one big time
+bin' so you don't have to know the exact timespan of an obs_seq.final file
+before converting to netCDF.</LI>
+</UL>
<!--==================================================================-->
@@ -258,15 +393,32 @@
<UL>
-<LI>The mkmf-generated Makefiles now take care of calling 'fixsystem' if needed so the mpi utilities code compiles without further user intervention.</LI>
-<LI>Make the default input.nml for the Lorenz 96 and Lorenz 63 model gives good assimilation results. Rename the original input.nml to input.workshop.nml. The workshop_setup script renames it back before doing anything else so this won't break the workshop instructions. Simplify all the workshop_setup.csh scripts to do the minimal work needed by the DART tutorial.</LI>
-<LI>Updates to the models/template directory with the start of a full 3d geophysical model template. Still under construction.</LI>
-<LI>Move the pdf files in the tutorial directory up a level. routines that generate diagrams/figures are moved to the DAReS/group/DART_tutorial dir. framemaker files removed because we don't have a working version to update or read them with.</LI>
-<LI>Enable netCDF large file support in the work/input.nml for models which are likely to have large state vectors.</LI>
-<LI>Minor updates to the doc.css file, make pages look identical in the safari and firefox browsers.</LI>
-<LI>Added a utility that sorts and reformats namelists, culls all comments to the bottom of the file. Useful for doing diffs and finding duplicated namelists in a file.</LI>
-<LI>Cleaned up mkmf files - removed files for obsolete platforms and compilers, updated suggested default flags for intel.</LI>
-<LI>Update the mkmf template for gfortran to allow fortran source lines longer than 132 characters.</LI>
+<LI>The mkmf-generated Makefiles now take care of calling 'fixsystem' if
+needed so the mpi utilities code compiles without further user
+intervention.</LI>
+<LI>Make the default input.nml for the Lorenz 96 and Lorenz 63 model gives
+good assimilation results. Rename the original input.nml to
+input.workshop.nml. The workshop_setup script renames it back before doing
+anything else so this won't break the workshop instructions. Simplify all the
+workshop_setup.csh scripts to do the minimal work needed by the DART
+tutorial.</LI>
+<LI>Updates to the models/template directory with the start of a full 3d
+geophysical model template. Still under construction.</LI>
+<LI>Move the pdf files in the tutorial directory up a level. Removed
+framemaker source files because we no longer have access to a working
+version of the Framemaker software. Moved routines that generate figures
+and diagrams to a non-distributed directory of the subversion repository.</LI>
+<LI>Enable netCDF large file support in the work/input.nml for models which
+are likely to have large state vectors.</LI>
+<LI>Minor updates to the doc.css file, make pages look identical in the
+safari and firefox browsers.</LI>
+<LI>Added a utility that sorts and reformats namelists, culls all comments to
+the bottom of the file. Useful for doing diffs and finding duplicated
+namelists in a file.</LI>
+<LI>Cleaned up mkmf files - removed files for obsolete platforms and
+compilers, updated suggested default flags for intel.</LI>
+<LI>Update the mkmf template for gfortran to allow fortran source lines
+longer than 132 characters.</LI>
</UL>
Modified: DART/trunk/doc/html/Lanai_release.html
===================================================================
--- DART/trunk/doc/html/Lanai_release.html 2013-08-12 17:51:26 UTC (rev 6396)
+++ DART/trunk/doc/html/Lanai_release.html 2013-08-12 22:55:09 UTC (rev 6397)
@@ -69,7 +69,7 @@
Comparing these estimates and their uncertainty to the observation and
its uncertainty ultimately results in the adjustments to the model states.
See the DARTLAB demos or read more in the tutorials included with the
-DART distribution. They are described below.<p>
+DART distribution. They are described below.</p>
<p>
DART diagnostic output includes two netCDF files containing
@@ -1964,6 +1964,92 @@
<ol>
+<li>The DART system uses a new random number generator based on
+the Mersenns Twister algorithm from the GNU scientific library.
+It is believed to have better behavior in general, and in particular
+when it is frequently reseeded, as may be the case in some
+perfect_model_obs experiments. The seed in perfect_model_obs
+is now based on the time-stamp associated with the data, so running
+single advances as separate invocations of the executable will still
+result in a good random distribution of the observation errors.
+The seeds in several other places in the code have been changed so
+they are more consistent in the face of different numbers of MPI
+tasks when executing. The random values should reproduce if an
+identical run is repeated, but there are still a few places in the
+code where changing the number of MPI tasks results in different
+seeds being created for the random number generator, and so the
+non-deterministic values will differ.</li>
+
+<li>The WRF model_mod now interpolates in the vertical in log(pressure)
+space instead of linear pressure space. This is the new default.
+There is a module global variable that can be set at compile time
+to restore the previous behavior.</li>
+
+<li>The POP model_mod used to interpolate sensible temperature observations
+using a potential temperature field in the state vector. The code now
+correctly does the conversion from potential temperature to sensible (in-situ)
+temperature during the forward operator process.</li>
+
+<li>The simple_advection model was incorrectly calling the random number
+generator initialization routines after generating some random numbers.
+It now correctly initializes the generator before getting any random values.</li>
+
+<li>The gts_to_dart converter now creates separate obs types for surface dewpoint
+vs obs aloft because they have different vertical coordinates. The obs_diag
+program (and other diagnostic routines) do not cope with the same obs type
+having different vertical coordinates because it is trying to bin observations
+in the vertical (it is unable to convert pressure to height after the fact,
+for example, or bin surface obs with a height with pressure obs).</li>
+
+<li>Shell scripts which used to contain MSS (mass store) commands for
+long-term archiving have been converted to call HSI (HPSS) commands.</li>
+
+<li>The 'wrf_dart_obs_preprocess' program will now refuse to superob observations
+which are too close to the poles. If the superob radius includes either pole,
+the computation of an average obs location becomes more complicated than the
+existing code is prepared to deal with. (If this case is of interest to you,
+write the DART development team. We have ideas on how to implement this.)</li>
+
+<li>The default namelist values for the 'obs_seq_to_netcdf' program has changed
+so the default is a single large time bin, which means you don't have to know
+the exact time extents when converting an obs_seq.final file into a netCDF file.
+You can still set specific bins and get multiple netCDF files as output if
+you prefer.</li>
+
+<li>The tutorial files are now directly in the DART/tutorial directory and
+no longer in separate subdirectories.</li>
+
+<li>The default flags in the mkmf_template.XXX files have been updated to
+be more consistent with current compiler versions.</li>
+
+<li>The default work/input.nml namelists for Lorenz 63 and Lorenz 96 have
+been changed to give good assimilation results by default. Originally these
+were set to work with a workshop tutorial in which the settings did not work
+and as part of the tutorial they were changed to good values. Now the
+workshop versions of the namelists are separate and copied into place by
+a workshop_setup script.</li>
+
+<li>filter now calls the end_model() subroutine in the model_mod for
+the first time. It should have been called all along, but was not.</li>
+
+<li>The 'rat_cri' namelist item has been removed from the &obs_diag namelist.</li>
+
+<li>The preprocess program has a new namelist item 'overwrite_output' and it
+is .true. by default. The program will no longer fail if the target
+obs_kind_mod.f90 or obs_def_mod.f90 files exist but will silently overwrite them.
+Set this namelist item to .false. to recover the previous behavior.</li>
+
+</ol>
+
+<p>
+For completeness, the following were changes in the Kodiak
+release (SVN revision number 5018, 15 June 2011), which were
+not backwards compatible with the previous Jamaica branch
+(SVN revision number 2884, 30 April 2007).
+</p>
+
+<ol>
+
<li> &filter_nml used to have a single entry to control whether
to read in both the inflation values and standard deviations
from a file or use the settings in the namelist. The old namelist
More information about the Dart-dev
mailing list