Version History – v53

The following is a brief summary of the ADCIRC development history.

167f586 | 2018-04-19 19:04:38 | Zach Cobell

Merge pull request #109 from zcobell/v53vectorSparseBug. Fixing bug for vector quantities in sparse output

accea7b | 2018-04-19 18:43:14 | Zach Cobell

Fixing bug for vector quantities in sparse output

If a vector quantity was written to a sparse file, only the first value of the vector was checked if it was a default value. This can cause errors when winds, for example, for winds with no U-component

360bfc4 | 2018-04-13 07:26:29 | Jason Fleming

Merge pull request #107 from jasonfleming/closingissues. Closingissues #40 and #41

122dd9b | 2018-04-07 16:28:46 | Jason Fleming

Added code to append ascii files in the proper place when appending output specifier 1 or 4); removed repeated code for writing headers to binary files.

a222de4 | 2018-04-07 08:35:55 | Jason Fleming

Fixed filename and logical unit number reporting in error messages from OWI wind module. Resolves #41.

9ac004f | 2018-03-18 14:03:20 | Jason Fleming

Merge pull request #104 from jasonfleming/boundary3Dfix. 3D fixes for div by zero and uninitialized boundaries

4402985 | 2018-03-17 15:00:56 | Jason Fleming

Fix a division by zero at dry nodes in the gaussian filter; fix failure to set values of lbcodei in mesh.F when running in 3D.

a8ba42a | 2018-03-16 19:22:33 | Zach Cobell

Merge pull request #103 from zcobell/bugfix_097

468e387 | 2018-03-16 16:09:30 | Zach Cobell

Fixing #97. Adding integer formatted write.

263379c | 2018-03-12 09:15:42 | Zach Cobell

Merge pull request #102 from WPringle/master

ad98f0e | 2018-03-10 09:43:01 | Zach Cobell

Merge branch ‘master’ into master

c87a468 | 2018-03-10 09:39:01 | William Pringle

Deleted unnecessary commented lines

815b4fa | 2018-03-10 09:39:01 | Zach Cobell

Adding check for intel compiler and applying heap-arrays option. Fixes #100

The `-heap-arrays [stacksize]` option for the intel fortran compiler forces Intel to use the heap for large implicitly created arrays. This is the default setting for most other compilers. The `work/makefile` has been modified to print a message warning the user to add this option and the CMake build system has been modified to add this setting automatically by querying the current machine’s stack size via `ulimit -s`. This option should also be added to the netcdf-fortran Fortran compiler flags as this is usually where this segmentation fault can occur. The option was added to the ADCIRC build process to prevent a simliar issue.

Additional changes were made to process the `NETCDFHOME` and `XDMFHOME` environment variables directly if they exist after processing the predefined variables that would exist from a cmake command line option, specifically `-DNETCDFHOME` and `-DXDMFHOME`. The command line options are given priority. If both of these cases fail, the user is prompted for the location of the
libraries before the options are allowed to be enabled.

af902c9 | 2018-03-10 09:39:01 | William Pringle

Warnelev to print off global node numbers

0f82b16 | 2018-03-10 09:31:59 | William Pringle

Adding tensor type internal_tide_friction (matrix friction coefficient)

Added support for the internal wave drag (internal_tide_friction fort.13 parameter). This is applied by calling Apply2DInternalWaveDrag subroutine (found in nodalattr.F) in timestep.F after applying the bottom friction to get TK. It returns a TKM 3 x n array which is the matrix of friction coefficients (XX, YY, XY = YX). Thus modifications in GWCE.F and momentum.F were made to support the matrix of friction coefficients. I only applied this into the Mom_Eqs_New_NC and GWCE_New subroutines (not the predictor corrector ones). Note that these modfications are similar to those used to support the old CSWAN_FRIC: Ethan/Joannes’s modified friction. Without internal_tide_friction the code still works as usually does where the offdiagonal cofficients are set to zero and the on-diagonal coefficients will be equal (equal to TK).

Fixed couple of bugs with the SAL and tidal potential in adcirc.F and read_input.F with the new mapping to local node numbers (using find and node_dict)

In timestep.F I added a line from Dam’s code to define the background pressure for a tidal only run, not sure how important this is. I also made sure the warnElevDump logical was used from global module and not defined locally in timestep.F

Fixed mistakes for SAL

In read_input.F the lines to convert the phase to radians and close fort.24 for reading were in the wrong location

Fixing bug for building without netcdf

in write_output.F

Attempted fix for xdmf

IT_Fric was matrix but used array

Attempt to fix essential boundary condition for scalar friction

I had implemented the non-symmetric form of the essential bc which could of been reason for difference in the shinnecok test problem. So I put a check for the scalar friction to then implement the symmetrical formula.

Made mistake in attempted fix for the essential bc

Was order dependent

Revert “Attempted fix for xdmf”

This reverts commit fd5da4c.

Revert “Revert “Attempted fix for xdmf””

This reverts commit 13966ef.

Revert “Made mistake in attempted fix for the essential bc”

This reverts commit 436fc19.

Revert “Attempt to fix essential boundary condition for scalar friction”

This reverts commit b552001.

Fixes in wetdry to update TKM

Wetting and drying can update the TK values which were not then put back into the TKM array. Made the fix here

Essential bc maintaing consistency for symmetrical formula

In momentum.F for the essential bc I put a check for the scalar friction which will then use the symmetric formula instead of the non-symmetrical formula for consistency when running checks

066f7af | 2018-03-10 09:31:21 | William Pringle

Adding prep for SAL and internal_tide

d21e018 | 2018-02-27 15:40:58 | Zach Cobell

Merge pull request #101 from zcobell/intel-heaparrays

4a72459 | 2018-02-23 13:33:52 | Zach Cobell

Adding check for intel compiler and applying heap-arrays option. Fix..
13c0a27 | 2018-02-22 13:32:11 | William Pringle

NetCDF changes to boundary and harmonic outputs, and small fixes for tidal potential and Holland model

Made NBVV and NBDV boundary matrices into vectors when outputting to netcdf file to make file sizes for large domains with many islands etc much smaller

Added in the capabilities for harmonic analysis outputs to fort.51-54.nc (station list or full nodal outputs)

Fixed HollandGet subroutine for the winds in the southern hemisphere (assuming the vortex is centered in northern hemisphere)

In timestep.F the formula for the tidal potential forcing is improved to a general formula for 0,1,2 species numbers.

L_N tidal potential multiple for each species is intialized in adcirc.F

Made the decomposed fort.24 files readable when using padcirc in read_input.F

62e8042 | 2017-12-18 09:04:32 | Jason Fleming

Fixed typo that was causing blended meteorology (NWS29 and NWS30) to have piecewise constant barometric pressure values (changed PRN2->PR2); added compiler flags for gfortran on x86_64.

ce505b0 | 2017-10-15 10:30:11 | Zach Cobell

Merge pull request #95 from zcobell/windCompileBug. Wind and Hotstart bugs

41b691a | 2017-10-14 20:37:07 | Zach Cobell

Merge branch ‘hotstartFix’ into windCompileBug

caf73d6 | 2017-10-14 00:57:46 | Zach Cobell

Fixing writing of hot start velocity data

Hot start velocity data was being written as -9999 instead of 0 which caused crashes upon hotstart. The
VelDescript variable was being used instead of HotstartVelDescript

a0b8ca8 | 2017-10-10 19:59:18 | Zach Cobell

CircleCI: Removing use of cache

Since multiple processor types are used by CircleCI, this can cause issues when using a cached build. Simple solution is to avoid cache and re-build on the target system

23502c3 | 2017-10-10 13:31:49 | Zach Cobell

Fixing initialization of Powell wind drag when hardwiring using -DPOWELL

<p4349b6c | 2017-09-29 11:53:06 | Zach Cobell

Merge pull request #92 from zcobell/hashtableParallel. Fixing bugs in writing output.

e1ad8f0 | 2017-09-29 08:41:33 | Zach Cobell

Fixing nodal_data_id reference in `netcdfio.F` for fort.64.nc

18fb2dd | 2017-09-28 20:53:12 | Zach Cobell

Fixing sparse output for serial run

0135aab | 2017-09-28 13:56:28 | Zach Cobell

Fixing -DHAVE_MPI_MOD comiler flag in CMake

a458e54 | 2017-09-28 01:05:52 | Zach Cobell

Creating routines in globalio.F for full output and binary output

57b7f97 | 2017-09-27 23:21:36 | Zach Cobell

Setting logical default values for alternate_value

Atmospheric pressure, wind velocity, and wind drag are computed even over dry cells and therefore a value of -9999 isn’t logical for sparse output files. Instead, atmospheric background pressure is specified for pressure output, 0.0 for velocity output, and 0.0 for wind drag which reduces file size of sparse output files. Without these values, they are always essentially full format files.

47540fe | 2017-09-27 22:57:56 | Zach Cobell

Fixing remaining issues with sparse/full format output files

e5fe77c | 2017-09-27 19:17:10 | Zach Cobell

Adding hashtable to SWAN

Since local PE directories are now written with global node numbering, the hashtable library needs to be used by SWAN when reading adcirc mesh files.

Note: This change should be brought to the SWAN development team so it can be used in future versions
of the SWAN code.

c201226 | 2017-09-27 14:34:06 | Zach Cobell

Removing reference to cache in circleci

9b87f9c | 2017-09-27 14:29:47 | Zach Cobell

Merge branch ‘master’ into hashtableParallel

c2da481 | 2017-09-27 12:10:04 | Jason Fleming

Merge pull request #90 from jasonfleming/sundryFixes

bcf42af | 2017-09-27 10:15:44 | Jason Fleming

Turned off writing of maxele.63 when NOUTGE.eq.0 and maxvel.63 when NOUTGV.eq.0; fixed negative values of Holland B from aswip for GAHM when central pressure is above 1013mb (issue #85).

246fe3f | 2017-09-27 07:54:53 | Jason Fleming

Fixed echo of element table when NABOUT.gt.1 (issue #82); added execution tracing to wind.F; added end-of-file handling at the beginning of a run to the fort.22 for nws8, nws19, and nws20 as well as a new CALM line type (issue #30); replaced where constructs with loops in collectInundationData and collectMinMax in write_output.F (issue #83).

0d69c8a | 2017-09-26 05:50:31 | Zach Cobell

Merge remote-tracking branch ‘origin/sparseOutputFix’ into hashtableParallel

5eac432 | 2017-09-26 05:40:11 | Zach Cobell

Fixing bug introduced when hashtable module was introduced

Hashtable module previously did not account for parallel simulations. ADCPREP and PADCIRC have been adjusted to use hashtable. This fixes issues in max outputs as well as timeseries output files.

However, for parallel simulations, node numbering is still enforced to be sequential via adcprep. Once the data is in the subdomains, the restriction is lifted.

Finally, note that subdomains now have the labels from the global nodes and not a new set of indicies. This is carried for the fort.13 and fort.14, however, other files currently still use indexing.

bff6729 | 2017-09-25 14:35:02 | Jason Fleming

Merge pull request #80 from zcobell/FluxSettlingTimeFix. Fixing EtaDisc initialization for type 52 flux boundaries

bd539bd | 2017-09-25 10:14:44 | Jason Fleming

Merge branch ‘FluxSettlingTimeFix’ of https://github.com/zcobell/adcirc-cg into FluxSettlingTimeFix

c8b0685 | 2017-09-25 10:13:53 | Jason Fleming

Added a check to ensure that fluxSettlingTime is specified when flux/radiation boundaries (ibtype=52) are present in the mesh file.

d8f5dcf | 2017-09-25 07:43:48 | Zach Cobell

Merge branch ‘master’ into FluxSettlingTimeFix

8f705d8 | 2017-09-25 09:09:07 | Jason Fleming

Merge pull request #88 from jasonfleming/master; fixes for two node island boundaries issue #78 and last subdomain fort.13 issue #86

8d47a36 | 2017-09-24 20:33:43 | Zach Cobell

Merge branch ‘master’ into FluxSettlingTimeFix

82cf142 | 2017-09-24 20:33:11 | Zach Cobell

Merge branch ‘master’ into master

37b4f34 | 2017-09-24 20:32:49 | Zach Cobell

Merge pull request #89 from zcobell/circleci2.0. Adding use of CircleCI v2.0 with Docker.

621d6d1 | 2017-09-24 16:00:44 | Zach Cobell

Adding no_output_timeout command

c8ba012 | 2017-09-24 15:52:30 | Zach Cobell

Adjusting timeout for slow test cases

126d61f | 2017-09-24 14:50:43 | Zach Cobell

Adding use of CircleCI v2.0 with Docker

Docker allows the build environment to function much
more smoothly. There is no longer a need to build the dependencies or install packages when the container starts. This is all handled through the docker environment. The docker environment used can be accessed using the command:

`docker pull zcobell/adcirc_20170924`

This should prevent issues that occured when the automated build scripts used previously encountered errors due to issues outside of their control

438071e | 2017-09-23 19:25:30 | Jason Fleming

Merge branch ‘master’ of https://github.com/jasonfleming/adcirc-cg

5908290 | 2017-09-23 19:24:24 | Jason Fleming

Merge remote-tracking branch ‘upstream/master’

13464cf | 2017-09-23 17:10:36 | Jason Fleming

Merge branch ‘master’ into master

0638347 | 2017-09-19 15:38:15 | Jason Fleming

Fixed issue #86 where the last subdomain fort.13 would not be created if the number of subdomains is divisible by 257 with a remainder of 1.

869bce8 | 2017-09-17 10:12:26 | Jason Fleming

Updated compiler flags for stampede2 at TACC.

a65a646 | 2017-09-17 08:37:22 | Jason Fleming

Disabled the rejection of a partitioning when the only fault is the existence of a single land boundary appearing as a ghost node on a separate subdomain.

e7dfd5d | 2017-09-06 09:46:05 | Zach Cobell

Fixing format specifier for full output format

a76e255 | 2017-09-05 22:09:53 | Zach Cobell

Matching format for files written through writer processors

e7d5360 | 2017-09-05 20:31:41 | Zach Cobell

Applying fix for sparse output files. Additional changes include further checks for the the considerWetDry attribute

a636963 | 2017-08-15 11:55:58 | Zach Cobell

Merge branch ‘master’ into FluxSettlingTimeFix

fc242ef | 2017-08-15 11:55:42 | Zach Cobell

Merge pull request #81 from zcobell/TimeVaryingWeir_ETAMaxFix. Fixing logic when checking for an eta_max type boundary change

bd8e4e3 | 2017-08-14 17:38:00 | jasonfleming

Fixed issue with identifying the correct node number for a hanging boundary node on a subdomain; increased the limit on repartitioning attempts from 10 to 100.

476863e | 2017-08-11 10:38:01 | Zach Cobell

Fixing logic when checking for an eta_max type boundary change

The code was originally set up to first check if either side of the weir was active and then check if either side the weir had an elevation that could trigger the boundary to change elevation. The check against nnodecode needs to occur at the same time as the check against eta2.

The problem becomes when one side of the weir is active and the other side has a topographic elevation greater than the weir change criteria, the code will sense this incorrectly because the active side of the weir will trigger the first check and the
topographic elevation will trigger the second, even though the node is not active. This change performs both checks simultaneously so that this is no longer an issue.

3681fc3 | 2017-08-11 08:23:50 | Zach Cobell

Ensuring correct nodes are reset for lbtype=52

74bbfac | 2017-08-10 14:44:46 | Zach Cobell

Fixing EtaDisc initialization for type 52 flux boundaries

The flux settling time is used to allow the river boundary conditions to equalize before beginning other forcing in the model. It does not appear that it is a hard requirement that it be used, but it is a good practice.

In recent versions of the code, the `EtaDisc` variable has been changes so it is initialized to `-9999.0`. However, if the flux settling time is not used, this value begins to find its way into the equations used to generate the `GWCE_LV` which causes fatal instabilities in the first few model time steps.

Previous versions of the model initialized this value to 0.0, though a more appropriate way to do this seems to be to initialize it to `ETA2` once the geoid offsets and `river_above_msl` attributes have been applied so it is consistent.

This commit initializes `EtaDisc` to `Eta2` during cold start. This does not appear to be necessary at hot start since those values will be read from the hot start file. This commit fixes #79.

0ffbbac | 2017-08-09 09:38:36 | Jason Fleming

Fixes for memory allocation and processing for weir boundaries with cross barrier pipes (ibtypes 5 and 25).

fddac22 | 2017-08-08 18:22:38 | Jason Fleming

Added new command line option –strict-boundaries that causes adcprep to detect when the mesh partitioning will leave one or more subdomains with only one node of any fulldomain boundary; it will also attempt to correct such partitioning by incrementing the weights of those vertices and repartitioning the mesh with the updated weights.

8137e6d | 2017-07-08 22:21:21 | Jason Fleming

Merge pull request #77 from jasonfleming/master

10feb6a | 2017-07-08 17:36:51 | Jason Fleming

Fixed allocation of offset stations output so it uses mnstae instead of nstae; added debugging flags for bus error and heap arrays issues for intel compilers; added a formatting check to NWS20 (GAHM) to detect format issues in fort.22 where there are more than 4 isotachs.

16daf77 | 2017-07-08 11:30:52 | Jason Fleming

Removed blanket USE statements from ADCIRC_Mod and added fine grained USE statements to each subroutine inside ADCIRC_Mod in an attempt to resolve bus error issue.

810d624 | 2017-06-27 10:21:07 | Jason Fleming

Merge pull request #75 from jasonfleming/blendoffset

0257d9c | 2017-06-26 14:48:38 | Jason Fleming

Merge branch ‘master’ into blendoffset

203f27a | 2017-06-26 14:43:44 | Jason Fleming

Fixed double allocation issue for vortexWVNX2 and vortexWVNY2.

6f91fe8 | 2017-06-26 08:46:34 | Zach Cobell

Merge pull request #74 from jasonfleming/master

5ae8227 | 2017-06-26 07:46:22 | Jason Fleming

Fixes for compilation issues.

f027f4d | 2017-06-26 06:56:09 | Jason Fleming

Fixed issue where ascii offset.61 header was not being written on hotstart and was always written on coldstart regardless of the sign of NOUTE; added feature where the offset surface file will be opened in globaldir, or if it is not found there, will be opened as if it contained the full path to the file.

c285edd | 2017-06-26 06:52:21 | Jason Fleming

Fixed issue where the offset.61 file output was active even when the water level offset namelist was not present in the fort.15.

a79dea7 | 2017-06-26 06:46:53 | Jason Fleming

Fixes for offset.61 file to write ascii header only when required; clarification of long name in netcdf metadata.

7187368 | 2017-06-26 06:46:34 | Jason Fleming

Added capability to write offset.63 and offset.61 in netcdf format.

9166f0c | 2017-06-26 06:46:05 | jasonfleming

Fixed compilation issues associated with offset.61; resolved conflicts between blendgahm and master.

6177a8f | 2017-06-26 06:29:10 | jasonfleming

Merge branch ‘blendgahm’ of https://github.com/jasonfleming/adcirc-cg into blendgahm; fix conflicts between blendgahm and master.

74ab1cf | 2017-06-26 06:02:39 | jasonfleming

Changes to support offset.61 file; resolved conflicts between blendgahm and master.

a1b3f0c | 2017-06-26 05:57:01 | jasonfleming

Fixed gahm+nam blending by enabling negative values of nws 30 in adcprep; resolved conflicts between blendgahm and master.

f4a93ec | 2017-06-25 12:31:42 | Jason Fleming

Fixed water level offset by subtracting it from the barotropic pressure in the momentum equation; created two time levels of water level offset data in adcirc; resolved conflicts between blendgahm and master.

1e4b419 | 2017-06-25 12:20:14 | jasonfleming

Fixed the way the offset is added to the boundary so that it uses the value at the boundary node rather than at node i; resolved conflicts between blendgahm and master.

6a080af | 2017-06-25 12:15:32 | Jason Fleming

Fixed ramping so that ramp multiplier value is zero prior to the start of the ramp period.

d04b8af | 2017-06-25 12:15:18 | jasonfleming

Added water level offset to elevation specified boundaries; resolved conflicts between blendgahm and master.

df15d99 | 2017-06-25 12:05:12 | Jason Fleming

Moved the initialization of the water level offset so it is before initOutput2D in adcirc.F to resolve a segmentation fault in output data descriptors when writing water level offset to the output file.

111c543 | 2017-06-25 12:04:09 | jasonfleming

Added initialization of offset.63 output file.

9fc6d68 | 2017-06-25 12:03:37 | jasonfleming

Added offsetTimeReference parameter to control whether ramp times are relative to hotstart or coldstart time.

a182b41 | 2017-06-25 12:03:21 | jasonfleming

Fixed error handling for water level offset files; resolved conflicts between blendgahm and master.

46c7f9d | 2017-06-25 09:59:25 | Zach Cobell

Updating link to HDF5 source code in installHdf5.sh

b367b08 | 2017-06-25 11:53:33 | Jason Fleming

Added support for water level offset control namelist to adcprep; added support for writing water level offset output data; added support for water level offset as a pseudo barometric pressure in gwce_new(); resolved conflicts between blendgahm and master.

3a8fc5a | 2017-06-25 11:30:05 | Jason Fleming

Added code to deal with NWS=-29 and NWS=-30; consolidated code for hotstarting NWS=+/-12; added code to compute average Rmax value for NWS=20.

3256d92 | 2017-06-25 10:36:25 | Jason Fleming

Fixes to resolve conflicts between blendgahm and master.

5dc1782 | 2017-06-25 05:59:50 | Jason Fleming

Merge branch ‘master’ of https://github.com/jasonfleming/adcirc-cg

383253a | 2017-06-25 05:58:56 | Jason Fleming

Added compiler flags for lonestar5.tacc.utexas.edu.

245e9ee | 2017-06-25 06:56:41 | Jason Fleming

Changed one REAL to REAL(4) declaration in couple2swan.F for use on intel-sgi platforms (e.g., Thunder at AFRL).

891ce87 | 2017-06-24 10:54:01 | Jason Fleming

Added compiler flags for thunder at AFRL.

e224129 | 2017-05-13 16:06:18 | Jason Fleming

Merged nws30 changes.

fe85e28 | 2017-05-13 15:41:38 | Jason Fleming

Added back missing variable declaration for etratio to hstart.F.

b52dca6 | 2017-05-13 15:25:21 | Jason Fleming

Added code to support NWS=30, blending GAHM (NWS=20) vortex winds with OWI (NWS=12) background winds

9cd5626 | 2017-05-08 15:16:28 | Zach Cobell

Merge pull request #70 from caseydietrich/master

eb51169 | 2017-05-08 11:53:25 | Casey Dietrich

Initializing the background pressure Penv inside src/owiwind.F

f3a4d5d | 2017-05-05 17:47:21 | Zach Cobell

Merge pull request #68 from zcobell/cmakeXdmf

0d5ec2e | 2017-05-05 13:56:53 | Zach Cobell

Merge branch ‘master’ into cmakeXdmf

1af6380 | 2017-05-05 06:35:45 | Jason Fleming

Merge pull request #67 from atdyer/master

a265d72 | 2017-05-04 10:50:11 | tristan

Harmonics files (53, 54) pass node numbers through hash table

87e419e | 2017-05-04 10:22:43 | tristan

Fix bug where station numbers are passed through the hash table before output

9bd483e | 2017-05-02 16:36:55 | Tristan

Remove leftover text from merge

dbf2191 | 2017-05-02 16:13:19 | Tristan

Fix conflicts to complete merge

4f9182a | 2017-05-02 07:49:45 | jasonfleming

Updated docs to add reference to MACHINENAME in cmplrflags.mk and to remove references to svn

495f8e5 | 2017-04-28 16:11:57 | Jason Fleming

Merge pull request #65 from jasonfleming/master

49b9443 | 2017-04-28 08:39:40 | Jason Fleming

Fixed compilation issues associated with v53dev refactoring of wind.F in comparison with the nws10init and nws10get subroutines from v51release

a81d0e8 | 2017-04-28 08:18:36 | Jason Fleming

Merged GFS changes that we made to v51release but failed to make in trunk prior to v52release

bb7022a | 2017-04-28 07:12:50 | Jason Fleming

Added command for merging upstream/master.

3d170ee | 2017-04-28 07:08:57 | Jason Fleming

Merge remote-tracking branch ‘upstream/master’

f7a91b7 | 2017-04-28 07:06:05 | Jason Fleming

Added recipe for keeping fork synchronized with upstream.

1452762 | 2017-04-28 05:40:12 | Jason Fleming

Merge pull request #64 from jasonfleming/master

cea6213 | 2017-04-27 16:20:35 | Zach Cobell

Merge branch ‘master’ into master

2e64ad7 | 2017-04-27 15:50:26 | Jason Fleming

Added recipe for looking at version history.

394977a | 2017-04-27 15:30:20 | Jason Fleming

Added recipe for cherry picking from branch-of-a-branch back to master/p>
3b362b0 | 2017-04-25 10:54:56 | Zach Cobell

Adding more explicit find method for xdmf

34d738c | 2017-04-25 10:39:40 | Zach Cobell

Merge pull request #62 from zcobell/xdmfHotfix

f00b6dc | 2017-04-24 19:09:08 | Zach Cobell

Updating linking for Xdmf on the build server which was causing issues

359a97e | 2017-03-07 09:11:07 | Zach Cobell

Merge pull request #33 from zcobell/automaticVersionTracking

d34d565 | 2017-03-06 09:23:55 | Zach Cobell

Fixing dependencies of aswip in cmake build

c73aa96 | 2017-03-05 20:16:58 | Zach Cobell

Removing Swan Modified Friction option

a32ef12 | 2017-03-05 20:16:58 | Zach Cobell

Using FindNetCDF.cmake from VTK repository to locate netCDF configuration

fb64c61 | 2017-03-05 20:16:58 | Zach Cobell

Adding CMake macros to simplify files for each executable

2dfe0ef | 2017-03-04 08:16:48 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

a4265ca | 2017-03-04 07:18:12 | Jason Fleming

Merge pull request #35 from zcobell/3dUpdates-UTexas

a4da2a2 | 2017-03-04 00:00:09 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

0f81c87 | 2017-03-03 23:52:50 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

8c317aa | 2017-03-03 23:52:37 | Zach Cobell

Merge pull request #57 from jasonfleming/nocswanfric. Removed code included by CSWANFRIC preprocessor symbol

ddff7f9 | 2017-03-03 21:23:34 | Jason Fleming

Removed code included by CSWANFRIC preprocessor symbol because it has been deemed vestigial.

6c93437 | 2017-03-03 15:09:51 | Jason Fleming

Added back changes to GWCE_New() lost in a code conflict; added BCFLAG_LNM to ONLY list for USE GLOBAL in timestep.F.

cbd84a5 | 2017-03-03 14:21:11 | Jason Fleming

Merge branch ‘master’ into 3dUpdates-UTexas

be76aff | 2017-03-03 11:03:13 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

1dd284b | 2017-03-03 11:02:50 | Zach Cobell

Merge pull request #50 from jasonfleming/modtimestep, modularized timestep() into modules: gwce, momentum, and wetdry

cb3e44d | 2017-03-03 09:32:37 | Jason Fleming

Changed the initializeBoundaries subroutine so that boundaries with less than the minimum interior angle will only have their boundary type value changed (i.e., ibtype=0 is changed to ibtype=10, or 1 -> 11 or 2 -> 12 or 3 -> 13, in the case of a 2DDI solution; the 3D solver does not seem to support boundary types with no tangential slip (10 to 13); the ultimate fix for this is to modify the 3D solver for these boundary types or for the 3D solver to reject such boundaries and exit with an error.

fc9be26 | 2017-03-02 21:33:00 | Jason Fleming

Merge branch ‘modtimestep’ of https://github.com/jasonfleming/adcirc-cg into modtimestep

cfed9e2 | 2017-03-02 21:31:49 | Jason Fleming

Reduced scope of USE SIZES in global module; removed unused variables from global module; removed duplicate variable declarations between global and momentum modules; pushed selection of gwce and momentum formulations out of subroutine timestep() and into those modules; moved slope limiter subroutines into gwce module; reduced scope of USE GLOBAL to only those variables needed in GWCE_New and Momentum_Eqs_New_NC; moved initialization of winddraglimit and draglawstring to wind module variable declaration instead of wind initialization subroutine so that it could be overridden by values from read_input().

eb2f4a0 | 2017-02-28 20:16:34 | Zach Cobell

Moving the versioning scripts to their own folder. Adding a warning to the user when exporting if the repository has been modified. Adding a modified tag to all version numbers generated.

23c2132 | 2017-02-27 08:57:43 | Zach Cobell

Updating dependencies in cmake build

afedc37 | 2017-02-27 10:07:17 | Jason Fleming

Removed duplicate variable declarations between global.F and wetdry.F to fix numerical instability issue in shinnecock inlet test; made further progress on modularization of variables out of global.F; made module dependencies of gwce.F and momentum.F on global.F explicit in use statements.

7b31a0c | 2017-02-25 09:41:17 | Jason Fleming

Added debugging statements to highlight the occurrence of a division by zero issue

55ec6c9 | 2017-02-25 08:42:06 | Jason Fleming

Removed double declaration and double allocation of wvnxout and wvnyout

b5bea76 | 2017-02-25 08:26:43 | Jason Fleming

Merge branch ‘modtimestep’ of https://github.com/jasonfleming/adcirc-cg into modtimestep

52367d4 | 2017-02-25 08:25:09 | Jason Fleming

Moved the initialization of the wind module from the top of read_input() to adcirc_init() so that the value of G could be read from fort.15 before the initialization of the meteorological forcing constants.

fa9b9c9 | 2017-02-24 20:47:35 | Jason Fleming

Merge branch ‘master’ into modtimestep

1cf56d5 | 2017-02-23 18:33:27 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

607f2ba | 2017-02-22 21:41:50 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

658d026 | 2017-02-22 14:23:48 | Zach Cobell

Merge pull request #56 from adcirc/ncsu-subdomain

c787eca | 2017-02-22 12:14:47 | Zach Cobell

Merge branch ‘master’ into ncsu-subdomain

8481ee3 | 2017-02-22 13:49:51 | Alper Altuntas

Removed duplicate calls to readFort019()

d433ca4 | 2017-02-21 23:47:59 | Zach Cobell

Merge pull request #53 from jasonfleming/swell. Capability to write windDrag output plus use SWELL drag coef from SWAN

96aef51 | 2017-02-21 21:15:07 | Jason Fleming

Merge branch ‘modtimestep’ of https://github.com/jasonfleming/adcirc-cg into modtimestep

66398e3 | 2017-02-21 21:14:21 | Jason Fleming

Re-implemented smagorinsky changes that were lost in a code conflict.

f060a5f | 2017-02-21 17:31:07 | Zach Cobell

Merge branch ‘master’ into swell

a7df2fb | 2017-02-21 17:30:32 | Zach Cobell

Merge pull request #55 from zcobell/sparseOutputFormat.

Fixing sparse output format for serial runs.

3aa8991 | 2017-02-21 17:29:21 | Jason Fleming

Merge branch ‘master’ into modtimestep

1db757a | 2017-02-21 13:07:43 | Zach Cobell

Fixing sparse format for vector files

d8b82bd | 2017-02-21 12:31:44 | Zach Cobell

Fixing sparse output format for serial runs. The sparse output format was not being correctly written for serial runs. The output file was blank except for the header for each time snap. Also, removing the “-DDEBUG” flag in favor of “-DEBUG”. (i.e. ifdef EBUG) This avoids a situation where the compiler defines variables within the code to “1”, which is not the intention

27c6886 | 2017-02-21 10:01:03 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

f47760b | 2017-02-21 06:43:56 | Zach Cobell

Merge branch ‘master’ into swell

4b714ec | 2017-02-21 06:40:09 | Zach Cobell

Merge pull request #51 from jasonfleming/3Dfix

982f044 | 2017-02-21 01:55:51 | Zach Cobell

Fixing dependencies in makefile

2fe3f71 | 2017-02-21 00:45:06 | Jason Fleming

Added capability to write windDrag values as a new output file by setting the outputWindDrag parameter to .true. in the &metControl namelist in the fort.15; added capability to use SWELL wind drag from SWAN in ADCIRC in coupled simulations (from @mattbilskie).

6824965 | 2017-02-20 18:15:20 | Jason Fleming

Fix for segmentation fault during initialization by making pointers from 3D variables to 2DDI counterparts explicit in global_3dvs.F and then fixing references to these variables throughout the code and adjusting makefile to reflect new dependency relationships.

a6948e1 | 2017-02-20 18:12:40 | Zach Cobell

Merge pull request #52 from jasonfleming/master

f123318 | 2017-02-20 16:16:02 | Zach Cobell

Adding updating Xdmf install script

a710f1e | 2017-02-20 14:24:11 | Zach Cobell

Fixing issues related to variable types in wind module

ace8f8c | 2017-02-20 14:36:14 | Chris Massey

Added a new namelist to the fort.15 file called Smag_Control.

This namelist has three variables associated with it 1) SMAG_COMP_FLAG (T/F), 2) SMAG_UPPER_LIM (real) and 3) SMAG_LOWER_LIM (real). If SMAG_COMP_FLAG = .true. then Smagorinski Eddy Viscosityis computed and it’s value tested against the user supplied Upper and Lower Limits. This new flag will override the existing Smagorinski flags that were triggered within the ADCIRC read_input.F but was not allowed for use in several of the Momentum EQ formulations for some reason. Namelist Default Values are as follows SMAG_UPPER_LIM = 100.d0 m^2/sec SMAG_LOWER_LIM = 1.0d-8 m^2/sec SMAG_COMP_FLAG = .FALSE.

051dad4 | 2017-02-20 11:13:45 | Zach Cobell

Fixing typo in Xdmf library path

f5a38d7 | 2017-02-20 10:19:01 | Zach Cobell

Fixing xdmf install location in circleci scripts

7f39ebe | 2017-02-20 06:58:44 | Zach Cobell

Updating CMake build files for new modules created from timestep.F

5c9bbc7 | 2017-02-20 06:40:47 | Zach Cobell

Merge branch ‘master’ into modtimestep

bc15663 | 2017-02-18 13:43:31 | Jason Fleming

Added new wetdry module.

29d3946 | 2017-02-18 13:42:28 | Jason Fleming

Modularized wetting and drying out of timestep() and into its own module.

7b015b6 | 2017-02-18 11:36:49 | Jason Fleming

Modularized met forcing data management out of timestep.F and into wind module.

c5b187c | 2017-02-17 09:45:19 | Jason Fleming

Created momentum module from timestep.F for solving momentum equations

0692c86 | 2017-02-17 09:23:52 | Jason Fleming

Created GWCE module from subroutines in timestep.F.

bdaa3e9 | 2017-01-27 10:42:12 | Zach Cobell

Handling version_cmake.F file when conducting Windows Native build.

The windows native build will assume that it does not have to generate the version file itself. Instead, it will fall back to a version file made available in the main source directory.

By default, it looks for version.F, however if not found it will fall back to version_default.F.

5874a2c | 2017-01-27 09:00:04 | Zach Cobell

Adding compression method option for adcirc export

244d618 | 2017-01-27 08:13:52 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

31e85dd | 2017-01-27 07:41:46 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

af08718 | 2017-01-23 20:44:48 | Alper Altuntas

Removed UnitNumber assign in subOpenPrep() to fix segfault with GNU compilers

ad348fb | 2017-01-15 22:19:15 | Zach Cobell

Setting CMake to look for NETCDFHOME and XDMFHOME (#39)

At configure time, CMake will look for these two environment variables and test the code against what it finds. If valid, the code will use what is found here or the user can set a different path. If these variables are not set, the user will need to set these manually if they want to enable these options.

6c6e4eb | 2017-01-15 12:15:49 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

953262a | 2017-01-15 12:14:59 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

f24ae52 | 2017-01-14 09:17:37 | Jason Fleming

Added MinGW compiler flags from @natedill.

6a4e4de | 2017-01-08 23:42:42 | Zach Cobell

Adding ‘-assume byterecl’ for intel based compilers

48cc021 | 2016-11-27 14:42:45 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

d80ccd3 | 2016-11-27 14:32:46 | Zach Cobell

Merge branch ‘master’ into automaticVersionTracking

0335c41 | 2016-11-27 13:34:17 | Zach Cobell

Addition of CMake build system (#36)

Traditional work/makefile with cmplrflags.mk is maintained for backwards compatibility. The adcirc_Xdmf.f file has been moved to the src folder so it is found by both cmake and traditional make. In order for GNU compilers to more easily process SWAN f90 files, some long lines have been wrapped

Enabling build caching on CircleCI to speed up builds. This includes some scripts set up to build the dependencies if they do not appear in the cache.

Updating the BUILD.md documentation and ensuring CMake compatibility with Windows.

Adding 32-64 bit compatibility checks and netCDF checks

Adding XDMF and NETCDF checks

Finalizing XDMF test code

Ensuring CMake generation doesn’t occur with incorrect netCDF of XDMF libs

Ensuring that netCDF and XDMF libs throw an error that doesn’t let CMake generate makefiles until the code compile tests pass.

Also adding the templib prefix to temporary libaries.

Adding more flags from ADCIRC code. Additional compat checking with MPI

Splitting full CMakeLists.txt into smaller files to make them more manageable

Using windows API for mkdir when compiling under windows

This removes the need for #15 using some compiler specific checks to apply Fortran compatibility flags for longer line lengths. Closes #15

acc49db | 2016-11-23 01:27:11 | Zach Cobell

Fixing dependencies for ASWIP

fd3acf7 | 2016-11-22 13:33:40 | Zach Cobell

Fixing missing reference to Boundaries module in global_3dvs.F

4c0d770 | 2016-11-21 16:33:01 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

cf3cc59 | 2016-11-21 16:19:45 | Zach Cobell

Updating link to Hdf5 source

af343bb | 2016-11-21 15:45:35 | Zach Cobell

Fixing broken link ot HDF5 source code

6a20db5 | 2016-10-24 12:44:48 | Zach Cobell

Removing #ifdef CASEY debugging statements

502f6ee | 2016-10-24 11:40:49 | Zach Cobell

Removing long lines in SWAN source that caused issues with GNU Fortran compiler

05daa13 | 2016-10-24 11:40:13 | Zach Cobell

Merge branch ‘master’ into 3dUpdates-UTexas

d54fef6 | 2016-10-24 10:29:29 | Zach Cobell

Adding changes for 3D from Arash @ UTexas. The following are addressed in this commit:

1. Evaluating the baroclinic pressure gradient using a cubic polynomial interpolation in subroutine bpg3d() in timestep.F.

2. Using a biharmonic operator (as opposed to a Laplacian operator) for viscosity in the momentum and diffusion in the transport equations (vsmy.F and transport.F).

3. Evaluating the viscosity and diffusion coefficients by using a modified Leith approach (vsmy.F).

4. Adaptive filtering scheme for the velocity field (vsmy.F).

5. Fixed a bug in adcprep that pertains to the baroclinic rivers (prep.F).

830cf05 | 2016-10-03 11:40:39 | Zach Cobell

Adding scripts for automatic version tracking

9e73815 | 2016-10-02 21:07:25 | Zach Cobell

Adding scripts to do installs for packages on circleCI. Adding scripts that perform installs only if necessary. Cached copies are used if available. This should speed up builds.

e81e6b1 | 2016-09-30 17:49:46 | Zach Cobell

Merge pull request #32 from zcobell/netcdfHarmonicsFix. Fixes issue #31: Harmonic analysis netCDF segfault

2cc2425 | 2016-09-30 15:07:35 | Jason Fleming

Fixes issue #31: when using a netcdf hotstart file containing harmonic analysis data in parallel, a segmentation fault will occur if there are zero stations.

457bae6 | 2016-09-16 09:50:45 | Zach Cobell

Merge pull request #28 from zcobell/master. Fixing logging in weir_boundary.F

7d8c7b4 | 2016-09-15 23:45:30 | Zach Cobell

Removing multiple references to ALL_TRACE

51fa915 | 2016-09-15 16:04:56 | Zach Cobell

Adding return logging to weir_boundary.F. Enter was correctly being logged, however, many of the returns were not logged using allMessage.

c7e2fa7 | 2016-09-15 12:36:19 | Zach Cobell

Merge pull request #27 from zcobell/weirBugs-issue25. Fixing issues described in #25 related to weir boundaries.

51a8b75 | 2016-09-15 12:53:29 | Zach Cobell

Fixing issues described in #25 related to weir boundaries.

1. Fixed variables referenced in mesh.F causing array bounds violations

2. Fixed missing calls to unsetMessageSource in weir_boundary.F

053ec87 | 2016-08-04 13:05:03 | Zach Cobell

Merge pull request #23 from zcobell/circleci-swan-xdmf. Fixing SWAN serial/parallel compilation issue. Adding SWAN to CircleCI tests. Adding gitignore.
2016-08-02 15:57:07 | Zach Cobell

Fixing SWAN serial/parallel compilation issue. Adding gitignore. SWAN runs a script to pre-process files as part of the build process. This process is different for serial and parallel modes. The process needs to be cleaned if changing from a serial compile (adcswan) to a parallel comple (padcswan). The makefile now keeps track of this via a hidden file .swanlastcompile.

The circle.yml file has been updated for SWAN and XDMF compiles. An issue was fixed with the XDMF library locations. It now expects the standard directory tree (lib/include/share/bin). Use XDMFHOME to specify the location of the directory just as you would with NETCDFHOME. The HDF5, netCDF, netCDF-Fortran, and XDMF libraries are now built by the build server since it was unable to use the pre-installed versions.

ASWIP did not compile with XDMF enabled. This has been resolved.

Finally, a .gitignore has been added to keep track of files that are not supposed to be included in the repository but might be included by mistake.

9aef131 | 2016-08-01 10:26:26 | Zach Cobell

Merge pull request #22 from caseydietrich/master. Updating to SWAN release version 41.10

88390aa | 2016-08-01 09:34:32 | Zach Cobell

Fixing typo in README.md

8fbee18 | 2016-08-01 09:03:48 | Zach Cobell

Deselecting intel compiler

d7352f9 | 2016-08-01 08:58:36 | Zach Cobell

Removing actualflags.txt

619c7bc | 2016-08-01 08:57:58 | Zach Cobell

Merge remote-tracking branch ‘upstream/master’

24a38a5 | 2016-08-01 09:27:44 | Casey Dietrich

Merging my branch for SWAN v41.10 into my forked master.

62360fc | 2016-07-29 18:01:21 | Zach Cobell

Adding bc to dependency list

54a1b86 | 2016-07-29 17:58:48 | Zach Cobell

Updating link to test suite for CI

e60c9c8 | 2016-07-29 17:15:25 | Zach Cobell

Adding status badge to README.md

2e40990 | 2016-07-29 15:27:08 | Zach Cobell

Updating reference for CircleCI

bb7381c | 2016-07-29 15:24:30 | Zach Cobell

Merge pull request #20 from zcobell/CircleCI_Setup. Adding CircleCI Continuous Integration files.

ee0ad02 | 2016-07-29 15:20:29 | Zach Cobell

Updating reference to test suite

a1fdad7 | 2016-07-29 14:25:40 | Zach Cobell

Adding development reference to version.F

c2bb467 | 2016-07-29 14:21:54 | Zach Cobell

Merge pull request #21 from zcobell/documentation

ff991c1 | 2016-07-29 14:18:39 | Zach Cobell

Adding reference to officially supported branches

4a6b47e | 2016-07-29 14:02:57 | Zach Cobell

Using full path names when compiling. This enables code coverage options but does not change anything for other users.

e8db83e | 2016-07-28 20:57:50 | Zach Cobell

Adding CircleCI Continuous Integration files

811140d | 2016-07-28 18:41:27 | Zach Cobell

Fixing enumeration

f8c5621 | 2016-07-28 17:54:56 | Zach Cobell

Adding section on CI

5ddb7c8 | 2016-07-28 12:44:17 | Casey Dietrich

Adding the latest release version 41.10 of SWAN, with changes to the ADCIRC makefile for compilation of padcswan.

4da1b38 | 2016-07-28 11:13:52 | Casey Dietrich

Removing the existing SWAN subdirectory to prepare for adding the newest release version.

f1e4fce | 2016-07-28 09:43:54 | Zach Cobell

Update README.md

92a7ba8 | 2016-07-27 14:05:41 | Zach Cobell

Merge pull request #19 from zcobell/adcprepNetCDFSegFault

d523bb0 | 2016-07-27 14:36:41 | Zach Cobell
Adding missing call to check_err

ef3a978 | 2016-07-26 23:59:52 | Zach Cobell

Removing other locations where implicit arrays were created

fcb9124 | 2016-07-26 23:16:43 | Zach Cobell

Fixes #17. Avoid creating implicit array using “spread” Implicit array creation seemed to cause segfault for some systems Also, prep.F had rads.64.nc set to be treated as a min/max file. Not sure of the consequence of this, but it is fixed anyway

1e4d47c | 2016-07-25 09:31:46 | Zach Cobell

Fixing numbering in git tools section

02cffa7 | 2016-07-20 17:50:50 | Zach Cobell

Adding more specific checkout instructions

61cb40d | 2016-07-20 15:28:06 | Zach Cobell

Adding markdown documentation files

744c6e7 | 2016-07-20 15:27:34 | Zach Cobell

Adding commit template

2f32ee9 | 2016-07-20 15:26:09 | Zach Cobell

Reorganizing files in preparation for markdown files

v53.00 – 07/20/2016 – jgf

Committing changes made to v52release branch revisions 1208:1223 (v52.30.01 to v52.30.07) to trunk:

v52.30.07 – 07/20/2016 – jgf

Applied tcm fix that originally went into v51.52.38 r1210 of v51release branch and includes makefile fixes for building XDMF support into libadc.a as well as:

06/07/2016 – tcm – Added unsetMessageSource calls after the writer_main and hswriter_main calls in ADCIRC_Run, otherwise for coupled simulations where ADCIRC_Run is called multiple times, the Message number will run out of dimensions.

v52.30.06 – 07/14/2016 – jgf

Fixed issue in adcprep where the finding of the time varying weirs list was erroneously reported.

v52.30.05 – 06/25/2016 – jgf

Extended workaround for the issue where ADCIRC opens i/o unit numbers currently in use by SWAN so that it covers any/all i/o unit numbers that ADCIRC might use (this issue was causing mysterious hangs in parallel); changed GFS read routine (NWS10) so that it always uses the same unit number to avoid collisions with SWAN.

v52.30.04 – 06/18/2016 – jgf

Enabled inverted barometer modification of elevation specified boundaries using metControl namelist in fort.15 with invertedBarometerOnElevationBoundary=.true.

v52.30.03 – 06/11/2016 – jgf

Fixed issue in reading levee boundaries with cross barrier pipes (ibtypes 5 and 25) where the counter variable for these boundaries was not being incremented.

v52.30.02 – 06/07/2016 – jgf

Moved handling of rads.64 file outside of #ifdef CSWAN so that it could be written without SWAN coupling.

v52.30.01 – 06/06/2016 – jgf

Fixed issue when reading min/max files in netCDF in parallel resulting from reading the ADCIRC time from the netCDF file into an unallocated array.