http://carson.math.uwm.edu/jenkins/blue/organizations/jenkins/cam_clubb_generalized_grid_scam_test/Performs a generalized grid test for CAM-CLUBB in single-column mode. This test is disabled for the time being, since the reversible grid direction code in CAM has been un-implemented. https://github.com/larson-group/cam/commit/51174997cc51ad7841c74ab1821e139a9f39936dcam_clubb_generalized_grid_scam_testcam_clubb_generalized_grid_scam_testcam_clubb_generalized_grid_scam_testcam_clubb_generalized_grid_scam_testhttp://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/falseStarted by an SCM change1532e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1514841891642148cam_clubb_generalized_grid_scam_test #1515false1511277SUCCESS1759392306858http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/15/latin_hypercube_driver_module.F9050f5a9aa2d2a1e4894e9d0aa40f653db463beb151759358185000http://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerhuebler@uwm.eduMaking the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true. 2025-10-01 17:36:25 -050050f5a9aa2d2a1e4894e9d0aa40f653db463beb15Making the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true.editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerfalse14http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/Started by an SCM change141be51d053905fea48cebe2371a516e113a4963121be51d053905fea48cebe2371a516e113a496312refs/remotes/origin/clubb_silhs_devel1be51d053905fea48cebe2371a516e113a4963121be51d053905fea48cebe2371a516e113a496312refs/remotes/origin/clubb_silhs_devel1be51d053905fea48cebe2371a516e113a4963121be51d053905fea48cebe2371a516e113a496312refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1415207101642148cam_clubb_generalized_grid_scam_test #1414false1411188SUCCESS1759305906745http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/latin_hypercube_driver_module.F90c65ac4caa14049aac533a666cae096b427ea43881759289999000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comSILHS reproducibility (#1258) * Making the SILHS random number generator on GPUs more reproducible by relying on the lh_seed input every call, which is the same as how the CPU randoms work. Also ensuring that the seed changes each timestep, just to prevent generating the same randoms on different timesteps. Since this does change the randoms, it is BIT_CHANGING for silhs enabled runs. * Improving comment 2025-09-30 22:39:59 -0500c65ac4caa14049aac533a666cae096b427ea4388SILHS reproducibility (#1258)editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse15http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/15/13http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/13/Started by an SCM change13aa62f2a748b177007297c0f42dc08bc7539f7cc0aa62f2a748b177007297c0f42dc08bc7539f7cc0refs/remotes/origin/clubb_silhs_develaa62f2a748b177007297c0f42dc08bc7539f7cc0aa62f2a748b177007297c0f42dc08bc7539f7cc0refs/remotes/origin/clubb_silhs_develaa62f2a748b177007297c0f42dc08bc7539f7cc0aa62f2a748b177007297c0f42dc08bc7539f7cc0refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1319215441642148cam_clubb_generalized_grid_scam_test #1313false1311048SUCCESS1759046708160http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/13/constants_clubb.F9006d501ccaa052143c7142816c8a30b354593ff061758974591000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comChange tolerance in way that affects only single precision builds (#1256) This fixes a problem with precip_frac exceeding a bound in single precision. The code change is BIT_CHANGING for single precision, but V. Larson wouldn't expect it to be bit changing for double precision, although V. Larson hasn't checked. 2025-09-27 07:03:11 -050006d501ccaa052143c7142816c8a30b354593ff06Change tolerance in way that affects only single precision builds (#1256)editconstants_clubb.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse14http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/12http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/12/Started by an SCM change12e94d063b32865a628aea8b86ae7a394883932653e94d063b32865a628aea8b86ae7a394883932653refs/remotes/origin/clubb_silhs_devele94d063b32865a628aea8b86ae7a394883932653e94d063b32865a628aea8b86ae7a394883932653refs/remotes/origin/clubb_silhs_devele94d063b32865a628aea8b86ae7a394883932653e94d063b32865a628aea8b86ae7a394883932653refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1215394111642148cam_clubb_generalized_grid_scam_test #1212false1210956SUCCESS1758960308384http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/12/grid_adaptation_module.F904660a606dd89b3b53da4c162f66d6571907d74ee1758908234000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comRemove module vars from grid_adaptation_module (#1255) * removed fixed_min_gr_dens module variables * remove gr_dens_old_global variables * remove counter and cumulative module variables 2025-09-26 19:37:14 +02004660a606dd89b3b53da4c162f66d6571907d74eeRemove module vars from grid_adaptation_module (#1255)editgrid_adaptation_module.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse13http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/13/11http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/11/Started by an SCM change11e6a8375659be8bb0f6850a8d0170d5834dc51f20e6a8375659be8bb0f6850a8d0170d5834dc51f20refs/remotes/origin/clubb_silhs_devele6a8375659be8bb0f6850a8d0170d5834dc51f20e6a8375659be8bb0f6850a8d0170d5834dc51f20refs/remotes/origin/clubb_silhs_devele6a8375659be8bb0f6850a8d0170d5834dc51f20e6a8375659be8bb0f6850a8d0170d5834dc51f20refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1118629691642148cam_clubb_generalized_grid_scam_test #1111false1110861SUCCESS1758873908498http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/11/precipitation_fraction.F90f68b263becf1bcfb48213729fd4ee43dac9d67801758846454000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comPrecip frac fix (#1254) * Using epsilon() to help check precip_frac against precip_frac_tol * Using clubb version of eps for checks Including BIT_CHANGING:06288ac66a3a26013c135d8aef776cc99202f64f because a ifort compiler option changed, and that changes the results. Otherwise it was all BFB. 2025-09-25 19:27:34 -0500f68b263becf1bcfb48213729fd4ee43dac9d6780Precip frac fix (#1254)editprecipitation_fraction.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse12http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/12/10http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/10/Started by an SCM change1031a990b91e4bb1507048326cb9b66eaa5de409a431a990b91e4bb1507048326cb9b66eaa5de409a4refs/remotes/origin/clubb_silhs_devel31a990b91e4bb1507048326cb9b66eaa5de409a431a990b91e4bb1507048326cb9b66eaa5de409a4refs/remotes/origin/clubb_silhs_devel31a990b91e4bb1507048326cb9b66eaa5de409a431a990b91e4bb1507048326cb9b66eaa5de409a4refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1015508971642148cam_clubb_generalized_grid_scam_test #1010false1010775SUCCESS1758787508737http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/10/CMakeLists.txt98d8c4814d6669dae533da157c1da701ec53e92c1758753854000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comThe cmake update (#1253) * Initial cmake additions * Reworking dependency handling and toolchains. * Improvements, intel, gfortran, and nvfortran work with and without debugging, and nvfortran works with openacc or openmp too. * Small changes to toolchains * Little various improvements * Making ctests work * Reworking ctests, we have decided to make the ctests run a small number of quick cases/tests just to facilitate development. * Introducing ctests. Big script updates - run_scm.py and compile.py are new * Introducing run_scm.py to replace run_scm.bash and CLUBBStandardsCheck.py to replace CLUBBStandardsCheck.pl. The new CLUBBStandardsCheck.py is much faster and should be even more capable at detecting issues than the perl version, it also has improved readability. The new run_scm.py is way more capable and managable than the bash version, it's used now in the new ctests. Also, the compile script has improved functionality, and ctests can be invoked when adding -run_tests to the compile.py call. Also the whole system supports maintaining builds with multiple compilers now, e.g. using nvfortran then ifx will not overwrite the nvfortran version now, and the new run_scm.py is smart enough to choose the right one depending on the compiler loaded. * Small changes and adding a -legacy flag to run_scm.py to run the bin/clubb_standalone version from the excruciatingly old compilation system. * Small cleanup and Cmake file simplifications - more to come * Mostly cleanup. Moving stuff around to put it in better places, renaming, removing unnecessary parts, etc. * More cleanup and adding comments * Hopefully final cleanup * Small bug fix * Small changes and new jenkins test * Small fix * Now that our source checking scripts return error codes, we have to ignore them in this make file. This doesn't change their output at all, but causes the error code to be ignored * Accidentally removed output dir * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Compile script improvements and new run_thread_test.py to replace bash version. * Changing default ENABLE_OMP to default off, and changing the default executable used by the new python version of the thread test to use the cmake install by default. * Making more general cmake test to test various compile options. * Making ctests output better in compile.py and making new jenkins test a little fancier. * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Reverting some of the fancyness of new test larson-group/clubb#1117 2025-09-24 17:44:14 -050098d8c4814d6669dae533da157c1da701ec53e92cThe cmake update (#1253)addCMakeLists.txtCMakeLists.txt3c612a8f18f66a849f81a234da6df214cd5667201758753854000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comThe cmake update (#1253) * Initial cmake additions * Reworking dependency handling and toolchains. * Improvements, intel, gfortran, and nvfortran work with and without debugging, and nvfortran works with openacc or openmp too. * Small changes to toolchains * Little various improvements * Making ctests work * Reworking ctests, we have decided to make the ctests run a small number of quick cases/tests just to facilitate development. * Introducing ctests. Big script updates - run_scm.py and compile.py are new * Introducing run_scm.py to replace run_scm.bash and CLUBBStandardsCheck.py to replace CLUBBStandardsCheck.pl. The new CLUBBStandardsCheck.py is much faster and should be even more capable at detecting issues than the perl version, it also has improved readability. The new run_scm.py is way more capable and managable than the bash version, it's used now in the new ctests. Also, the compile script has improved functionality, and ctests can be invoked when adding -run_tests to the compile.py call. Also the whole system supports maintaining builds with multiple compilers now, e.g. using nvfortran then ifx will not overwrite the nvfortran version now, and the new run_scm.py is smart enough to choose the right one depending on the compiler loaded. * Small changes and adding a -legacy flag to run_scm.py to run the bin/clubb_standalone version from the excruciatingly old compilation system. * Small cleanup and Cmake file simplifications - more to come * Mostly cleanup. Moving stuff around to put it in better places, renaming, removing unnecessary parts, etc. * More cleanup and adding comments * Hopefully final cleanup * Small bug fix * Small changes and new jenkins test * Small fix * Now that our source checking scripts return error codes, we have to ignore them in this make file. This doesn't change their output at all, but causes the error code to be ignored * Accidentally removed output dir * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Compile script improvements and new run_thread_test.py to replace bash version. * Changing default ENABLE_OMP to default off, and changing the default executable used by the new python version of the thread test to use the cmake install by default. * Making more general cmake test to test various compile options. * Making ctests output better in compile.py and making new jenkins test a little fancier. * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Small fix to jenkins test * Reverting some of the fancyness of new test larson-group/clubb#1117 2025-09-24 17:44:14 -05003c612a8f18f66a849f81a234da6df214cd566720The cmake update (#1253)addCMakeLists.txtgithttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse11http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/11/9http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/9/Started by an SCM change93435d5ba58c9bd61872a20bdde6ee026743287083435d5ba58c9bd61872a20bdde6ee02674328708refs/remotes/origin/clubb_silhs_devel3435d5ba58c9bd61872a20bdde6ee026743287083435d5ba58c9bd61872a20bdde6ee02674328708refs/remotes/origin/clubb_silhs_devel3435d5ba58c9bd61872a20bdde6ee026743287083435d5ba58c9bd61872a20bdde6ee02674328708refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#91226151642148cam_clubb_generalized_grid_scam_test #99false910663SUCCESS1758701108496http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/9/stats_variables.F90advance_clubb_core_module.F90stats_zt_module.F90123a31f755ebf9511f169f5fe1d6312c3b1cdc7a1758642876000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comAdding expldiff budget terms for thlm and rtm (#1252) * Adding expldiff budget terms for thlm and rtm * Removing unintentional line. * Fixing initialization-to-zero problem * Adding some comments. * Adding thlm_ed and rtm_ed to standard_stats.in * Adding thlm_ed and rtm_ed to all_stats.in * Fixing two bugs. Budget balance test passes and results are b4b. This commit is BIT_CHANGING, but only because it adds additional stats fields to the _zt.nc files. Answers would only change if l_do_expldiff_rtm_thlm=T. 2025-09-23 09:54:36 -0600123a31f755ebf9511f169f5fe1d6312c3b1cdc7aAdding expldiff budget terms for thlm and rtm (#1252)editstats_variables.F90editadvance_clubb_core_module.F90editstats_zt_module.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse10http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/10/8http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/8/Started by user Brian GriffingriffinbBrian Griffin81644e80604055cf7a4910e9ace4213ac15f38fbf1644e80604055cf7a4910e9ace4213ac15f38fbfrefs/remotes/origin/clubb_silhs_devel1644e80604055cf7a4910e9ace4213ac15f38fbf1644e80604055cf7a4910e9ace4213ac15f38fbfrefs/remotes/origin/clubb_silhs_devel1644e80604055cf7a4910e9ace4213ac15f38fbf1644e80604055cf7a4910e9ace4213ac15f38fbfrefs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#814921291642148cam_clubb_generalized_grid_scam_test #88false810366SUCCESS1748038020657http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/8/src/physics/cam/clubb_intr.F901644e80604055cf7a4910e9ace4213ac15f38fbf1748037675000http://carson.math.uwm.edu/jenkins/user/bmg2bmg2bmg2@uwm.eduI am officially setting l_ascending_grid to false in CAM-CLUBB, meaning that CLUBB's grid is now officially in descending mode in CAM, matching CAM's grid direction. With -O0 optimization and with CLUBB's l_test_grid_generalization turned on, results are a bit-for-bit match to the previous revision. However, normally CAM-CLUBB doesn't run with these options, which means these results should be BIT_CHANGING. This is probably my final commit to our battery of repositories. Thank you for 23 years, and God bless you! 2025-05-23 17:01:15 -05001644e80604055cf7a4910e9ace4213ac15f38fbfI am officially setting l_ascending_grid to false in CAM-CLUBB, meaningeditsrc/physics/cam/clubb_intr.F90githttp://carson.math.uwm.edu/jenkins/user/bmg2bmg2false9http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/9/7http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/7/Started by an SCM change7294f7921fa437e0b541244f89b96a679d039d4ba294f7921fa437e0b541244f89b96a679d039d4barefs/remotes/origin/clubb_silhs_devel294f7921fa437e0b541244f89b96a679d039d4ba294f7921fa437e0b541244f89b96a679d039d4barefs/remotes/origin/clubb_silhs_devel294f7921fa437e0b541244f89b96a679d039d4ba294f7921fa437e0b541244f89b96a679d039d4barefs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#715147671642148cam_clubb_generalized_grid_scam_test #77false710293SUCCESS1747987507693http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/7/output_netcdf.F90177ee2016e7e4584f9c7c1d420ab3e1e013969601747955296000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comfix precision error of reals 2025-05-22 18:08:16 -0500177ee2016e7e4584f9c7c1d420ab3e1e01396960fix precision error of realseditoutput_netcdf.F90grid_adaptation_module.F90grid_adaptation.F90remapping.F90remapping_module.F900ae10ffa778fe9951dfd3de81a1ffcd38b3e5bdd1747962119000http://carson.math.uwm.edu/jenkins/user/0-freundlich-adenin0-freundlich-adenin0-freundlich-adenin@icloud.comrename grid_adaptation and remapping modules 2025-05-22 20:01:59 -05000ae10ffa778fe9951dfd3de81a1ffcd38b3e5bddrename grid_adaptation and remapping modulesaddremapping_module.F90addgrid_adaptation_module.F90deleteremapping.F90deletegrid_adaptation.F90grid_adaptation_module.F903481c17db470604c208604778051bb11dfa7f9431747963509000http://carson.math.uwm.edu/jenkins/user/0-freundlich-adenin0-freundlich-adenin0-freundlich-adenin@icloud.comadd fix for grid generalization 2025-05-22 20:25:09 -05003481c17db470604c208604778051bb11dfa7f943add fix for grid generalizationeditgrid_adaptation_module.F90githttp://carson.math.uwm.edu/jenkins/user/0-freundlich-adenin0-freundlich-adeninhttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse8http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/8/6http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/6/Started by user Brian GriffingriffinbBrian Griffin6e5ec9afb7da53442bdee3952105faf142b9c3007e5ec9afb7da53442bdee3952105faf142b9c3007refs/remotes/origin/clubb_silhs_devele5ec9afb7da53442bdee3952105faf142b9c3007e5ec9afb7da53442bdee3952105faf142b9c3007refs/remotes/origin/clubb_silhs_devele5ec9afb7da53442bdee3952105faf142b9c3007e5ec9afb7da53442bdee3952105faf142b9c3007refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#615632431642148cam_clubb_generalized_grid_scam_test #66false610268SUCCESS1747957990527http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/6/src/physics/cam/clubb_intr.F90f3e4feee74e7ead32bdc9e6373c1217b7dbb5e3c1747956100000http://carson.math.uwm.edu/jenkins/user/bmg2bmg2bmg2@uwm.eduI've got the magic fix I know if I can fix it once, I can fix twice I fix the baddest bugs ... Changing the pbuf_get_field calls to only go from 1 to ncol in the column dimension for the variables that are newly being passed directly into advance_clubb_core_api thanks to our removal of the _in variables (e.g. wp2_in) in a recent commit. This solves the problem of trying to shove a (pverp-top_lev+1 x pcols) size box into the hole for a (pverp-top_lev+1 x ncol) size box. G-UNIT!!! 2025-05-22 18:21:40 -0500f3e4feee74e7ead32bdc9e6373c1217b7dbb5e3cI've got the magic fixeditsrc/physics/cam/clubb_intr.F90githttp://carson.math.uwm.edu/jenkins/user/bmg2bmg2false7http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/7/5http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/5/Started by an SCM change5151044a3d1ce02bfa4b602cd3f491b2878307dfa151044a3d1ce02bfa4b602cd3f491b2878307dfarefs/remotes/origin/clubb_silhs_devel151044a3d1ce02bfa4b602cd3f491b2878307dfa151044a3d1ce02bfa4b602cd3f491b2878307dfarefs/remotes/origin/clubb_silhs_devel151044a3d1ce02bfa4b602cd3f491b2878307dfa151044a3d1ce02bfa4b602cd3f491b2878307dfarefs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#515308741642148cam_clubb_generalized_grid_scam_test #55false510162SUCCESS1747901108841http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/5/model_flags.F90remapping.F90constants_clubb.F90output_netcdf.F90stats_zm_module.F90interpolation.F90stats_variables.F90stats_clubb_utilities.F90grid_adaptation.F90b558b034dae7a6c45872fe3f91ac6e163472eaaa1747866722000http://carson.math.uwm.edu/jenkins/user/noreplynoreplynoreply@github.comMerge grid adaptation implementation (#1249) * add tunable parameter interp_from_dycore_grid_method * add dycore interpolation to atex case with conservative remapping by ullrich * remove new gpu commands * add acc annotation for gpu * remove acc annotation again * add acc annotation for parallel loop * add acc annotations for parallel loop execution * change remapping subroutine to take number of levels and levels as input instead of whole grid object * add more general functions to interpolate * add and refactor functions for remapping values given on zm levels * fix compiler errors and refactoring to stay under 100 chars per line * remove TODOs and used new interpolating function in interpolate_forcings * add functions for adaptive grid generation and grid density function normalization * add conservative ullrich remapping to cases with forcing time dependent inputs * add grid adaptation and flag * clean up and add warning for case when remapping from dycore flag cannot be used * some refactoring * add logical flag for dycore and renamed flags for dycore and grid adaptation * remove ifdefs * refactoring * add units * changed Lscale to inverse * add write to file for grid adaptation (still WIP) * clean up * change parameter for grid adaptation and add file and plot script for grid adaptation * add script for animation * add e3sm dycore grid * refactoring for efficiency * small fix for grid adapt plot script * add new grid adaptation method with prescribed minimum density profile * fix call to setup_grid * add gabls2 case * add refinement criterion and adaptation trigger * small fix and add minimum grid density profile to animation * add grid density to pyplots * add general refinement criteria for astex and gabls2 case (and arm) * update create_evenly_grid script * unfinished save * WIP: save progress * save working configuration * save working config without Lscale * save working config wo the use of Lscale * change call to map1_ppm to use correct iv parameter and order 4 * save code for results in thesis * add code for results * refactor remapping module for grid adaptation * add list of functions/subroutines to remapping module * adjust use statements * add model files for arm, astex and gabls2 * refactor grid_adaptation module * refactor and adjust tim stopping in clubb_driver * clean up * clean up * add grids * add scripts * add changes made to pyplotgen to generate plots for paper * add configurable model flags file that was used to obtain results * clean up and save tunable parameters file * fix error * fix error * add grid adapt output for debug level 2 * change dycore grid * add plot generation script for paper plots * add small changes * add normalization explaination plot to paper plots script * fix plot * clean up for stats output * add comment * fix use of clubb_at_least_debug_api * small deallocation fix * fix merge problems * revert changes made to test grid adaptation * revert tunable_parameters for Lscale * revert changes made to create plots for grid adaptation * add new line * add new line * revert changes * revert changes made to generate grid adaptation plots 2025-05-21 17:32:02 -0500b558b034dae7a6c45872fe3f91ac6e163472eaaaMerge grid adaptation implementation (#1249)editstats_zm_module.F90editinterpolation.F90editmodel_flags.F90editstats_variables.F90editgrid_adaptation.F90editstats_clubb_utilities.F90editconstants_clubb.F90addremapping.F90editoutput_netcdf.F90githttp://carson.math.uwm.edu/jenkins/user/noreplynoreplyfalse6http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/6/4http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/4/Started by an SCM change4154fe0431dc356aebf5c39550a0b8c8a9b05e597154fe0431dc356aebf5c39550a0b8c8a9b05e597refs/remotes/origin/clubb_silhs_devel154fe0431dc356aebf5c39550a0b8c8a9b05e597154fe0431dc356aebf5c39550a0b8c8a9b05e597refs/remotes/origin/clubb_silhs_devel154fe0431dc356aebf5c39550a0b8c8a9b05e597154fe0431dc356aebf5c39550a0b8c8a9b05e597refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#414904061642148cam_clubb_generalized_grid_scam_test #44false410094SUCCESS1747814709641http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/4/advance_xm_wpxp_module.F90a1ccb74c23ed0ba31458eab3cd2101a82c39dcf31747720486000http://carson.math.uwm.edu/jenkins/user/bmg2bmg2bmg2@uwm.eduzm and zt were mixed up in the error prints argument list. 2025-05-20 00:54:46 -0500a1ccb74c23ed0ba31458eab3cd2101a82c39dcf3zm and zt were mixed up in the error prints argument list.editadvance_xm_wpxp_module.F90githttp://carson.math.uwm.edu/jenkins/user/bmg2bmg2false5http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/5/3http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/3/Started by an SCM change30a7b056e7e3fb54e702de4e1ea97494a9fff85a90a7b056e7e3fb54e702de4e1ea97494a9fff85a9refs/remotes/origin/clubb_silhs_devel0a7b056e7e3fb54e702de4e1ea97494a9fff85a90a7b056e7e3fb54e702de4e1ea97494a9fff85a9refs/remotes/origin/clubb_silhs_devel0a7b056e7e3fb54e702de4e1ea97494a9fff85a90a7b056e7e3fb54e702de4e1ea97494a9fff85a9refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#315065291642148cam_clubb_generalized_grid_scam_test #33false310029SUCCESS1747728310654http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/3/src/physics/cam/clubb_intr.F900a7b056e7e3fb54e702de4e1ea97494a9fff85a91747689630000http://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerhuebler@uwm.eduBug fix to dimensions that was causing crashes. This is technically BIT_CHANGING. 2025-05-19 16:20:30 -05000a7b056e7e3fb54e702de4e1ea97494a9fff85a9Bug fix to dimensions that was causing crashes. This is technically BIT_CHANGING.editsrc/physics/cam/clubb_intr.F90githttp://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerfalse4http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/4/2http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/2/Started by user Brian GriffingriffinbBrian Griffin25afa946afdb36ccab6009d735fd59a243a881f435afa946afdb36ccab6009d735fd59a243a881f43refs/remotes/origin/clubb_silhs_devel5afa946afdb36ccab6009d735fd59a243a881f435afa946afdb36ccab6009d735fd59a243a881f43refs/remotes/origin/clubb_silhs_devel5afa946afdb36ccab6009d735fd59a243a881f435afa946afdb36ccab6009d735fd59a243a881f43refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#214374521642148cam_clubb_generalized_grid_scam_test #22false29942SUCCESS1747412052046http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/2/false3http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/3/1http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/1/Started by an SCM changefalse#14721642148cam_clubb_generalized_grid_scam_test #11false19941FAILURE1747382709197http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/1/false2http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/2/disabledStarted by an SCM changefalse#14721642148cam_clubb_generalized_grid_scam_test #11false19941FAILURE1747382709197http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/1/false2http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/2/Build stability: No recent builds failed.icon-health-80plushealth-80plus.png100falseStarted by an SCM change1532e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1514841891642148cam_clubb_generalized_grid_scam_test #1515false1511277SUCCESS1759392306858http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/15/latin_hypercube_driver_module.F9050f5a9aa2d2a1e4894e9d0aa40f653db463beb151759358185000http://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerhuebler@uwm.eduMaking the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true. 2025-10-01 17:36:25 -050050f5a9aa2d2a1e4894e9d0aa40f653db463beb15Making the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true.editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerfalse14http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/Started by an SCM change1532e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1514841891642148cam_clubb_generalized_grid_scam_test #1515false1511277SUCCESS1759392306858http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/15/latin_hypercube_driver_module.F9050f5a9aa2d2a1e4894e9d0aa40f653db463beb151759358185000http://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerhuebler@uwm.eduMaking the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true. 2025-10-01 17:36:25 -050050f5a9aa2d2a1e4894e9d0aa40f653db463beb15Making the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true.editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerfalse14http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/Started by an SCM changefalse#14721642148cam_clubb_generalized_grid_scam_test #11false19941FAILURE1747382709197http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/1/false2http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/2/Started by an SCM change1532e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1514841891642148cam_clubb_generalized_grid_scam_test #1515false1511277SUCCESS1759392306858http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/15/latin_hypercube_driver_module.F9050f5a9aa2d2a1e4894e9d0aa40f653db463beb151759358185000http://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerhuebler@uwm.eduMaking the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true. 2025-10-01 17:36:25 -050050f5a9aa2d2a1e4894e9d0aa40f653db463beb15Making the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true.editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerfalse14http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/Started by an SCM change1532e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_devel32e7d4adf103e76b7fe05242a69588eb5a6fa29632e7d4adf103e76b7fe05242a69588eb5a6fa296refs/remotes/origin/clubb_silhs_develhttps://github.com/larson-group/cam.gitfalse#1514841891642148cam_clubb_generalized_grid_scam_test #1515false1511277SUCCESS1759392306858http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/15/latin_hypercube_driver_module.F9050f5a9aa2d2a1e4894e9d0aa40f653db463beb151759358185000http://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerhuebler@uwm.eduMaking the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true. 2025-10-01 17:36:25 -050050f5a9aa2d2a1e4894e9d0aa40f653db463beb15Making the l_random_k_lh_start set the random generator seed before generating a random. This should fix the generalized grid test, which failed for the only flag set that turned on l_random_k_lh_start. Before, the test passed because we used the same random seed for all timesteps, preventing differences in the random call. But now we use a different seed each timestep, causing differences. To ensure reproducibility, we use the seed each timestep to generate rand_pool, which is an effective solution, but we need to do that for the rand_integer_in_range call in the l_random_k_lh_start code for the same reason. Since this does change the random that comes out, this is BIT_CHANGING, but only for runs using both SILHS and l_random_k_lh_start=.true.editlatin_hypercube_driver_module.F90githttp://carson.math.uwm.edu/jenkins/user/hueblerGunther Hueblerfalse14http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/14/Started by an SCM changefalse#14721642148cam_clubb_generalized_grid_scam_test #11false19941FAILURE1747382709197http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/1/false2http://carson.math.uwm.edu/jenkins/view/all/job/cam_clubb_generalized_grid_scam_test/2/16truetruefalsefalse