Adding WRF-Hydro PRs by scrasmussen · Pull Request #2263 · wrf-model/WRF

scrasmussen · 2025-11-26T20:20:01Z

TYPE: bugfix, enhancement

KEYWORDS: syncing WRF-Hydro/hydro, gage-assisted diversion, bugfixes

SOURCE: internal hydro team + contributor

Soren Rasmussen, NSF NCAR
Ryan Cabell, NSF NCAR
Andrew Song, Inner Mongolia Meteorological Service / China Meteorological Administration (Wrote PR#829)

DESCRIPTION OF CHANGES: This PR updates the hydro directory to the NCAR/wrf_hydro_nwm_public repository. It includes the following PRs

LIST OF MODIFIED FILES:

$ git diff --name-status develop
M       hydro/CMakeLists.txt
M       hydro/CPL/WRF_cpl/CMakeLists.txt
M       hydro/Data_Rec/module_namelist.F90
M       hydro/Data_Rec/module_namelist_inc.F90
M       hydro/HYDRO_drv/module_HYDRO_drv.F90
M       hydro/Makefile
M       hydro/OrchestratorLayer/config.F90
M       hydro/Routing/CMakeLists.txt
A       hydro/Routing/Diversions/CMakeLists.txt
A       hydro/Routing/Diversions/Makefile
A       hydro/Routing/Diversions/module_diversions.F90
A       hydro/Routing/Diversions/module_diversions_timeslice.F90
M       hydro/Routing/Makefile
M       hydro/Routing/module_channel_routing.F90
M       hydro/arc/Makefile.mpp
D       hydro/arc/Makefile.seq
M       hydro/arc/macros.mpp.cray_fortran
M       hydro/arc/macros.mpp.gfort
M       hydro/arc/macros.mpp.ifort
D       hydro/arc/macros.mpp.ifort.luna
D       hydro/arc/macros.mpp.ifort.summit_has
D       hydro/arc/macros.mpp.ifort.theia
M       hydro/arc/macros.mpp.linux
D       hydro/arc/macros.seq.IBM.xlf90_r
D       hydro/arc/macros.seq.gfort
D       hydro/arc/macros.seq.ifort
D       hydro/arc/macros.seq.linux
D       hydro/arc/macros.theia.debug
D       hydro/arc/macros.yellowstone
D       hydro/arc/macros.yellowstone.debug
M       hydro/configure
M       hydro/nudging/CMakeLists.txt
A       hydro/nudging/io/CMakeLists.txt
R098    hydro/nudging/module_nudging_io.F90     hydro/nudging/io/module_nudging_io.F90
M       hydro/utils/CMakeLists.txt
A       hydro/utils/fortglob/CMakeLists.txt
A       hydro/utils/fortglob/Makefile
A       hydro/utils/fortglob/fortglob.F90
A       hydro/utils/fortglob/libfortglob.c
A       hydro/utils/fortglob/libfortglob.h

TESTS CONDUCTED: NOTE THAT THE PR IS IN DRAFT UNTIL TESTING IS COMPLETE. Current building and running

WRF-Hydro Front Range, CO testcase
Testing multiple configurations of Make and CMake build systems

RELEASE NOTE: Hydro bug fixes plus modifications to add initial support for gage-assisted diversions in channel routing, which requires a new optional Diversion netCDF parameter file. This adds a C compiler dependency

- NCAR/wrf_hydro_nwm_public#756 Add initial support for gage-assisted diversions in channel routing - NCAR/wrf_hydro_nwm_public#811 bugfix: Diversions Makefiles for configure build - NCAR/wrf_hydro_nwm_public#816 Revert TYPEL(k) check - NCAR/wrf_hydro_nwm_public#823 make clean new directories - NCAR/wrf_hydro_nwm_public#829 Rename fortglob.f90 to fortglob.F90 for preprocessor - NCAR/wrf_hydro_nwm_public#827 Dropping .exe extension from wrf_hydro binary - NCAR/wrf_hydro_nwm_public#855 Properly handle unexpected NaNs in diversion observations - NCAR/wrf_hydro_nwm_public#856 Make diversion init errors non-fatal, proceed without diversions - NCAR/wrf_hydro_nwm_public#861 Nudging, Fixing type mismatch when nudging turned on - NCAR/wrf_hydro_nwm_public#862 OUTPUT_CHAN_CONN uses nudging io but not nudging Co-authored-by: Ryan Cabell <rcabell@users.noreply.github.com> Co-authored-by: Andrew Song <andrewsoong@users.noreply.github.com>

weiwangncar · 2025-11-27T01:48:17Z

The regression test results:

Test Type              | Expected  | Received |  Failed
= = = = = = = = = = = = = = = = = = = = = = = =  = = = =
Number of Tests        : 23           24
Number of Builds       : 60           57
Number of Simulations  : 158           150        0
Number of Comparisons  : 95           86        0

Failed Simulations are: 
None
Which comparisons are not bit-for-bit: 
None

weiwangncar · 2025-12-17T06:32:16Z

@scrasmussen Is it possible to find another hydro person to review this PR?

scrasmussen · 2025-12-17T15:29:08Z

@scrasmussen Is it possible to find another hydro person to review this PR?

Sure thing, thanks for the reminder! Just pinged @rcabell about this

… configure: ifx/openapi

islas · 2026-01-28T22:44:18Z

hydro/Routing/Diversions/CMakeLists.txt

@@ -0,0 +1,10 @@
+add_library(hydro_routing_diversions STATIC
+    module_diversions.F90
+    module_diversions_timeslice.F90


I think this might be missing some compile / link info for netcdf. When I try to compile hydro from the top-level WRF build I get:

[ 42%] Building Fortran object hydro/Routing/Diversions/CMakeFiles/hydro_routing_diversions.dir/module_diversions_timeslice.F90.o /home/aislas/wrf-model/wrf_extra/hydro/Routing/Diversions/module_diversions_timeslice.F90:5:9: 5 | use netcdf | 1 Fatal Error: Cannot open module file ‘netcdf.mod’ for reading at (1): No such file or directory compilation terminated. make[2]: *** [hydro/Routing/Diversions/CMakeFiles/hydro_routing_diversions.dir/build.make:88: hydro/Routing/Diversions/CMakeFiles/hydro_routing_diversions.dir/module_diversions_timeslice.F90.o] Error 1 make[1]: *** [CMakeFiles/Makefile2:2078: hydro/Routing/Diversions/CMakeFiles/hydro_routing_diversions.dir/all] Error 2 make[1]: *** Waiting for unfinished jobs.... [ 42%] Linking Fortran static library libhydro_routing_reservoirs.a [ 42%] Built target hydro_routing_reservoirs [ 42%] Building Fortran object external/io_netcdf/CMakeFiles/io_netcdf.dir/field_routines.F90.o [ 42%] Linking Fortran static library libio_netcdf.a [ 42%] Built target io_netcdf [ 42%] Built target registry_code make: *** [Makefile:136: all] Error 2

I'll look into this, thanks for finding this issue!

@islas I was not able to reproduce this error. What environment/setup are you using? How did you build (CMake or Makefile configure)?

I've been using

module purge module load ncarenv gcc ncarcompilers cray-mpich craype netcdf cmake

That's from following the Old Method compile instructions in the readthedocs, without the number two step of replacing the hydro directory with WRF-Hydro repo.

I'm definitely interested in reproducing the bug so I can fix it, thanks!

I'm building locally on my machine where I've installed mpich, netcdf, and netcdf-fortran to directories, making sure that they are found by the cmake build. I usually use this setup to test for anything that derecho hides.

This is the verbose compile output:

[ 6%] Building Fortran object hydro/Routing/Diversions/CMakeFiles/hydro_routing_diversions.dir/module_diversions_timeslice.F90.o cd /home/aislas/wrf-model/wrf_extra/_build/hydro/Routing/Diversions && /usr/bin/gfortran -DBUILD_RRTMG_FAST=0 -DBUILD_RRTMK=0 -DBUILD_SBM_FAST=1 -DCONFIG_BUF_LEN=65536 -DDA_CORE=0 -DDM_PARALLEL -DDWORDSIZE=8 -DEM_CORE=1 -DFSEEKO64_OK -DGRIB1 -DINTIO -DIWORDSIZE=4 -DKEEP_INT_AROUND -DLANDREAD_STUB -DLIMIT_ARGS -DLWORDSIZE=4 -DMAX_DOMAINS_F=21 -DMAX_HISTORY=25 -DMPP_LAND -DNETCDF -DNMM_CORE=0 -DNMM_MAX_DIM=2600 -DNONSTANDARD_SYSTEM_SUBR -DRWORDSIZE=4 -DSHOW_ALL_VARS_USED=0 -DUSE_ALLOCATABLES -DUSE_MPI=1 -DUSE_NETCDF4_FEATURES=1 -DWRFIO_NCD_LARGE_FILE_SUPPORT=1 -DWRFPLUS=0 -DWRF_HYDRO=1 -DWRF_USE_CLM -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE=1 -Dwrfmodel -w -fconvert=big-endian -frecord-marker=4 -J../../mods -O2 -ftree-vectorize -funroll-loops -ffree-line-length-none -fallow-argument-mismatch -fallow-invalid-boz -cpp -c /home/aislas/wrf-model/wrf_extra/hydro/Routing/Diversions/module_diversions_timeslice.F90 -o CMakeFiles/hydro_routing_diversions.dir/module_diversions_timeslice.F90.o /home/aislas/wrf-model/wrf_extra/hydro/Routing/Diversions/module_diversions_timeslice.F90:5:9: 5 | use netcdf | 1 Fatal Error: Cannot open module file ‘netcdf.mod’ for reading at (1): No such file or directory compilation terminated.

I am able to recreate it on derecho, however, by doing:

module purge module load gcc cmake netcdf cray-mpich ./configure_new <stuff, prompt select hydro> ./compile_new

I do have to fix finding CURL in the cmake build... if you do run into that just wholly comment that find_package( CURL ) line in the top-level CMakeLists.txt

Here is my WRF build on derecho, if that helps: /glade/work/aislas/wrf-model/wrf_upstream

module purge module load gcc cmake netcdf cray-mpich ./configure_new <stuff, prompt select hydro> ./compile_new

I do have to fix finding CURL in the cmake build... if you do run into that just wholly comment that find_package( CURL ) line in the top-level CMakeLists.txt

Just an fyi if I also load ncarcompilers in addition to the ones you use, I don't get the CURL package error.

ml purge ml gcc cmake netcdf cray-mpich ncarcompilers

Still working on replicating the error, will chat with you in the release meeting soon!

scrasmussen requested a review from a team as a code owner November 26, 2025 20:20

scrasmussen marked this pull request as draft November 26, 2025 20:20

dudhia previously approved these changes Dec 16, 2025

View reviewed changes

scrasmussen requested a review from rcabell December 17, 2025 15:28

WRF-Hydro PR NCAR/wrf_hydro_nwm_public#859 Add new Intel compilers to…

defcbf2

… configure: ifx/openapi

scrasmussen dismissed dudhia’s stale review via defcbf2 December 24, 2025 17:58

scrasmussen marked this pull request as ready for review December 24, 2025 17:58

rcabell approved these changes Jan 7, 2026

View reviewed changes

scrasmussen mentioned this pull request Jan 20, 2026

doc: coupled WRF + WRF-Hydro Intel's ifx update NCAR/wrf_hydro_nwm_public#876

Merged

weiwangncar approved these changes Jan 28, 2026

View reviewed changes

islas reviewed Jan 28, 2026

View reviewed changes

islas merged commit f2c697e into wrf-model:develop Feb 12, 2026
2 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Adding WRF-Hydro PRs#2263

Adding WRF-Hydro PRs#2263
islas merged 2 commits intowrf-model:developfrom
scrasmussen:develop-v4.8-hydro

scrasmussen commented Nov 26, 2025

Uh oh!

weiwangncar commented Nov 27, 2025

Uh oh!

weiwangncar commented Dec 17, 2025

Uh oh!

scrasmussen commented Dec 17, 2025

Uh oh!

islas Jan 28, 2026

Uh oh!

scrasmussen Jan 28, 2026

Uh oh!

scrasmussen Jan 29, 2026

Uh oh!

islas Jan 29, 2026

Uh oh!

islas Jan 29, 2026

Uh oh!

scrasmussen Feb 4, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

scrasmussen commented Nov 26, 2025

Uh oh!

weiwangncar commented Nov 27, 2025

Uh oh!

weiwangncar commented Dec 17, 2025

Uh oh!

scrasmussen commented Dec 17, 2025

Uh oh!

islas Jan 28, 2026

Choose a reason for hiding this comment

Uh oh!

scrasmussen Jan 28, 2026

Choose a reason for hiding this comment

Uh oh!

scrasmussen Jan 29, 2026

Choose a reason for hiding this comment

Uh oh!

islas Jan 29, 2026

Choose a reason for hiding this comment

Uh oh!

islas Jan 29, 2026

Choose a reason for hiding this comment

Uh oh!

scrasmussen Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants