Initialise ssnow variables in mpiworker

[SeanBryan51]

Variables ssnow%rtevap_sat, ssnow%rtevap_unsat and ssnow%qrecharge are uninitialised in all MPI worker processes and crashes the model when runtime memory checks are enabled (see #397 for more details). These variables are specifically initialised (to zero) in cable_parameters.F90 for the serial case (and for the MPI master process). Rather than communicating the initialised variables from master to worker, this change initialises the variables in the worker directly.

Although we plan to deprecate the MPI drivers in the future, this change will still be useful when comparing the behaviour of MPI vs serial during development.

Fixes #397

Type of change

Please delete options that are not relevant.

• Bug fix

Checklist

• I have checked my code/text and corrected any misspellings

---

📚 Documentation preview 📚: https://cable--471.org.readthedocs.build/en/471/

[SeanBryan51]

Note to reproduce the error, the following temporary patch was used to "fix" #396:

diff --git a/src/offline/cable_parameters.F90 b/src/offline/cable_parameters.F90
index 8117f3f..f9c242e 100644
--- a/src/offline/cable_parameters.F90
+++ b/src/offline/cable_parameters.F90
@@ -2294,22 +2294,22 @@ CONTAINS
   endwhere
 
   ! ____________________ MMY comment out as we don't use hys ___________________________
-  do k=1,ms
-     do i=1,mp
-        if (ssnow%wb_hys(i,k) .lt. 0._r_2) then
-           ssnow%wb_hys(i,k) = ssnow%wb(i,k)
-        end if
-        ssnow%wb_hys(i,k)  = max(soil%watr(i,k) ,min(soil%ssat_vec(i,k), ssnow%wb_hys(i,k)))
-
-       if (ssnow%smp_hys(i,k) .lt. -1.0e+30_r_2) then  !set to missing, calc
-          ssnow%smp_hys(i,k) = -soil%sucs_vec(i,k)*  &
-                               ( (ssnow%wb_hys(i,k)-ssnow%watr_hys(i,k))/&
-                                 (ssnow%ssat_hys(i,k)-ssnow%watr_hys(i,k)) )**&
-                                (-1._r_2/soil%bch_vec(i,k) )
-       end if
-       ssnow%smp_hys(i,k) = max(-1.0e10,min(-soil%sucs_vec(i,k),ssnow%smp_hys(i,k) ))
-    end do
-  end do
+!   do k=1,ms
+!      do i=1,mp
+!         if (ssnow%wb_hys(i,k) .lt. 0._r_2) then
+!            ssnow%wb_hys(i,k) = ssnow%wb(i,k)
+!         end if
+!         ssnow%wb_hys(i,k)  = max(soil%watr(i,k) ,min(soil%ssat_vec(i,k), ssnow%wb_hys(i,k)))
+
+!        if (ssnow%smp_hys(i,k) .lt. -1.0e+30_r_2) then  !set to missing, calc
+!           ssnow%smp_hys(i,k) = -soil%sucs_vec(i,k)*  &
+!                                ( (ssnow%wb_hys(i,k)-ssnow%watr_hys(i,k))/&
+!                                  (ssnow%ssat_hys(i,k)-ssnow%watr_hys(i,k)) )**&
+!                                 (-1._r_2/soil%bch_vec(i,k) )
+!        end if
+!        ssnow%smp_hys(i,k) = max(-1.0e10,min(-soil%sucs_vec(i,k),ssnow%smp_hys(i,k) ))
+!     end do
+!   end do
 
  ! MMY note that after commenting out, wb_hys and smp_hys are not defined any more
 

[SeanBryan51]

Benchcab regression tests pass for both site and spatial runs:

2025-02-11 17:25:58,521 - INFO - benchcab.benchcab.py:380 - Running comparison tasks...
2025-02-11 17:25:58,544 - INFO - benchcab.benchcab.py:381 - tasks: 168 (models: 2, sites: 42, science configurations: 4)
2025-02-11 17:29:02,317 - INFO - benchcab.benchcab.py:391 - 0 failed, 168 passed

config.yaml:

realisations:
  - repo:
      git:
        branch: main
  - repo:
      git:
        branch: 397-uninitialised-ssnowrtevap_sat-and-ssnowrtevap_unsat-in-worker-processes

modules: [
  intel-compiler/2021.1.1,
  netcdf/4.7.4,
  openmpi/4.1.0
]