Closed ekluzek closed 8 months ago
As suspected when I remove the output about negative flow the test works fine:
diff --git a/src/riverroute/MOSART_physics_mod.F90 b/src/riverroute/MOSART_physics_mod.F90
index a2d327f..66ea867 100644
--- a/src/riverroute/MOSART_physics_mod.F90
+++ b/src/riverroute/MOSART_physics_mod.F90
@@ -324,10 +324,10 @@ subroutine Routing_KW(iunit, nt, theDeltaT)
if((TRunoff%wr(iunit,nt)/theDeltaT &
+ TRunoff%dwr(iunit,nt)) < -TINYVALUE) then
- write(iulog,*) 'mosart: ERROR main channel going negative: ', iunit, nt
- write(iulog,*) theDeltaT, TRunoff%wr(iunit,nt), &
- TRunoff%wr(iunit,nt)/theDeltaT, TRunoff%dwr(iunit,nt), temp_gwl
- write(iulog,*) ' '
+ !write(iulog,*) 'mosart: ERROR main channel going negative: ', iunit, nt
+ !write(iulog,*) theDeltaT, TRunoff%wr(iunit,nt), &
+ ! TRunoff%wr(iunit,nt)/theDeltaT, TRunoff%dwr(iunit,nt), temp_gwl
+ !write(iulog,*) ' '
! call shr_sys_abort('mosart: ERROR main channel going negative')
endif
The test
ERI_D.f10_f10_mg37.I1850Clm50Sp.derecho_intel.mosart-nobypass
fails to run on Derecho in mosart1_0_48 because it times out after the overly generous 3:40. I think the problem may be because there is so much output reporting on negative runoff.