Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

good way to get familiar with WRF code

I am running wrf with different number of MPI processes. The resulting model has slightly different statistics. what function could cause numeric instability?

Regards,
 
compile wrf 4.2.2 fail with double precision after (configure -r8)

The first fail is (please help) -

module_ra_goddard.f90
module_ra_goddard.f90:1781:128:

1781 | flx_out=flx, flxd_out=flxd,flxu_out=flxu, flxd_surf = flxd_surf, lmask=lmask, irestrict=min(CHUNK,ite-ii+1) )
| 1
Error: Type mismatch in argument ‘fcld’ at (1); passed REAL(4) to REAL(8)
module_ra_goddard.f90:1973:98:

1973 | flx_out=flx, acflxd_out=flxd, acflxu_out=flxu, irestrict=min(CHUNK,ite-ii+1) )
| 1
Error: Type mismatch in argument ‘tb’ at (1); passed REAL(4) to REAL(8)
Command exited with non-zero status 1
0.15user 0.03system 0:00.18elapsed 100%CPU (0avgtext+0avgdata 51476maxresident)k
0inputs+8outputs (0major+9815minor)pagefaults 0swaps
../configure.wrf:344: recipe for target 'module_ra_goddard.o' failed
 
Note that WRF outputs could be slightly different if you compile the code with different compilers and/or run with different number of processors. Such difference is acceptable. Please don't worry for it.
 
There is no quantitative metrics to describe how large the difference could be. It depends on many factors (number of processors, integration period, physics and dynamic options, etc.)
 
Top