search

NGEET / fates

repository for the Functionally Assembled Terrestrial Ecosystem Simulator (FATES)
Other
99 stars 92 forks source link

Restart error for FATES hydro (wierd values in v_aroot_layer) #878

Open mariuslam opened 2 years ago

mariuslam commented 2 years ago

Hi,

I would like to report a bug and a fix for it when restarting with FATES-HYDRO. @huitang-earth has found that moving some part of the code higher up in clmfates_interface worked.

Move code: https://github.com/ESCOMP/CTSM/blob/19c57a73099f0712622a7b07170f1a3c648f31aa/src/utils/clmfates_interfaceMod.F90#L1557-L1586 to the front of https://github.com/ESCOMP/CTSM/blob/19c57a73099f0712622a7b07170f1a3c648f31aa/src/utils/clmfates_interfaceMod.F90#L1515

version: ctsm5.1.dev092 and fates-sci.1.57.0_api.23.0.0 Regards,

Marius

glemieux commented 2 years ago

Thanks @mariuslam and @huitang-earth. Would you be able to provide some information about the error you are seeing?

mariuslam commented 2 years ago

Here you will find the log files: logs.zip

huitang-earth commented 2 years ago

Hi @glemieux,

I think @mariuslam shared the log files when the model is in debug mode and also with a lot of the prints. The line 3225-3227 in the "lnd.log." file shared by @mariuslam starts to show the problem. This is because the initialization of "v_aroot_layer" has not been done properly before FATES start to fuse the cohort when calling "ed_update_site" in https://github.com/ESCOMP/CTSM/blob/19c57a73099f0712622a7b07170f1a3c648f31aa/src/utils/clmfates_interfaceMod.F90#L1524

If you want to the see the original error when we ran into this issue, here it is log_original.zip

Line 3321 in "lnd.log.5699783.220521-171227" indicates the error: "Site plant water balance does not close"