Up to Specific issues in real coupled models
Hello,
I'm using OASIS3-MCT to couple the GRIMs atmospheric model with NEMO (v3.6), and encountering two issues.
1. I set $RUNTIME to match the model integration period (e.g., $RUNTIME 432000 for a 5-day integration from the initial state). However, the model abnormally stops at the end of the time step with the error message: "model time must be strictly smaller than namcouple $RUNTIME..."
Increasing $RUNTIME to something like 432001 resolves this issue, and the model performs well, stopping normally.
2. Another issue arises when restarting the model using the restart files from the previous 5-day experiment. The model abnormally stops near the end of integration with the following debug file output. Notably, the time step jumps from 428400 to 864000.
I would appreciate any help you can provide.
BK Moon
-------- NOTE (oasis_advance_run) compute field index and sizes
(oasis_advance_run) DEBUG nfav,nsav,nsa = 1 798 798
-------- NOTE (oasis_advance_run) comm_now compute
(oasis_advance_run) at 428400 428400 SKIP: O_dQnsdT
(oasis_advance_run) at 428400 428400 KINF: O_dQnsdT 0
------ EXIT (oasis_advance_run)
---- EXIT (oasis_get_worker)
-- EXIT (oasis_get_r28)
-- ENTER (oasis_put_r28)
---- ENTER (oasis_put_worker)
------ ENTER (oasis_advance_run)
-------- ENTER (oasis_coupler_bldvarname)
(oasis_coupler_bldvarname) check vname 1 O_SSTSST
-------- EXIT (oasis_coupler_bldvarname)
(oasis_advance_run) DEBUG vname 2 1 O_SSTSST
-------- NOTE (oasis_advance_run) loop over var ncpl
(oasis_advance_run) lreadrest : F arrayon = T F F F F
-------- NOTE (oasis_advance_run) set nx and ny
(oasis_advance_run) DEBUG nx, ny = 182 149
-------- NOTE (oasis_advance_run) check for lag restart
-------- NOTE (oasis_advance_run) set mseclag
(oasis_advance_run) DEBUG msec,mseclag = 864000 864000
(oasis_advance_run) ERROR: at 864000 864000 for var = O_SSTSST
(oasis_advance_run) ERROR:
model time must be strictly smaller than namcouple $RUNTIME = 864000
432001
(oasis_abort) ABORT: file = mod_oasis_advance.F90
(oasis_abort) ABORT: line = 676
(oasis_abort) ABORT: on model = oceanx
(oasis_abort) ABORT: on global rank = 0
(oasis_abort) ABORT: on local rank = 0
(oasis_abort) ABORT: CALLING ABORT FROM OASIS LAYER NOW
Hello, The model time is not decided by oasis but by the "date" argument (i.e. the second argument) provided to the oasis_put call in your model. I think you should check how the oasis_put are called in your code to understand what is happening. Regards, Sophie
Thanks for your comments. I've fixed the problem and really appreciate your help.