Hello,
I'm trying to continue with a previous computation of our mixed sediment model but it seems not successful. So I checked code and found that
When the keyword COMPUTATION CONTINUED is set to YES in SISYPHE, the CONC will not be initialized. Because SISYPHE firstly calls INIT_SEDIMENT, then INIT_SEDIMENT calls INIT_MIXTE to deal with the mixed sediment. But in the subroutine INIT_MIXTE, there is a condition which is
IF(.NOT.DEBU) THEN
!
CALL INIT_COMPO_COH(ES,CONC_VASE,CONC,NPOIN,
* NOMBLAY,NSICLA,AVAIL,AVA0)
...
ELSE
!Check that sum of layers (simple precision) is equal to ZF-ZR
...
If DEBU is true, the program skip INIT_COMPO_COH and goes to "ELSE" block. The problem is that CONC is only initialized in INIT_COMPO_COH which is bypassed.
So as far as I got from my tests, CONC(I,J) remains zero everywhere in continued computation and it causes problems in some subroutines. For example, in SUSPENSION_EVOL,
ZFCL_S%R(I)=QFLUX%R(I)/CONC(I,1)
It will give NaN problems.
At this moment, I modified the subroutine INIT_MIXTE by adding a line
after line 177. It works in my case.
Kind regards,
Qilong