Opened 4 years ago
Last modified 4 years ago
#68 assigned defect
Enable lslin after and before zrecast
| Reported by: | abierjon | Owned by: | abierjon |
|---|---|---|---|
| Priority: | major | Milestone: | |
| Component: | MARS GCM | Version: | |
| Keywords: | Cc: |
Description
To be used after zrecast, lslin must take into account the missing values generated by zrecast when performing its time interpolations and binning (look zrecast or simu_MCS for inspiration)
To be used before zrecast, lslin must rewrite in the output file the fields aps, bps and phisinit.
Change History (4)
comment:1 Changed 4 years ago by
| Resolution: | → fixed |
|---|---|
| Status: | assigned → closed |
comment:2 Changed 4 years ago by
| Resolution: | fixed |
|---|---|
| Status: | closed → reopened |
Actually, only PARTIALLY solved by r2432 , when the user wants to average the values into Ls bins.
When the user wants to interpolate at exact Ls values, the interpolation routine in lslin.F90 doesn't deal correctly with missing_values in the input data.
+ the boolean-ish string "average" is not initialized when answer='y' at the question "Automatic Ls timetsep (y/n)?"
+ it would be good to clean lslin.F90 code a bit
comment:3 Changed 4 years ago by
| Status: | reopened → assigned |
|---|
comment:4 Changed 4 years ago by
NB : the problem occurs when the missing_value in the input file is 1e20, but not when it is -9.99e33. It would be good to find why...
actually, already solved by r2432