Ou Wang
Ou Wang
@jm-c Thanks for making the changes. Looks good to me. Please push the changes.
@jm-c The cause of what you experienced is a combination of a bug in ecco_phys.F of this PR and some non-ideal name lists for the new total volume cost. First,...
@jm-c Thanks. LGTM.
> @owang01 I cannot run your forward experiment for two reasons > > 1. my compiler requires that the macros do not contain white spaces, in stir_init_fixed.F: > > >...
> @owang01 for easier reviewing it would be nice if you could remove all trailing white spaces in the new files (if possible). (with something like `sed -i 's/[ \t]*$//'...
@mjlosch Thanks. I don't get warnings or errors with or without "-devel": > link mitgcmuv from dir ../build > STOP NORMAL END My testreport build uses gfortran and the optfile...
I will take a look at why pLoc (or R_Shelfice) is zero. Those pieces of code are inside the IF block `IF (K .GT. 0) THEN`; supposedly, R_Shelfice is larger...
This is still a mystery. Before calling STIC_SOLVE4FLUXES, I inserted a check of pLoc to force the run to stop if pLoc is smaller than 1e-3. The run didn't trigger...
@mjlosch I can reproduce the error on a Mac. I am able to get rid of the error by replacing `shiTransCoeffT(CURI,CURJ,K,bi,bj)` with `gammaT` etc. as an argument in calling `STIC_SOLVE4FLUXES`,...
@mjlosch Thanks. The odd behavior is probably related to the fact that the sizes of variable shelficeForcingT/S and the common block STIC_FIELDS_RL (in STIC.h) are incorrect, if a subroutine, e.g.,...