A review of the the persistence and bowl computation is needed.
Running April 20th version on Eric's ciclad account (old CDAT) and drive_IPSL.py 1 year test leaves a few places with non filled points on bowl (zero value for T,S, sigma, depth,... see for instance a ncview of ptopxy). Need to investigate why.
Running the same version with the new CDAT8.1 installation on Yona's account (but for the new call to ESMP) generates additionnal issues of a different sort (masked points on ptop<>xy in the middle of the ocean). A change had to be made to remove a (integer minus boolean) statement, apparently not accepted anymore by CDAT8.1.
A review of the the persistence and bowl computation is needed.
Running April 20th version on Eric's ciclad account (old CDAT) and drive_IPSL.py 1 year test leaves a few places with non filled points on bowl (zero value for T,S, sigma, depth,... see for instance a ncview of ptopxy). Need to investigate why.
Running the same version with the new CDAT8.1 installation on Yona's account (but for the new call to ESMP) generates additionnal issues of a different sort (masked points on ptop<>xy in the middle of the ocean). A change had to be made to remove a (integer minus boolean) statement, apparently not accepted anymore by CDAT8.1.