Closed penguian closed 1 year ago
@inh599@nci.org.au changed milestone from 6. Report
to 3. Implementation
@inh599@nci.org.au commented
branch@ branches/Users/inh599/Ticket337
fix applied via the argument list method - this restores the old code structure/intent but is less efficient.
limited, single site, testing so far indicates that this branch successfully restores the functionality of the trunk @8873 - running benchcab for further analysis.
@inh599@nci.org.au changed milestone from 3. Implementation
to 2. Submission
@inh599@nci.org.au commented
benchcab shows that this branch bitwise reproduces the trunk prior to the restructure at all single sites (170) considered, across 16 variables (energy fluxes, carbon fluxes, temperature and humidity diagnostics, soil temperature and moisture) and four science configurations
No spatial run testing has been undertaken - but this this is not expected to cause an issue.
Pass to Jhan for consideration/implementation in trunk.
@ccc561@nci.org.au changed status from assigned
to closed
@ccc561@nci.org.au set resolution to fixed
@ccc561@nci.org.au changed milestone from 2. Submission
to 1. Closed
@ccc561@nci.org.au commented
Close this, since #359 implements a fix.
@ccc561@nci.org.au changed keywords from screen level temperature
to screen level temperature nogit
keyword_screen_level_temperature_nogit
owner:jxs599@nci.org.au
resolution_fixed
type_defect
| by inh599The code re-organisation @8873 has inadvertently broken the calculation of the screen-level temperature (canopy%tscrn) and then the screen-level humidity (canopy%qscrn). This is because the looping index iterplus is no longer set within cable_canopy and this is used to evaluate terms contributing to %tscrn.
There are two possible solutions:
The second option has the added advantage of a) removing the need for iterplus and b) (eventually) we could simplify the code so that %zetar and %zetarsh are single dimension (mp)-arrays.
Also - argument canopy_zetar and canopy_zetarsh in cbl_zetar() need to be INOUT variables. The trunk should/will fail rigorous compilation as it currently stands.
Issue migrated from trac:337 at 2023-11-27 11:42:09 +1100