Open penguian opened 3 years ago
@jxs599@nci.org.au changed owner from jxs599
to somebody
@jxs599@nci.org.au changed type from defect
to model improvement
@jxs599@nci.org.au changed component from JAC readiness
to model
@jxs599@nci.org.au edited the issue description
@jxs599@nci.org.au set milestone to 6. Report
@ccc561@nci.org.au set keywords to maygit
keyword_maygit
type_model improvement
| by srb001@csiro.aucable_constants.F90: cable_data.F90: : cable_math_constants_mod.F90 : cable_other_constants_mod.F90 : cable_photo_constants_mod.F90 : cable_phys_constants_mod.F90 : cable_params_mod.F90
This has come full circle. We are now going to have a seperate file and module for each constants type. These will then moer easily be promoted to runtime definable quantities.
There were 3 places where CABLE was getting its constants (for e.g. PI=3.1412.....) from. Following ACCESS-1.3, as a proof of concept that each called function could be embedded with informmation of what data that it should go to the "library" and collect, we initiated this with process with constants. As a self contained initiative a file called cable_data was born to contain the constants and also the functions to gather up what was to be dispatched when needed. Shortly thereafter we were sidetracked. SLI re-introduced the concept of USEing constants through modules again however these modules were all bundled into a single file cable_constants.
So now we intend to have all the constants coming from only one source.
Issue migrated from trac:268 at 2023-11-27 11:34:36 +1100