Outstanding issues yet to be resolved

From Geos-chem
Jump to: navigation, search


NOTE: This page is now obsolete. Please see our list of Currently unresolved issues in GEOS-Chem for the most up-to-date listing of issues that have not yet been fixed in the current version of GEOS-Chem.

On this page we list some issues that have not been resolved in the most recent version of GEOS-Chem.

Order of tracers in ND36 diagnostic

Dylan Millet (dbm@umn.edu) wrote:

I think there's a bug in the anthro emissions diagnostic (nd36). Long story short, not all tracers are getting saved out that should be.
Here's the sequence as I understand it:
  • TMAX(36) gets set to 10 since NEMANTHRO is 10.
  • TINDEX(36,1:10) then gets set to the values 1:10 (line 2769) -- not sure why.
  • in the nd36 area, the main M loop is over 1:10 - i.e., 1:TMAX(36).
  • N then gets set as TINDEX(36,M). But from input_mod.f, this is just 1:10.
  • Line 2165, subloop over NMAX=1:NEMANTHRO (i.e. 1:10)
    • here we check whether N = IDEMS(NMAX)
    • IDEMS(1:10) are the input.geos tracer #'s for the 10 anthropogenic compounds to be included.
  • Several of these are > 10, but N is only going up to 10 so these get skipped.

Philippe Le Sager (plesager@seas.harvard.edu) wrote:

In input_mod.f, TINDEX is not set to 1:10, but to something like (1 4 5 9 10 18 19 21), as you can see from the log file from any run that uses ND36:
   Diag    L   Tracers being saved to disk
   ND36    2   1  4  5  9 10 18 19 21
   ND45   47   1 - 54
   ND70    1   1 -  0
The trick is in the IF."all".THEN..ELSE of the SET_TINDEX routine in input_mod.f. The ELSE command should be executed... in other words, you cannot use the "all" word in input.geos to get all emitted tracers, you need to explicitly give the tracer numbers. That's something we could improve, but for the moment we will only emphasize that we cannot specify all tracers with "all" for ND36.
Same story for ND34, and ND28.
For these 3 diagnostics, the manual says: So you can just list these tracer numbers individually.
I guess it should say So you must list these tracer numbers individually, and never use 'all' to specify all of them.

--Bob Y. 14:37, 2 September 2008 (EDT)