Difference between revisions of "CABLE"
(→CABLE - CSIRO Atmosphere Biosphere Land Exchange model) |
|||
(2 intermediate revisions by one other user not shown) | |||
Line 1: | Line 1: | ||
− | CABLE - CSIRO Atmosphere Biosphere Land Exchange model | + | == CABLE - CSIRO Atmosphere Biosphere Land Exchange model == |
+ | |||
I have fluxes provided by Chris Lu and Rachel Law that I want to use to investigate differences in NEE over Australia, particularly in comparison to CarbonTracker. | I have fluxes provided by Chris Lu and Rachel Law that I want to use to investigate differences in NEE over Australia, particularly in comparison to CarbonTracker. | ||
Line 17: | Line 18: | ||
But, I am still getting two compilation errors: | But, I am still getting two compilation errors: | ||
− | co2_mod.F(294): error #6460: This is not a field name that is defined in the encompassing structure. [LBIOCABLE] | + | co2_mod.F(294): error #6460: This is not a field name that is defined in the encompassing structure. [LBIOCABLE] |
− | + | LBIOCABLE = Input_Opt%LBIOCABLE | |
− | ------------------------------^ | + | ------------------------------^ |
− | co2_mod.F(3661): error #6460: This is not a field name that is defined in the encompassing structure. [LBIOCABLE] | + | co2_mod.F(3661): error #6460: This is not a field name that is defined in the encompassing structure. [LBIOCABLE] |
− | + | LBIOCABLE = Input_Opt%LBIOCABLE | |
− | ------------------------------^ | + | ------------------------------^ |
− | compilation aborted for co2_mod.F (code 1) | + | compilation aborted for co2_mod.F (code 1) |
− | make[3]: *** [co2_mod.o] Error 1 | + | make[3]: *** [co2_mod.o] Error 1 |
− | make[3]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' | + | make[3]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' |
− | make[2]: *** [lib] Error 2 | + | make[2]: *** [lib] Error 2 |
− | make[2]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' | + | make[2]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' |
− | make[1]: *** [all] Error 2 | + | make[1]: *** [all] Error 2 |
− | make[1]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' | + | make[1]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' |
− | make: *** [all] Error 2 | + | make: *** [all] Error 2 |
Line 38: | Line 39: | ||
Assuming this works, I still need to re-format the CABLE flux files into something more logically readable by GEOS-Chem. | Assuming this works, I still need to re-format the CABLE flux files into something more logically readable by GEOS-Chem. | ||
+ | |||
+ | ... I just watched co2_mod.f roll by in the compilation. Success?! | ||
+ | |||
+ | Onto the CABLE handling... | ||
+ | |||
+ | [[Re-gridding CABLE fluxes to GEOS-Chem compatible files]] |
Latest revision as of 19:18, 16 February 2016
CABLE - CSIRO Atmosphere Biosphere Land Exchange model
I have fluxes provided by Chris Lu and Rachel Law that I want to use to investigate differences in NEE over Australia, particularly in comparison to CarbonTracker.
These fluxes are currently on a land only grid, with hourly resolution. The grid is 0.5x0.5°, so potentially 720 x 360 (=259200 points). File size is saved by only including land point fluxes (it is a biosphere model, after all), leaving 62433 points.
Of course, this creates the headache of mapping the flux to the correct grid point, especially as the lat/lon coordinates are not given for each netCDF file, except for the original 0.5x0.5° resolution. So I will regrid to 0.5x0.5° (basically map the flux to match the corresponding points if it is a land point, or set to 0 if not). Then I can worry about aggregating to other resolutions.
The mapping of the 62433 land points to lat/lon is provided in the file latlon_cable.csv, in the format (point lat lon).
In the GEOS-Chem code, I am modifying co2_mod.f, and have added an option "LBIOCABLE" and subroutine "READ_BIO_CABLE"
At present I have made it is far as where I need to build a filename to read the fluxes. Until I set the re-gridded files up, that's all I can probably do.
Also, with the current changes it did not yet compile. I needed to add the Input_Opt%LBIOCABLE = LBIOCABLE line to Input_mod.F I also needed to add the LBIOCABLE logical to logical_mod.F
But, I am still getting two compilation errors:
co2_mod.F(294): error #6460: This is not a field name that is defined in the encompassing structure. [LBIOCABLE] LBIOCABLE = Input_Opt%LBIOCABLE ------------------------------^ co2_mod.F(3661): error #6460: This is not a field name that is defined in the encompassing structure. [LBIOCABLE] LBIOCABLE = Input_Opt%LBIOCABLE ------------------------------^ compilation aborted for co2_mod.F (code 1) make[3]: *** [co2_mod.o] Error 1 make[3]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' make[2]: *** [lib] Error 2 make[2]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/ndeutsch/CO2_CABLE/CO2_CABLE/GeosCore' make: *** [all] Error 2
I have made these references as similar as possible to others (e.g. LBIODIURNAL), but I still cannot see what is the problem.
Okay, previously I was only searching *.F files for references - it turns out there is also a .F90 (gigc_input_opt_mod.F90) that requires modification. I have done that, and I am now recompiling.
Assuming this works, I still need to re-format the CABLE flux files into something more logically readable by GEOS-Chem.
... I just watched co2_mod.f roll by in the compilation. Success?!
Onto the CABLE handling...