Need to provide Joran with definitions of fluxes for leaching

[rosiealice]

Joeran says:

thanks to the great initial work of Mariana and Matvey on the MOSART code, we eventually implemented a generic tracer advection scheme and got MOSART technically(!) ready to transport tracers in a mass-conserving manner. With this community effort, we got one (big!) step further to advance NorESMs capability towards coupling across the land-ocean continuum. This will aid, amongst others, in coupling land-river-ocean biogeochemistry.

In order to step further forward, I would like to suggest/lay out the next steps and would like to kindly ask for your further support. I envisage to implement a first MOSART river biogeochemistry early next year and I am currently investigating which pathway to take in order to bridge the different modelling strategies on the land and ocean biogeochemistry side via the river biogeochemistry.

1. For this decision process, it would be helpful in a first step, if you could provide information on which leaching fluxes are/will be provided from the land side (i.e. inorganic/organic, dissolved/particulate, potentially 'quality' of organic matter in terms of stoichiometry (flexible or fixed) and about potential (decay) classes leaching on the land side - so some overview on expected leaching fluxes lost to the river network). This will aid in connecting the dots between leaching fluxes and ocean biogeochemistry by a simplified or more elaborated river biogeochemistry.

2. As a second step, it would be helpful if those fluxes could be written out to enable providing them -even untuned- as forcing fields to MOSART so that initial test-runs with a new river biogeochemistry could be carried out - sort of a climatological field of leaching and water fluxes.

[rosiealice]

The leaching fluxes I can find in the output file are:

SMIN_NO3_LEACHED
SMIN_NO3_RUNOFF
SOM_C_LEACHED

Additional variables we could print out (but are not done so by default) include:

SOM_N_LEACHED
decomp_npools_leached
decomp_cpools_leached

The latter two appear to break down the organic leaching per soil organic matter pool. They are calcualted from the decomposition of the SOM pools here:

CTSM/src/soilbiogeochem/SoilBiogeochemNitrogenFluxType.F90

Line 1233 in ca5cddb

do j = 1, nlevdecomp

SOM_N_LEACHED is the sum of decomp_npools_leached

These do not appear in the tech note description:
https://escomp.github.io/ctsm-docs/versions/release-clm5.0/html/tech_note/External_Nitrogen_Cycle/CLM50_Tech_Note_External_Nitrogen_Cycle.html#leaching-losses-of-nitrogen

The SMIN_NO3_LEACHED flux is calculated in a completely different way, from the soil water mineral N concentration and the drainage
https://github.com/NorESMhub/CTSM/blob/ca5cddbcfe543b89e8f9e6b5fb1941b8c8297925/src/soilbiogeochem/SoilBiogeochemNLeachingMod.F90#L218C42-L218C73

So it seems that these are indeed different pathways. Maybe the organic matter pathway was added after the tech note description. The git blame shows that all this code was refactored by Mariana a few months after CLM5 freeze
https://github.com/NorESMhub/CTSM/blame/ca5cddbcfe543b89e8f9e6b5fb1941b8c8297925/src/soilbiogeochem/SoilBiogeochemNitrogenFluxType.F90#L1233

One thing we could do is test with a CLM-BGC run whether this works or not?

[jmaerz]

Related thread for diagnostics: NorESMhub/noresm3_dev_simulations#48 (comment)
Maybe we can change the discussion thread title to some more meaningful: "Definition of leaching fluxes to MOSART" ?