Nitrogen cycle problem

[jinmuluo]

Hi, @wwieder

I've been working on the hydrology cycle and nitrogen cycle recently and have some questions on the water flux and nitrogen cycle. In CLM's hydrological modules, I saw it evaluates the surface water runoff and water drainage(sub-surface runoff) at the whole column level, and it makes sense to use these two fluxes to evaluate the nitrogen loss at the surface and sub-surface. But, there is another module that calculates the water diffusion between different depths called soilwater_zengdecker2009 in SoilWaterMovementMod.F90. Is this different with the sub-surface water runoff mentioned before, and which one is the real water drainage?
If I understand correctly, the SoilWaterMovementMod.F90 calculates the fast aqueous water transport vertically, can we attach the soluble inorganic nitrogen to this flux to evaluate the vertical transport of nitrogen/carbon? CLM hasn't been done yet, any considerations behind this, or if we apply the vertical fast aqueous transport of nitrogen in CLM, what would that happen? Thank you so much,

Best,

Jinmu

 

[slevis]

Regarding the sub-surface runoff:
Think of this variable as the column output from the "bottom" of the land model that becomes one of the inputs to the river model.

 

[wwieder]

Thanks for asking this question @jinmuluo.

I think you and Sam are correct that leaching losses of NO3 are handed as a function of soil layer NO3 concentrations and moisture status of each soil layer in SoilBiogeochemNLeachingMod.

I'd also assume this means that NO3 transfers between soil layers use the same diffusion code used for the rest of the soil BGC code, but I haven't tracked this carefully. If this is accurate, then yes, assigning vertical transport as a function of water fluxes between soil layers from SoilWaterMovementMod would be a more accurate solution. I'm not sure if this is how the Norwegian group was thinking about this for DOC fluxes?

At the same time, I also worry a larger issue with the soil BGC code is that we sequentially handle inorganic N fluxes to plants and heterotrophs, followed by nitrifiers and denitrifiers, with leaching happening last. As a result of this mathematical error we're always going to have small leaching losses of NO3. Bill Riley and others have thought about this with ELM, applying an ECA approach where competition for inorganic N is handled simultaneously, which could also be nice to apply. There's a bit of discussion in this issue on the CTSM github page.

If this is a topic you're going to be able to spend some time on, maybe it's worth a group discussion to see what the best scientific and software implementation may be to address your questions?