Check (and fix) mass balance of woodCreation

[dlebauer]

fluxes.woodCreation is calcuated from a running average of NPP.

C used in 'woodCreation' is not accounted for by removal from another pool.

We have discussed some confusion surrounding fluxes.woodCreation in the past, and added code comments to help clarify. But these don't seem to address the mass balance, so I am not sure if this was specifically addressed

sipnet/src/sipnet/sipnet.c

Lines 1495 to 1502 in 072a967

	void updateMainPools() {
	// Update the stocks, with fluxes adjusted for length of time step.
	// Notes:
	// - GPP shows up twice (direct + via NPP --> woodCreation), but
	// the math works out to:
	// envi.plantWoodC += NPP_allocation_to_wood − woodLitter.
	// - The soil C pool(s) (envi.soil, envi.fineRootC, envi.CoarseRootC)
	// are updated in updatePoolsForSoil().

Bug (?): It does not seem that the fluxes.woodCreation are ever subtracted from a donor pool.

Tasks:

• Ensure woodCreation is a transfer, not a net C input. The fact that woodCreation is calculated from a moving average GPP might make this more complicated, but the math should work on some time scale. Maybe remove woodCreation dependence on GPP and subtract from NPP?
• Test system C balance
• Document

[Divya-A10]

Hi @dlebauer ,
I’ve been reading through the SIPNET code and the comments around fluxes.woodCreation, and I want to make sure I’m understanding the intended carbon bookkeeping correctly before attempting any changes.
From what I can tell, woodCreation contributes to plantWoodC, but I’m not seeing an explicit donor pool from which this carbon is subtracted. That made me wonder whether the mass balance is handled implicitly elsewhere via the NPP/GPP accounting, or if this is something that still needs to be made explicit.
Before trying to address this, I wanted to ask:
– Is the expectation that woodCreation should ultimately be treated as a transfer from NPP (or another pool), rather than as a standalone flux derived from GPP?
– Would it make sense to first focus on the sub-issue (#209) and establish a system-level carbon balance test before modifying the flux logic itself?
Thanks!

[Alomir]

Current task is to update biomass carbon pool models to match doc from #232.

[Alomir]

@Divya-A10 Thanks for your interest and comments! To answer your questions:

1. Since NPP is the fraction of GPP allocated to growth, I think it's correct to continue using that for the wood pool creation (i.e., growth) term. All four biomass carbon pools are treated this way.
2. I can see the point in getting Test C balance closure #209 done first - however, since we have known issues here to correct, I think I would prefer that this is done ASAP; or, at least, as soon as Updates for N limiting and biomass C pool models #232 is finished.

I hope the new documentation in #232 makes the model process clearer.

I might want to tackle this one myself, let's see where we are when the doc is finalized.

[Divya-A10]

Hi @Alomir , thanks for the clarification.
Understood on continuing to use NPP for biomass growth across all pools, and I’ll go through the updated documentation in #232 once it’s finalized to better align with the intended process.
Happy to help with testing or C balance diagnostics (e.g., #209) once the biomass pool updates land, if that would be useful.