Simulate soil organic carbon with the LPJmL two-pool soil model.
Source:R/soc_models.R
calculate_soc_lpjml.RdAnnual trajectory of LPJmL's soil carbon submodel: a fast and a slow
mineral-soil pool, each losing first-order decomposition and gaining a share
of the carbon that survives litter respiration. Of the litter carbon that
DECOMPOSES, atmosphere_fraction is respired straight to the
atmosphere and the remainder is split fast_fraction to the fast pool
and the rest to the slow pool. Neither soil pool transfers to the other and
there is no inert pool, so soil respiration leaves the system entirely.
Usage
calculate_soc_lpjml(
initial_soc_mgc_ha,
c_input_mgc_ha_yr,
years,
clay_pct = NA,
climate_modifier = 1
)Source
Schaphoff, S., von Bloh, W., Rammig, A., Thonicke, K., Biemans, H.,
Forkel, M., ... Waha, K. (2018). LPJmL4 - a dynamic global vegetation model
with managed land - Part 1: Model description. Geoscientific Model
Development, 11, 1343-1375. doi:10.5194/gmd-11-1343-2018
, Sect. 2.5 and
Eqs. 90-100. Rate constants and fractions are taken from the WHEP LPJmL
6.1.1 run configuration, which differs from the published values for
atmosphere_fraction, fast_fraction and the fast-pool rate;
see soc_turnover_params.
Arguments
- initial_soc_mgc_ha
Initial soil organic carbon stock (Mg C per ha), split between the two pools by their steady-state proportions.
- c_input_mgc_ha_yr
Annual carbon input to the litter layer (Mg C per ha per year), before the atmospheric respiration share.
- years
Number of years to simulate.
- clay_pct
Soil clay content (percent); unused, kept for contract.
- climate_modifier
Annual decomposition response (dimensionless), scaling both pool rates. See
soc_rate_modifier_lpjml.
What the atmospheric fraction applies to
In LPJmL the fraction multiplies the flux LEAVING the litter pool by decomposition, not the litterfall entering it. The two are equal only when the litter pool is at steady state, and litterfall is strictly the larger wherever litter burns or methanogenises. WHEP has no litter pool, so this function applies the fraction to the carbon input it is given and is therefore a steady-state reading of that chain. Over a transient the approximation is good for leaf and fine-root litter, which transit in one to three years, and poor for woody litter, whose 10-26 year transit is comparable to the 22.25-year soil-bound residence itself.
Earlier wording here said the fraction applied to carbon "entering the
litter layer", which contradicted soc_turnover_params's own
description of it ("Share of decomposed litter respired straight to the
atmosphere"). The parameter table was right.
This applies LPJmL's kinetics to the carbon input WHEP supplies, exactly as
calculate_soc_rothc applies RothC's. It is not a
reproduction of LPJmL's own soil carbon stock, which is reported over
0-300 cm and distributed down the profile; the stock returned here is
over the same layer as WHEP's other models. The litter pool is excluded, as
it is from LPJmL's own soilc output.
That vertical distribution is not a rooting-depth function, as this page previously said: LPJmL sums leaf, woody and belowground decomposition into one flux before applying the layer weights, so surface litter is carried down too, and Schaphoff et al. (2018) attribute the weights to cryoturbation and bioturbation.