Too much carbon... in soils, now?

Musings on:
Stewart CE, Paustian K, Conant RT, Plant AF, Six A. 2007. Soil carbon saturation: concept, evidence and evaluation. Biogeochemistry 86:19-31.

On first glance I thought that this was too obvious to be significant - if carbon input fluxes (e.g. litter fall) are increased (in a single step), then of course soil carbon will increase, but asymptotically to a new maximum, caused by establishment of a new steady state with increased losses due to mineralisation, etc.
I was wrong; this simplistic understanding assumes, as Stewart et al. point out, that every increase in input flux will eventually increase soil C content. In fact, with this assumption the relationship between input flux and new equilibrium soil carbon concentration turns out to be linear; this is what most models of soil organic carbon dynamics (e.g. CENTURY or RothC) do.

Stewart et al.'s hypothesis is that with increasing input fluxes to soil, the soil C content does not keep increasing but reaches a maximum value. Using sites having a range of annual carbon inputs (0-6 Mg C/ha/y), with treatments ranging in duration from 12-51 years, they find that the data as a whole are described best by a model that assumes that soil carbon content, or at least one pool of soil carbon, reaches a limiting value.

Clearly (if correct) Stewart et al.'s findings have implications for carbon sequestration in soils for controlling CO₂ concentrations in Earth's atmosphere (again, the authors make a point of emphasising this). Soils, then, behave the same way as reforestation/revegetation in terms of carbon sequestration (ecosystems reach steady states eventually in terms of productivity vs. loss). Soil is not an infinite sink for carbon; neither is the biomass growing on it.

It's early days before this hypothesis is fully evaluated across a range of biomes, but it needs to be taken notice of. There may be some issues with the treatment of data; it wasn't clear, for example, whether clustering of data for different sites affected the fitting procedure for the carbon saturation model. But we'd better understand stuff like this if we're to be serious about exploiting the role of soils in global carbon dynamics.

Soil image from Australian Society of Soil Science Inc.