Contribution of dissolved organic matter to carbon storage in forest mineral soils

Abstract

Dissolved organic matter (DOM) is often considered the most labile portion of organic matter in soil and to be negligible with respect to the accumulation of soil C. In this short review, we present recent evidence that this view is invalid. The stability of DOM from forest floor horizons, peats, and topsoils against microbial degradation increases with advanced decomposition of the parent organic matter (OM). Aromatic compounds, deriving from lignin, likely are the most stable components of DOM while plant‐derived carbohydrates seem easily degradable. Carbohydrates and N‐rich compounds of microbial origin produced during the degradation of DOM can be relatively stable. Such components contribute much to DOM in the mineral subsoil. Sorption of DOM to soil minerals and (co‐)precipitation with Al (and probably also with Fe), especially of the inherently stable aromatic moieties, result in distinct stabilization. In laboratory incubation experiments, the mean residence time of DOM from the Oa horizon of a Haplic Podzol increased from <30 y in solution to >90 y after sorption to a subsoil. We combined DOM fluxes and mineralization rate constants for DOM sorbed to minerals and a subsoil horizon, and (co‐)precipitated with Al to estimate the potential contribution of DOM to total C in the mineral soil of a Haplic Podzol in Germany. The contribution of roots to DOM was not considered because of lack of data. The DOM‐derived soil C ranges from 20 to 55 Mg ha–1 in the mineral soil, which represents 19%–50% of the total soil C. The variation of the estimate reflects the variation in mineralization rate constants obtained for sorbed and (co‐)precipitated DOM. Nevertheless, the estimates indicate that DOM contributes significantly to the accumulation of stable OM in soil. A more precise estimation of DOM‐derived C in soils requires mineralization rate constants for DOM sorbed to all relevant minerals or (co‐)precipitated with Fe. Additionally, we need information on the contribution of sorption to distinct minerals as well as of (co‐)precipitation with Al and Fe to DOM retention.