Modelling the production and transport of dissolved organic carbon in forest soils

DyDOC describes soil carbon dynamics, with a focus on dissolved organic carbon (DOC). The model treats the soil as a three-horizon profile, and simulates metabolic carbon transformations, sorption reactions and water transport. Humic substances are partitioned into three fractions, one of which is immobile, while the other two ( hydrophilic and hydrophobic) can pass into solution as DOC. DyDOC requires site-specific soil characteristics, and is driven by inputs of litter and water, and air and soil temperatures. The model operates on hourly and daily time steps, and can simulate carbon cycling over both long (hundreds-to-thousands of years) and short (daily) time scales. An important feature of DyDOC is the tracking of C-14, from its entry in litter to its loss as (DOC)-C-14 in drainage water, enabling information about C dynamics to be obtained from both long-term radioactive decay, and the characteristic C-14 pulse caused by thermonuclear weapon testing during the 1960s ("bomb carbon"). Parameterisation is performed by assuming a current steady state. Values of a range of variables, including C pools, annual DOC fluxes, and C-14 signals, are combined into objective functions for least-squares minimisation. DyDOC has been applied successfully to spruce forest sites at Birkenes ( Norway) and Waldstein ( Germany), and most of the parameters have similar values at the two sites. The results indicate that the supply of DOC from the surface soil horizon to percolating water depends upon the continual metabolic production of easily leached humic material. In contrast, concentrations and fluxes of DOC in the deeper soil horizons are controlled by sorption processes, involving comparatively large pools of leachable organic matter. Times to reach steady state are calculated to be several hundred years in the organic layer, and hundreds-to-thousands of years in the deeper mineral layers. It is estimated that DOC supplies 89% of the mineral soil carbon at Birkenes, and 73% at Waldstein. The model, parameterised with "steady state" data, simulates short-term variations in DOC concentrations and fluxes, and in (DOC)-C-14, which are in approximate agreement with observations.