Modelling the effects of temperature and moisture on CO2 evolution from top- and subsoil using a multi-compartment approach

A novel approach, at least for laboratory conditions, for analysis of the dependence of soil C evolution on temperature is presented. A two-component (labile and refractory organic C) parallel first-order model was fitted to CO2 evolution rates from top- and subsoil, incubated at different combinations of temperature (constant −4, 0.3, 5, 15, 25, weekly fluctuating between −4 and +5°C) and moisture (17, 26, 36 and 50% H2O for the topsoil and 16, 23, 31 and 41% for the subsoil) and to the evolution of CO2 after the addition of roots or stubble of Phalaris arundinacea in the topsoil, measured at 25°C and 36% H2O (Lomander et al., 1998). The size of the pools and their respective first-order rate constants were optimized simultaneously by a least-squares method. The optimization was carried out separately for top- and subsoil. Quadratic functions were fitted to the temperature and moisture responses. For topsoil samples in which roots or stubble were added, a three-component model (labile, refractory and stubble or roots) was used. The initial partitioning of the soil C, the decomposition rate constants for each partition and the temperature and moisture responses were all assumed to be identical to those of pure topsoil, while the initial pool sizes of added roots and straw were measured. The calculated temperature at which CO2 evolution ceased (Tmin) was −0.83°C, and a recalculation to Q10-values resulted in increasing temperature response with decreasing temperature (Q10=2.2 at 25°C and 12.7 at 0.3°C). Simulated CO2 evolution rates agreed well with the measurements (Radj2=0.96 and 0.81) for top- and subsoil, respectively. The multi-compartment approach was superior to the single-compartment approach, which gave Radj2=0.88 and 0.76 for top- and subsoil, respectively. In general, CO2 evolution rates obtained from the laboratory experiment were higher than those measured in the field, even after differences in temperature and moisture were taken into account. After 300 d in the laboratory at 25°C and 36% H2O, 99% and 86% of the added straw and roots, respectively, had disappeared according to the described model. The CO2-evolution rate per unit of soil carbon was about two times higher for topsoil than for subsoil.