Modelling of carbon cycle and fire regime in an east Siberian larch forest

Abstract

Carbon budget of Siberian boreal forests is one of the largest uncertainties in the global carbon cycle under changing environment and then climate prediction. A coupled carbon cycle and fire regime model, which is applicable to simulate carbon budget of a fire-prone boreal forest in eastern Siberia, was developed in this paper. The carbon model consists of 10 pools and represents carbon dynamics including biomass burning, on the basis of an ecophysiological carbon cycle model (Sim-CYCLE). The fire model simulates fire behavior (ignition, expansion, and extinction) at a landscape scale with a spatially explicit cellular automaton, in a stochastic manner. Fire spread and burning intensity are parameterized as functions of temperature, moisture, and fuel availability, and ground fire and crown fire are dealt with separately. Interactions between the carbon cycle and fire behavior models are included, such that the carbon model simulates fuel dynamics and post-fire vegetation growth and the fire model simulates biomass burning (carbon dioxide emission, detritus production, and charcoal formation). The model was calibrated at a deciduous needle-leaved (larch) forest near Yakutsk, Russia (62°N, 130°E), in which Japanese researchers conducted surveys of carbon dynamics. The long-term simulation (1200 years) indicated that the mean fire interval of the larch forest was as short as 64 years, and most fires (94% of burnt area) occurred as ground fires. Although crown fires were less frequent (6% of burnt area), they exerted more severe impacts on carbon storage than ground fires. On average, the biomass burning released about 12% of carbon fixed by plant photosynthesis directly and resulted in acceleration of carbon cycling of the boreal forest. Young post-fire forests exhibited higher photosynthetic activity and compensated for the loss of old stands. Sensitivity of the model simulation to variation in parameter values and environmental conditions was examined. Finally, the simulation analysis confirmed that the fire regime is an important ecosystem attribute, and that a coupled carbon and fire model is useful to investigate the carbon budget of boreal forest in eastern Siberia.