Comparison of fallow season CO2 efflux from paddy soil estimated using laboratory incubation with eddy covariance-based flux

To improve our understanding of the carbon dioxide (CO2) efflux dynamics in paddy fields during the drained fall to-spring fallow season, we conducted incubation experiments with paddy soil to determine the dependence of the soil CO2 emission rate on temperature and moisture. The 37-day integrated CO2 emission rate per unit soil weight was 1.45 mu mol g(-1) d(-1) at a soil temperature of 15 degrees C and a soil moisture content of 0.65 g g(-1). The temperature dependence of the soil CO2 emission rate was expressed using the Arrhenius equation, but its activation energy was likely to be underestimated if the data obtained at a high temperature were included. Soil CO2 emissions were linearly dependent on the soil moisture. We occasionally observed inverse relationships between the CO2 efflux estimated using the incubation experiments and those determined using the eddy covariance method, which was probably due to the effects of CO2 storage in the soil. However, when the data were averaged over a relatively long-time scale (10-15 days), the empirical model based on the incubation experiments successfully reproduced the soil CO2 efflux determined by the eddy covariance method, probably because the long-time scale canceled the CO2 storage effect.