Soil porosity is usually taken as a constant over time for a given field, although in reality it decreases with time after tillage. For the gradient method, estimating soil CO2 production with a fixed porosity may lead to large errors when soil porosity varies over time. In this study, we compared soil air-filled porosity, gas diffusivity, and CO2 production based on a temporally variable soil porosity (?V) with those based on a constant porosity, either initial porosity just after soil tillage (?i) or final porosity at harvest after a tilled soil has settled (?f). Soil porosity was measured seven times during a maize (Zea mays L.) growing season, and an exponential relationship of soil porosity with time was developed to describe ?V for the 0- to 5-cm soil layer. Soil CO2 production was estimated from the gradient method and the mass conservation law. Soil-surface CO2 efflux was measured with a dynamic chamber throughout the growing season. The ?i value was 0.49 m3 m-3 and the ?f value was 0.43 m3 m-3. Compared with results obtained from ?V, soil air-filled porosity, gas diffusivity, and CO2 production values obtained from ?f were 6, 11, and 22% lower, whereas values obtained from ?i were 17, 36, and 70% larger. The soil-surface CO2 effluxes estimated with ?V better matched the chamber values than did the estimates with ?i or ?f. We conclude that use of variable soil porosity improves estimations of soil-surface CO 2 effluxes and soil CO2 production with the gradient method. © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA All rights reserved.
