Evaluation of CO2 fluxes from an agricultural field using a process-based numerical model

During 2004, soil CO2 fluxes, and meteorological and soil variables were measured at multiple locations in a 30-ha agricultural field in the Sacramento Valley, California, to evaluate the effects of different tillage practices on CO2 emissions at the field scale. Field scale CO2 fluxes were then evaluated using the one-dimensional process-based SOILCO2 module of the HYDRUS-1D software package. This model simulates dynamic interactions between soil water contents, temperature, and soil respiration by numerically solving partial–differential water flow (Richards) and heat and CO2 transport (convection–dispersion) equations using the finite element method. The model assumes that the overall CO2 production in the soil profile is the sum of soil and plant respiration, whose optimal values are affected by time, depth, water content, temperature, and CO2 concentration in the soil profile. The effect of each variable is introduced using various reduction functions that multiply the optimal soil CO2 production. Our results show that the numerical model could predict CO2 fluxes across the soil surface reasonably well using soil hydraulic parameters determined from textural characteristics and the HYDRUS-1D software default values for heat transport, CO2 transport and production parameters without any additional calibration. An uncertainty analysis was performed to quantify the effects of input parameters and soil heterogeneity on predicted soil water contents and CO2 fluxes. Both simulated volumetric water contents and surface CO2 fluxes show a significant dependency on soil hydraulic properties.