Estimation of CO2 effluxes from suburban forest floor and grass using a process-based model

Carbon dioxide is an important greenhouse gas, and its atmospheric concentration has been predicted to increase in the future. The objective of this study was to quantify the soil CO2 efflux in a suburban area including mixed deciduous forest and grass by numerically modeling the CO2 transport through the soil profile. Three stations per land-cover (forest and grass) were selected at the Cub Hill site (MD, USA), where the US Forest Service operates an urban flux tower. Six VAISALA CO2 sensors (Vaisala Inc., Finland) per monitoring station were horizontally installed at 6 different depths (soil surface, 0.02, 0.05, 0.10, 0.20, and 0.30 m from the soil surface) in the mid of May, 2011. Temperature and volumetric soil moisture measurements were taken using thermistors and EC-5 sensors (Decagon devices, Pullman, WA, USA) that were installed at the same depths as the CO2 sensors except for the soil surface. These data were recorded every 10 min. To evaluate the numerical model (SOILCO2), CO2 efflux using the standard chamber method was measured once a week. The CO2 effluxes from the standard chamber method ranged from 3.32 x 10(-9) to 7.28 x 10(-8) m(3) m(-2) s(-1) and 6.79 x 10(-9) to 1.45 x 10(-7) m(3) m(-2) S-1 for forest and grass, respectively. The CO2 effluxes from "bare" soil at the grass site varied with the range of 3.63 x 10(-8) to 9.37 x 10(-8) m(3) m(-2) s(-1). The "pulse effect" (a rapid increase of CO2 concentrations right after rainfall events) in grass, where changes in soil moisture were larger than in the forest, was more apparent than in the forest. Diurnal patterns similar to those of temperature were observed from CO2 profiles in soils. The SOILCO2 model estimated the soil CO2 effluxes with coefficients of correlation of 0.64 and 0.76 at forest and grass, respectively, and root mean square error (RMSE) of 1.58 x 10(-8) and 2.06 x 10(-8) m(3) m(-2) s(-1) for forest and grass, respectively. This study suggests that the SOILCO2 model can provide a better understanding of the contribution of the soil ecosystem to the carbon cycle in suburban environments including mixed deciduous forest and grass. (C) 2014 Elsevier Ltd. All rights reserved.