Dependence of soil respiration on soil moisture, clay content, soil organic matter, and CO2 uptake in dry grasslands

The effects of abiotic and biotic drivers on soil respiration (R_s) were studied in four grassland and one forest sites in Hungary in field measurement campaigns (duration of studies by sites 2-7 years) between 2000 and 2008. The sites are within a 100 km distance of each other, with nearly the same climate, but with different soils and vegetation. Soil respiration model with soil temperature (T_s) and soil water content (SWC) as independent variables explained larger part of variance (range 0.47-0.81) than the Lloyd and Taylor model (explained variance: 0.31-0.76). Direct effect of SWC on R_s at much smaller temporal and spatial scale (1.5 h, and a few meters, respectively) was verified.Soil water content optimal for R_s (SWC_opt) was shown to significantly (positively) depend on soil clay content, while parameter related to activation energy (E₀) was significantly (negatively) correlated to the total organic carbon content (TOC) in the upper 10 cm soil layer. Dependence of model parameters on soil properties could easily be utilized in models of soil respiration. The effect of current (a few hours earlier) assimilation rates on soil respiration after removing the effect of abiotic covariates (i.e. temperature and water supply) is shown. The correlation maximum between the R_s residuals (R_{s_res}, from the R_s (SWC, T_s) model) and net ecosystem exchange (NEE) was found at 13.5 h time lag at the sandy grassland. Incorporating the time-lagged effect of NEE on R_s into the model of soil respiration improved the agreement between the simulated vs. measured R_s data. Use of SWC_opt and E₀ parameters and consideration of current assimilation in soil respiration models are proposed.