Effects of soil moisture and temperature on CO2 and CH4 soil–atmosphere exchange of various land use/cover types in a semi-arid grassland in Inner Mongolia, China

The aim of this study was to investigate the combined effects of soil moisture and temperature as well as drying/re-wetting and freezing/thawing on soil–atmosphere exchange of CO2 and CH4 of the four dominant land use/cover types (typical steppe, TS; sand dune, SD; mountain meadow, MM; marshland, ML) in the Xilin River catchment, China. For this purpose, intact soil cores were incubated in the laboratory under varying soil moisture and temperature levels according to field conditions in the Xilin River catchment. CO2 and CH4 fluxes were determined approximately daily, while soil CH4 gas profile measurements at four soil depths (5 cm, 10 cm, 20 cm and 30 cm) were measured at least twice per week. Land use/cover generally had a substantial influence on CO2 and CH4 fluxes, with the order of CH4 uptake and CO2 emission rates of the different land use/cover types being TS ≥ MM ≥ SD > ML and MM > TS ≥ SD > ML, respectively. Significant negative soil moisture and positive temperature effects on CH4 uptake were found for most soils, except for ML soils. As for CO2 flux, both significant positive soil moisture and temperature effects were observed for all the soils. The combination of soil moisture and temperature could explain a large part of the variation in CO2 (up to 87%) and CH4 (up to 68%) fluxes for most soils. Drying/re-wetting showed a pronounced stimulation of CO2 emissions for all the soils —with maximum fluxes of 28.4 ± 2.6, 50.0 ± 5.7, 81.9 ± 2.7 and 10.6 ± 1.2 mg C m−2 h−1 for TS, SD, MM and ML soils, respectively—but had a negligible effect on CH4 fluxes (TS: −3.6 ± 0.2; SD: 1.0 ± 0.9; MM: −4.1 ± 1.3; ML: −5.6 ± 0.8; all fluxes in μg C m−2 h−1). Enhanced CO2 emission and CH4 oxidation were observed for all soils during thawing periods. In addition, a very distinct vertical gradient of soil air CH4 concentrations was observed for all land use/cover types, with gradually decreasing CH4 concentrations down to 30 cm soil depth. The changes in soil air CH4 concentration gradients were in accordance with the changes of CH4 fluxes during the entire incubation experiment for all soils.