Soil CO2 efflux and extractable organic carbon fractions under simulated precipitation events in a Mediterranean Dehesa

The magnitude of CO2 efflux pulses after rewetting a dry soil is highly variable and the factors regulating these pulses are poorly understood. In this field experiment, we aimed to study the C dynamics after simulated summer rainstorms in a Mediterranean open holm oak woodland (dehesa). We hypothesized that because the herbaceous cover is mostly dead during the summer in this ecosystem, the short-term CO2 efflux (SR) after rewetting could mainly be explained by different measurable soil C fractions: i) K2SO4-extracted soil C (EOC); ii) microbial biomass C (MBC); or iii) chloroform-fumigated extracted C (CFE). On both grazed and abandoned dehesa sites, we simulated three summer rain events at two-week intervals and we measured SR discontinuously in three plots under tree canopy and in another three plots in open grassland. In each plot, C fractions and water content were estimated before (2 h) and after (36 h) each irrigation event. Following rewettings, SR increased up to ten times compared with non-irrigated plots. The CFE actually increased after rewetting in the first two irrigations but not in the third event, suggesting that the capacity of the soil to release labile organic C from soil aggregates or litter was reduced after each irrigation event. Overall, the C released as CO2 in the first 24 h was related to the CFE existing before rewetting, which may help to explain the spatial variability in SR. However, the explained variability decreased after each irrigation, suggesting a change to a less labile composition of the CFE fraction as a consequence of multiple drying-rewetting cycles.