Altered Seasonality and Magnitude of Rainfall Affects Soil Respiration and Nitrous Oxide Fluxes in California Annual Grassland

Currently, climate models do not agree on how rising concentrations of CO₂ and other greenhouse gases will affect rainfall in California. Changes in moisture regime will likely alter rates of carbon (C) loss via soil respiration, as well as fluxes of N₂O. Moisture availability can also affect plant productivity in highly seasonal environments. We examined the consequences of wetter conditions in an annual grassland in the Sierra foothills of northern California by extending the duration of the wet season by about 5 weeks and augmenting total annual rainfall by approximately 50 %. Discrete wet-up events took place prior to the onset of natural rains (early October 2003) and early in the drought period (May 2004). Soil respiration, N₂O and CH₄ effluxes, N mineralization, and above- and belowground plant production were measured in treatment and control plots over a one-year period. Soil CO₂ fluxes for the first treatment year, though large, were not statistically different between wet and control plots (1078 \pm148 g C m^-2 and 1006 \pm138 g C m^-2, respectively). The combined wet-up events comprised 17 % of the soil respiration over the 12-month period in treated plots, about twice as much C released by control plots during the same time interval. Aboveground biomass was similar between wetted and control plots (415 \pm45 g m^-2 y^-1 and 374 \pm36 g m^-2 y^-1, respectively), while root biomass increased significantly with wetting during the first year of treatment (179 \pm23 g m^-2 y^-1 and 111 \pm13 g m^-2 y^-1 for treatment and control plots, respectively). The additional biomass C gained in treatment plots (53 g C m^-2) partly offset the greater losses from respired C observed in treatment plots (72 g C m^-2). Nitrous oxide emissions were low to negligible during the year with the exception of the time directly following wet-up, when N₂O emissions averaged over 78\pm13 ng N cm^-2 h^-1. Our first year of water manipulation in annual grasslands suggests that increased water availability via early and late rainfall events releases large pulses of CO₂, increases belowground C inputs, and increases N₂O emissions.