The soil respiration flux in drylands is closely coupled to water availability and is characterized by brief pulses after rain separated by variable dry periods with low flux. The environmental drivers of these respiration pulses and their importance to overall soil carbon cycling is not well understood. We assembled continuous soil CO2 concentration profile and surface flux (soil respiration) data from 8 locations spanning the Mojave, Sonoran, Chihuahuan, and Great Basin deserts of North America. All of these studies used comparable, but not identical, measurement techniques that yielded time series of observed soil CO2 flux. At a subset of sites, fluxes at multiple soil depths were available. From these time series we used statistical techniques to identify respiration pulses associated with precipitation and soil wetting events, and then examined variations in the seasonality of these events, and the contribution of these events to the total soil CO2 flux at each site.
Results/Conclusions
Summer respiration pulses accounted for a higher fraction of the total soil CO2 flux at sites within the North American Monsoon boundary (Sonoran and Chihuahuan deserts), while spring respiration pulses were more important in the Mojave and Great Basin deserts. Across all sites, respiration pulses accounted for less time at the warm, dry sites (4-8% of time) than at cooler, wetter sites (up to 15% of time), yet produced a significantly higher percentage of total CO2 flux at warm, dry sites (31% vs 25%). When we analyzed sites with fluxes available for multiple soil depths, CO2 released from soils during pulses was largely produced in the top layer of the soil profile (~50-99%), with a greater fraction coming from the surface at warmer sites. Between pulses, the majority of CO2 efflux was produced in middle to lower depths (50-70%). Pulse events, though sometimes infrequent, were larger (in terms of CO2 produced) and more brief during warm months, as compared to respiration pulses during cooler months. Respiration pulse size was significantly related to precipitation event size, but was not positively correlated with time since the last pulse at most sites. These results, based on spatially and temporally diverse in-situ soil measurements, provide a rich picture of respiration pulse events in North America's deserts. As the frequency and timing of precipitation events in this region shift, we anticipate consequent shifts in the amount and timing of soil CO2 fluxes and in the soil carbon pools contributing to this flux.