Effects of Elevated CO2 on Root Function and Soil Respiration in a Mojave Desert Ecosystem

Increases in atmospheric CO{sub 2} concentration during the last 250 years are unequivocal, and CO{sub 2} will continue to increase at least for the next several decades (Houghton et al. 2001, Keeling & Whorf 2002). Arid ecosystems are some of the most important biomes globally on a land surface area basis, are increasing in area at an alarming pace (Dregne 1991), and have a strong coupling with regional climate (Asner & Heidebrecht 2005). These water-limited ecosystems also are predicted to be the most sensitive to elevated CO{sub 2}, in part because they are stressful environments where plant responses to elevated CO{sub 2} may be amplified (Strain & Bazzaz 1983). Indeed, all C{sub 3} species examined at the Nevada Desert FACE Facility (NDFF) have shown increased A{sub net} under elevated CO{sub 2} (Ellsworth et al. 2004, Naumburg et al. 2003, Nowak et al. 2004). Furthermore, increased shoot growth for individual species under elevated CO{sub 2} was spectacular in a very wet year (Smith et al. 2000), although the response in low to average precipitation years has been smaller (Housman et al. 2006). Increases in perennial cover and biomass at the NDFF are consistent with long term trends in the Mojave Desert and elsewhere in the Southwest, indicating C sequestration in woody biomass (Potter et al. 2006). Elevated CO{sub 2} also increases belowground net primary production (BNPP), with average increases of 70%, 21%, and 11% for forests, bogs, and grasslands, respectively (Nowak et al. 2004). Although detailed studies of elevated CO{sub 2} responses for desert root systems were virtually non-existent prior to our research, we anticipated that C sequestration may occur by desert root systems for several reasons. First, desert ecosystems exhibit increases in net photosynthesis and primary production at elevated CO{sub 2}. If large quantities of root litter enter the ecosystem at a time when most decomposers are inactive, significant quantities of carbon may be stored belowground in relatively recalcitrant forms. Indeed, a model-based analysis predicted that the arid/semiarid southwestern bioclimatic region had one of the highest rates of net carbon storage in the United States over the past century (Schimel et al. 2000). Second, root systems of desert plants are often extensive (Foxx et al. 1984, Hartle et al. 2006) with relatively large proportions of roots deep in the soil (Schenk & Jackson 2002). Thus, an understanding of belowground processes in desert ecosystems provides information on the potential for terrestrial carbon sequestration in desert ecosystems