Experimental Warming and Precipitation Effects on Plant Community Composition, Productivity, Nutrient Availability, and Soil Respiration in Pacific Northwest Prairies along a Natural Climate Gradient

Climate change effects on soil respiration and carbon stores in grasslands globally may have significant implications for future atmospheric carbon dioxide concentrations. Climate change may also may negatively impact native plant species and favor exotic species. We are experimentally increasing temperature by 3 degrees C and increasing precipitation by 25% above ambient in three upland prairie sites along a natural climate gradient from southwestern Oregon to central-western Washington to determine how future climate change will affect (i) plant community composition and the relative success of native versus introduced plant species and (ii) above- and belowground carbon and nutrient dynamics. Sixty plots (20 at each site) were restored by mowing, raking, and herbicide application followed by the sowing of the same 34 native grass and forb species in each plot. Differences in total cover, net primary productivity, and community composition were much greater among sites than among treatments within sites in both 2010--the establishment year, and 2011-the first full year of treatment. Strong successional dynamics occurred over the two years as competition intensified, but these were dependent on a site-treatment interaction, with lower native plant survival in heated plots because of competitive exclusion by exotic, invasive plants. A strong treatment - season interaction in canopy cover (as determined by canopy reflectance) also occurred, with heating causing greater cover during the wet season and lower cover during the dry season. This effect was strongest in the southernmost site which experiences earlier and more intense drought conditions. There were also strong site, treatment, and season interactions on nutrient availability as determined by cation-anion exchange resins. Heating increased nutrient availability in all but the northernmost site during the growing season, and that site also had much lower nutrient availability, but overall availability and site and treatment effects were less during the winter. Plant dynamics among the sites are likely strongly controlled by these differences in nutrient availability. In the winter, heating increased soil respiration, but this effect was lost as the sites warmed and dried during the growing season, with this switch in treatment effect occurring earlier in the spring in the southernmost site. Overall, we have observed many significant effects of the warming and precipitation treatments on plant and ecosystem dynamics, but these effects are often seasonally dependent and of lesser importance than strong differences among sites, driven by differences in soils and the pressure from invasive species. Our experiment shows the importance of placing climate change impacts on natural ecosystems within a context of local ecosystem controls.