Local effects of climate change (CC) and elevated CO2 (2 x CO2, 660 mu mol mol(-1)) on managed temperate grasslands were assessed by forcing a dynamic ecosystem model with weather scenarios. The aims of the study were to compare the relative importance of individual and combined effects of CC, 2 x CO2, and photosynthetic acclimation, and to assess the importance of local site conditions. The model was driven by hourly means for temperature (T), precipitation (P), global radiation (G), vapour pressure (VP), and wind speed (U). Local climate scenarios were derived by statistical downscaling techniques from a 2 x CO2 simulation with the General Circulation Model of the Canadian Climate Centre (CCC-GCMII). Simulations over 14 growing seasons to account for year-to-year variability of climate were carried out for a low, relatively dry site, and a high, more humid site. At both sites, shoot dry matter responded positively to 2 x CO2 with the site at low elevation being more sensitive than the higher site. The effect of assumed changes in climate was negative at the lower, but positive at the higher site. Shoot dry matter was more sensitive to the effects of 2 x CO2 than to CC. Both effects combined increased shoot dry matter by up to 20%. This was attributed to direct effects of 2 x CO2 and increased T,and indirect stimulation via increased soil N availability. Biomass partitioning to roots increased with 2 x CO2 but decreased with CC, while an intermediate response resulted from the combination. Leaf area index (LAI) increased under 2 x CO2, but not enough to compensate fully for a decrease in leaf conductance. Under the 2 x CO2 scenario evapotranspiration (ET) decreased, but increased under CC. Photosynthetic acclimation reduced the effect of 2 x CO2 on shoot growth, but had little effect on ET. The seasonal water use efficiency (WUE) was improved under 2 x CO2, and reduced under CC. With the combination of both factors, the change was small but still positive, especially at the high elevation site with more favourable soil water conditions. This reflects the stronger positive yield response in combination with a smaller increase in ET under cooler, more humid conditions. The results for the combination of factors suggest that except for shoot growth, effects of 2 x CO2 and CC tend to offset each other. While CC determines the sign of the ET response, the sign of the biomass response is determined by 2 x CO2. The results highlight the importance of a site-specific analysis of ecosystem responses by using a flexible approach based on a combination of state-of-the-art downscaling, spatially resolved data sets, and a mechanistic model to obtain quantitative and reproducible assessments of climate change impacts at the ecosystem level.
