Abstract
Climate change impact on pasture floristic composition needs to be carefully assessed, given its key role for the resilience of pastoral systems and related ecosystem services. Nevertheless, variations in floristic composition are rarely taken into account in climate change impact studies. Here, we used the plant community model CoSMo to simulate future dynamics of biomass accumulation and floristic composition for high-altitude semi-natural alpine pastures. Dedicated multi-site field activities were conducted to collect data for model calibration. Simulations were run for four 20-year climate scenarios centered on 2040, resulting from the combination of two general circulation models (GISS-ES and HadGEM2) and two representative concentration pathways (RCP4.5 and RCP8.5). Results highlighted the capability of CoSMo to successfully reproduce the productivity and floristic composition of semi-natural pastures, modeling efficiency and R 2 being higher than 0.90 for aboveground biomass accumulation and relative abundance of species. CoSMo simulated an overall positive effect of increasing temperatures on pasture productivity (+ 10.7% on average), due to higher biomass accumulation rates and longer growing seasons. However, these benefits were highly heterogeneous among the monitored pastures (ranging from − 2.5 to + 16.2%), because of differences in floristic composition and in species-specific thermal requirements that led to complex, non-linear reactions to climate variations. A negative impact of climate change was simulated for grazing value (− 11.1% on average), due to the higher suitability to future conditions of Nardus stricta , which has low grazing value compared to other species. Our results highlight that floristic composition should be explicitly considered while assessing climate change impacts on semi-natural pasture productivity and connected ecosystem services.