The rate of ecological acclimation is the dominant uncertainty in long-term projections of a key ecosystem service

Summary The potential for ecosystems to continue providing society with essential services may depend on their ability to acclimate to climate change through multiple processes operating from cells to landscapes1. While models to predict climate change impacts on ecosystem services often consider uncertainty among greenhouse gas emission scenarios or global circulation models (GCMs)2,3, they rarely consider the rate of ecological acclimation, which depends on decadal-scale processes such as species turnover4,5. Here we show that uncertainty due to the unknown rate of ecological acclimation is larger than other sources of uncertainty in late-century projections of forage production in US rangelands. Combining statistical models fit to historical climate data and remotely-sensed estimates of herbaceous productivity with an ensemble of GCMs, we projected changes in forage production using two approaches. The time-series approach, which assumes minimal acclimation, projects widespread decreases in forage production. The spatial gradient approach, which assumes ecological acclimation keeps pace with climate, predicts widespread increases in forage production. This first attempt to quantify the magnitude of a critical uncertainty emphasizes that better understanding of ecological acclimation is essential to improve long-term forecasts of ecosystem services, and shows that management to facilitate ecological acclimation6 may be necessary to maintain ecosystem services at historical baselines.