ABSTRACT. Ensembles of process-based crop models are increasingly used to simulate crop growth for scenariosof temperature and/or precipitation changes corresponding to different projections of atmospheric CO2concentrations. This approach generates large datasets with thousands of simulated crop yield data. Suchdatasets potentially provide new information but it is difficult to summarize them in a useful way due totheir structural complexities. An associated issue is that it is not straightforward to compare crops and tointerpolate the results to alternative climate scenarios not initially included in the simulation protocols.Here we demonstrate that statistical models based on random-coefficient regressions are able to emulateensembles of process-based crop models. An important advantage of the proposed statistical models isthat they can interpolate between temperature levels and between CO2concentration levels, and canthus be used to calculate temperature and [CO2] thresholds leading to yield loss or yield gain, without re-running the original complex crop models. Our approach is illustrated with three yield datasets simulatedby 19 maize models, 26 wheat models, and 13 rice models. Several statistical models are fitted to thesedatasets, and are then used to analyze the variability of the yield response to [CO2] and temperature.Based on our results, we show that, for wheat, a [CO2] increase is likely to outweigh the negative effectof a temperature increase of +2◦C in the considered sites. Compared to wheat, required levels of [CO2]increase are much higher for maize, and intermediate for rice. For all crops, uncertainties in simulatingclimate change impacts increase more with temperature than with elevated [CO2]. © 2015 Elsevier B.V. All rights reserved.