As the crucial tool for understanding the responses of crops to environmental changes, crop models are widely used in predicting crop yields and assessing risks of climate change. However, the complex model structure and parametrizations are major sources of uncertainty in predicting crop yield. Recent studies have shown that Eco-Evolutionary Optimality (EEO) can be used as a model constraint, which not only simplifies the structure and parameters of natural vegetation models, but also naturally accounts for the acclimation/adaptation processes of plants. To reliably assess the responses of global wheat yields to climate change, this study developed a new model for wheat constrained by the EEO principle, and then, applied it to explore the effects of climate change on global wheat yields and sowing dates. In this study, starting from an EEO-based productivity model (P model), we firstly developed a new model for wheat growth, named PC model - the Productivity model for Crops. The PC model considers the acclimation/adaptation of photosynthetic capacity and leaf area index, and predicts wheat yield parsimoniously. This PC model was then applied to predict potential wheat yields globally and diagnose the impacts of climate change and elevated CO2 concentration. Finally, we applied the PC model to predict the optimal sowing date corresponding to maximum wheat yield and investigated the long-term adaptation of sowing activity to climatology at the global scale. The results show that the EEO-based PC model can well predict wheat growth at the site level. It successfully accounts for the acclimation/adaptation processes of wheat and largely simplifies the model structure and tunable parameters. The PC model can well capture the global pattern of potential wheat yield, and has better performance than some complex process-based crop models. The changes in global wheat yields are the consequences of the combination of changes in climate and CO2 concentration, with warming reducing wheat yields in almost all regions, while CO2 enrichment promoting them. The global pattern of wheat sowing dates reflects the adaptation of sowing activities to long-term climate regimes for maximizing local wheat yields. Temperature is the first consideration for wheat sowing in extra-tropical regions, while rainfall intensity plays an important role in monsoon regions. Overall, the PC model provides a new tool for predicting wheat yield and assessing the impacts of climate change. The proposed scheme for predicting wheat sowing dates can potentially provide a scientific basis for the development of climate change adaptation strategies.
