Uncertainties in Scaling Up Crop Models for Large Area Climate Change Impact Assessments.

Crop models developed for field scale application are increasingly applied at larger spatial and temporal scales for which they have not originally been developed. Several methods exist to scale up crop models from the field to larger areas of a region, country, continent, or globe but uncertainties associated with these different methods are largely unclear and are the focus of this paper. Particular emphasis is on the impact of the different scaling methods on simulated yield under present and future climate conditions. Attention is also given to the interaction between scaling method and crop models used, as we hypothesize that different crop models will respond differently to the applied scaling methods.

The study provides an overview of results from recent studies on this topic and presents first results of a scaling exercise performed within the AgMIP (The Agricultural Model Intercomparison and Improvement Project). These results will allow better quantification of uncertainties in crop model application for large area climate change impact assessment due to the used crop model and method for up scaling. Although main emphasis of the paper is on wheat and selected regions mainly in Europe, generalization of the results obtain for different crops and regions across the world is discussed.