Regional yield simulations with process-based models often rely on input data of coarse spatial resolution (Ewert et al., 2015; Zhao et al., 2015). Using aggregated data as input for process-based models entails the risks of introducing so-called aggregation errors (AE). Such AE depend on the model structure in combination with the aggregation method, the type of aggregated data as well as its spatial heterogeneity. While the regional crop yield bias is usually