Validation: an objective or a tool? Results on a winter wheat simulation model application

Abstract

The present paper describes a validation exercise for the CERES-Wheat simulation model using the SCP 179 data set provided by the research team of Technical University of Braunschweig (Germany). The data set used is a common data set to facilitate the performance of different crop growth simulation models. The present paper carries out a validation exercise with the aim of illustrating the capabilities of CERES-Wheat in a West European scenario, and, on the other hand, to illustrate the use of some statistical methods in model validation. Four simulation scenarios are set up with altogether two different soil types and two weather data sets for a winter wheat variety called Kanzler. The goodness of fit between model outputs and field observations is assessed for anthesis and maturity dates, biomass and grain production, leaf area index development and soil water and nutrient balance. A sensitivity analysis is included to illustrate the model's special facility to make distinctions between wheat varieties of different characteristics. Full validation was not possible to be carried out due to the relatively limited number of observations available, thus the paper is an exercise in validation methodology rather than a complete assessment of the model's performance and validity.