Abstract: Crop models are important tools for impact assessment of climate change, as well as for  exploring management options under current climate. It is essential to evaluate the  uncertainty associated with predictions of these models. Several ways of quantifying  prediction uncertainty have been explored in the literature, but there have been no  studies of how the different approaches are related to one another, and how they are  related to some overall measure of prediction uncertainty. Here we show that all the  different approaches can be related to two different viewpoints about the model; either  the model is treated as a fixed predictor with some average error, or the model can be  treated as a random variable with uncertainty in one or more of model structure, model  inputs and model parameters. We discuss the differences, and show how mean squared  error of prediction can be estimated in both cases. The results can be used to put  uncertainty estimates into a more general framework and to relate different uncertainty  estimates to one another and to overall prediction uncertainty. This should lead to a  better understanding of crop model prediction uncertainty and the underlying causes of  that uncertainty. This study was published as (Wallach et al. 2016)

Abstract: The need to answer new scientific questions can be satisfied by an increased knowledge of physiological mechanisms which, in turn, can be used for improving the accuracy of simulations of process-based models. In this context, this report highlights areas that need to be further improved to facilitate the operational use of simulation models. It describes missing approaches within simulation models which, if implemented, would likely improve the representation of the dynamics of processes underlying different compartments of crop and grassland systems (e.g. plant growth and development, yield production, GHG emissions), as well as of the livestock production systems.  The following rationale has been used in the organization of this report. We first briefly introduced the need to improve the reliability of existing models. Then, we indicated climate change and its influence on the global carbon balance as the main issue to be addressed by existing crop and grassland (section 2), and livestock (section 3) models. In section 2, among the major aspects that if implemented may reduce the uncertainty inherent to model outputs, we suggested: i) quantifying the effects of climate extremes on biological systems; ii) modelling of multi-species sward; iii) coupling of pest and disease sub-models; iv) improvement of the carry-over effect. In section 3, as the most important aspects to consider in livestock models we indicated: i) impacts and dynamics of pathogens and disease; ii) heat stress effects on livestock; iii) effects on grassland productivity and nutritional values; iv) improvement of GHG emissions dynamics.  In Section 4, remarks are made concerning the need to implement the suggested aspects into the existing models.

Abstract: Priorities in addressing research gaps and challenges should follow the order of im­por­tance, which in itself would be a matter of defining goals and metrics of importance, e.g. the extent, impact and likelihood of occurrence. For improving assessments of cli­mate change impacts on agriculture for achieving food security and other sustainable develop­ment goals across the European continent, the most important research gaps and challen­ges appear to be the agreement on goals with a wide range of stakeholders from policy, science, producers and society, better reflection of political and societal prefer­ences in the modelling process, and the reflection of economic decisions in farm manage­ment within models. These and other challenges could be approached in phase 3 of MACSUR.