Abstract: This review examines relevant impact assessments identified by a literature search from 1991to date. A bibliographic search was applied to the CAB Abstracts database with a given searchstring. Resultant papers were checked for relevance, based on expert judgment. This yielded 91 papers, which were subjected to further analysis. Firstly, publication intensity over time and distribution by geographic location and cereal crop were examined. Next, for a given crop, the assessments and their outcomes were grouped by type and number of the change variables considered – that is, effects of climate change only, elevated CO 2 and technological progress(improved breeds, management). Finally, separately for individual countries/subregions and Europe as a whole, we examined whether and to what extent study results have changed over time, for example become more positive/negative. Based on our sample, we found that publication intensity increased exponentially during thelast 4 years, the majority of studies are Europe-wide, but some concentrated on a few countries(Italy, Spain and UK), whereby studies on wheat are clearly most popular. Taking the factor of technological progress into account has an overruling influence on results. Finally, over time, projected yield impacts have become more negative. This is in line with finding from global analyses, as reflected by the most recent comparison of agricultural impact chapters, of the 4thand 5th Assessment Reports of Intergovernmental Panel on Climate Change, Working Group II.In the future, there is particular need to consider impacts under various incremental and transformational adaptation measures in more depth (e.g. their interconnections across scales)and with more breadth (e.g. anticipated new breeds). Follow-up reviews should also examine how projected impacts are changing with the new climate scenario data sets (CMIP5) and with improved impact models and assessment approaches.

Abstract: Various attempts have been made to determine the relative importance of uncertainties in climate change impact assessments stemming from climate projections and crop models, respectively, and to analyse yield outputs probabilistically. For example, in the ENSEMBLES project, probabilistic climate projections (Harris et al. 2010) have been applied in conjunction with impact response surfaces (IRS), constructed by using impact models, to estimate the future likelihood (risk) of exceeding critical thresholds of crop yield impact (see, Fronzek et al., 2011, for an explanation of the method). In this task, we aimed to further develop and operationalize these methods and testing them in different case study regions in Europe. The method combines results of a sensitivity analysis of (one or more) impact model(s) with probabilistic projections of future temperature and precipitation (Fronzek et al., 2011). Such an overlay is one way of portraying probabilistic estimates of future impacts. By further accounting for the uncertainties in crop and biophysical parameters (using perturbed parameter approaches), the outcome represents an ensemble of impact risk estimates, encapsulating both climate and crop model uncertainties. No Label

Abstract: Data of diverse crop rotations from five locations across Europe were distributed to modelers to investigate the capability of models to handle complex crop rotations and management interactions. Crop rotations comprise various main crops (winter/spring wheat, winter/spring barley, rye, oat, maize, sugar beet, oil seed rape and potatoes) plus several catch crops. The experimental setup of the datasets included treatments such as modified soils, crops exchanged within the rotations, irrigation/rainfed, nitrogen fertilization, residue management, tillage and atmospheric CO2 concentration. 19 modeling teams registered to model either the whole rotation or single crops. Models which are capable to run the whole rotation should provide transient as well as single year simulations with a reset of initial conditions. In the first step only initial soil conditions (water and soil mineral N) of the first year and key phenological stages were provided to the modelers. For calibration, crop yields and biomass were provided for selected years but not for all seasons. In total the combination of treatments and seasons results in 301 years of simulation. Results were analyzed to evaluate the effect of transient simulation versus single-year simulation regarding crop yield, biomass, water and nitrogen balance components. Model results will be evaluated crop-specifically to identify crops with highest uncertainty and potential for model improvement. Full data will be provided to modelers for model-improvement and results will provide insights into model capabilities to reproduce treatments and crops. Further, the question of error propagation along the transient simulation of crop rotations will be addressed.