Abstract: Justifiable usage of large-scale crop model simulations requires transparent, comprehensive and spatially extensive evaluations of their performance and associated accuracy. Simulated crop yields of a Pan-European implementation of the Environmental Policy Integrated Climate (EPIC) crop model were satisfactorily evaluated with reported regional yield data from EUROSTAT for four major crops, including winter wheat, rainfed and irrigated maize, spring barley and winter rye. European-wide land use, elevation, soil and daily meteorological gridded data were integrated in GIS and coupled with EPIC. Default EPIC crop and biophysical process parameter values were used with some minor adjustments according to suggestions from scientific literature. The model performance was improved by spatial calculations of crop sowing densities, potential heat units, operation schedules, and nutrient application rates. EPIC performed reasonable in the simulation of regional crop yields, with long-term averages predicted better than inter-annual variability: linear regression R-2 ranged from 0.58 (maize) to 0.91 (spring barley) and relative estimation errors were between +/- 30% for most of the European regions. The modelled and reported crop yields demonstrated similar responses to driving meteorological variables. However, EPIC performed better in dry compared to wet years. A yield sensitivity analysis of crop nutrient and irrigation management factors and cultivar specific characteristics for contrasting regions in Europe revealed a range in model response and attainable yields. We also show that modelled crop yield is strongly dependent on the chosen PET method. The simulated crop yield variability was lower compared to reported crop yields. This assessment should contribute to the availability of harmonised and transparently evaluated agricultural modelling tools in the EU as well as the establishment of modelling benchmarks as a requirement for sound and ongoing policy evaluations in the agricultural and environmental domains. (C) 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

Abstract: Apart from measurements, numerical models are the most convenient instruments to analyze the carbon and water balance of terrestrial ecosystems and their interactions with changing environmental conditions. The process-based Biome-BGC model is widely used to simulate the storage and flux of water, carbon, and nitrogen within the vegetation, litter, and soil of unmanaged terrestrial ecosystems. Considering herbaceous vegetation related simulations with Biome-BGC, soil moisture and growing season control on ecosystem functioning is inaccurate due to the simple soil hydrology and plant phenology representation within the model. Consequently, Biome-BGC has limited applicability in herbaceous ecosystems because (1) they are usually managed; (2) they are sensitive to soil processes, most of all hydrology; and (3) their carbon balance is closely connected with the growing season length. Our aim was to improve the applicability of Biome-BGC for managed herbaceous ecosystems by implementing several new modules, including management. A new index (heatsum growing season index) was defined to accurately estimate the first and the final days of the growing season. Instead of a simple bucket soil sub-model, a multilayer soil sub-model was implemented, which can handle the processes of runoff, diffusion and percolation. A new module was implemented to simulate the ecophysiological effect of drought stress on plant mortality. Mowing and grazing modules were integrated in order to quantify the functioning of managed ecosystems. After modifications, the Biome-BGC model was calibrated and validated using eddy covariance-based measurement data collected in Hungarian managed grassland ecosystems. Model calibration was performed based on the Bayes theorem. As a result of these developments and calibration, the performance of the model was substantially improved. Comparison with measurement-based estimate showed that the start and the end of the growing season are now predicted with an average accuracy of 5 and 4 days instead of 46 and 85 days as in the original model. Regarding the different sites and modeled fluxes (gross primary production, total ecosystem respiration, evapotranspiration), relative errors were between 18-60% using the original model and 10-18% using the developed model; squares of the correlation coefficients were between 0.02-0.49 using the original model and 0.50-0.81 using the developed model. (c) 2011 Elsevier B.V. All rights reserved.

Abstract: Field-scale crop models are increasingly applied at spatio-temporal scales that range from regions to the globe and from decades up to 100 years. Sufficiently detailed data to capture the prevailing spatio-temporal heterogeneity in weather, soil, and management conditions as needed by crop models are rarely available. Effective sampling may overcome the problem of missing data but has rarely been investigated. In this study the effect of sampling weather data has been evaluated for simulating yields of winter wheat in a region in Germany over a 30-year period (1982-2011) using 12 process-based crop models. A stratified sampling was applied to compare the effect of different sizes of spatially sampled weather data (10, 30, 50,100, 500, 1000 and full coverage of 34,078 sampling points) on simulated wheat yields. Stratified sampling was further compared with random sampling. Possible interactions between sample size and crop model were evaluated. The results showed differences in simulated yields among crop models but all models reproduced well the pattern of the stratification. Importantly, the regional mean of simulated yields based on full coverage could already be reproduced by a small sample of 10 points. This was also true for reproducing the temporal variability in simulated yields but more sampling points (about 100) were required to accurately reproduce spatial yield variability. The number of sampling points can be smaller when a stratified sampling is applied as compared to a random sampling. However, differences between crop models were observed including some interaction between the effect of sampling on simulated yields and the model used. We concluded that stratified sampling can considerably reduce the number of required simulations. But, differences between crop models must be considered as the choice for a specific model can have larger effects on simulated yields than the sampling strategy. Assessing the impact of sampling soil and crop management data for regional simulations of crop yields is still needed.