Abstract: The available evidence suggests that the current increasing trend in global surface temperatures will continue during this century, which will be accompanied by a greater frequency of extreme events. The IPCC has projected that higher temperatures may outscore the known optimal and maximum temperatures for maize. The purpose of this study was to improve the ability of the maize model CSM-IXIM to simulate crop development, growth, and yield under hot conditions, especially with regards to the impact of above-optimal temperatures around anthesis. Field and greenhouse experiments that were performed over three years (2014-2016) using the same short-season hybrid, PR37N01 (FAO 300), provided the data for this work. Maize was sown at a target population density of 5 plants M-2 on two sowing dates in 2014 and 2015 and on one in 2016 at three locations in Spain (northern, central, and southern Spain) with a well-defined thermal gradient. The same hybrid was also sown in two greenhouse chambers with daytime target temperatures of approximately 25 and above 35 degrees C. During the nighttime, the temperature in both chambers was allowed to equilibrate with the outside temperature. The greenhouse treatments consisted of moving 18 plants at selected phenological stages (V4, V9, anthesis, lag phase, early grain filling) from the cool chamber to the hot chamber over a week and then returning the plants back to the cool chamber. An additional control treatment remained in the cool chamber all season, and in 2015 and 2016, one treatment remained permanently in the hot chamber. Two maize models in the Decision Support System for Agrotechnology Transfer (DSSAT) V4.6 were compared, namely CERES and IXIM. The HUM version included additional components that were previously developed to improve the crop N simulation and to incorporate the anthesis-silking interval (ASI). A new thermal time calculation, a heat stress index, the impact of pollen-sterilizing temperatures, and the explicit simulation of male and female flowering as affected by the daily heat conditions were added to IXIM. The phenology simulation in field experiments by IXIM improved substantially. The RMSE for silking and maturity in CERES were 7.9 and 13.7 days, decreasing in DCIM to 2.8 and 7.3 days, respectively. Similarly, the estimated kernel numbers, kernel weight, grain yield and final biomass were always closer to the measurements in HUM than in CERES. The worst simulations were for kernel weight, and for that reason, the differences in grain yield between the models were small (the RMSE in CERES was 1219 kg ha(-1) vs. 1082 kg ha(-1) in IXIM). The greenhouse results also supported the improved estimations of crop development by IXIM (RMSE of 2.6 days) relative to CERES (7.4 days). The impact of the heat treatments on grain yield was consistently overestimated by CERES, while HUM captured the general trend. The new HUM model improved the CERES simulations when elevated temperatures were included in the evaluation data. Additional model testing with measurements from a wider latitudinal range and relevant heat conditions are required.

Abstract: During the past few years, several studies have compared the performance of crop simulation models to assess the uncertainties in model-based climate change impact assessments and other modelling studies. Many of these studies have concentrated on cereal crops, while fewer model comparisons have been conducted for grasses. We compared the predictions for timothy grass (Phleum pratertse L.) yields for first and second cuts along with the dynamics of above-ground biomass for the grass simulation models BASGRA and CATIMO, and the soil -crop model STICS. The models were calibrated and evaluated using field data from seven sites across Northern Europe and Canada with different climates, soil conditions and management practices. Altogether the models were compared using data on timothy grass from 33 combinations of sites, cultivars and management regimes. Model performances with two calibration approaches, cultivar-specific and generic calibrations, were compared. All the models studied estimated the dynamics of above-ground biomass and the leaf area index satisfactorily, but tended to underestimate the first cut yield. Cultivar-specific calibration resulted in more accurate first cut yield predictions than the generic calibration achieving root mean square errors approximately one third lower for the cultivar-specific calibration. For the second cut, the difference between the calibration methods was small. The results indicate that detailed soil process descriptions improved the overall model performance and the model responses to management, such as nitrogen applications. The results also suggest that taking the genetic variability into account between cultivars of timothy grass also improves the yield estimates. Calibrations using both spring and summer growth data simultaneously revealed that processes determining the growth in these two periods require further attention in model development.

Abstract: In this study, the performance of nine widely used and accessible crop growth simulation models (APES-ACE, CROPSYST, DAISY, DSSAT-CERES, FASSET, HERMES, MONICA, STICS and WOFOST) was compared during 44 growing seasons of spring barley (Hordeum vulgare L) at seven sites in Northern and Central Europe. The aims of this model comparison were to examine how different process-based crop models perform at multiple sites across Europe when applied with minimal information for model calibration of spring barley at field scale, whether individual models perform better than the multi-model mean, and what the uncertainty ranges are in simulated grain yields. The reasons for differences among the models and how results for barley compare to winter wheat are discussed. Regarding yield estimation, best performing based on the root mean square error (RMSE) were models HERMES, MONICA and WOFOST with lowest values of 1124, 1282 and 1325 (kg ha(-1)), respectively. Applying the index of agreement (IA), models WOFOST, DAISY and HERMES scored best having highest values (0.632, 0.631 and 0.585, respectively). Most models systematically underestimated yields, whereby CROPSYST showed the highest deviation as indicated by the mean bias error (MBE) (-1159 kg ha(-1)). While the wide range of simulated yields across all sites and years shows the high uncertainties in model estimates with only restricted calibration, mean predictions from the nine models agreed well with observations. Results of this paper also show that models that were more accurate in predicting phenology were not necessarily the ones better estimating grain yields. Total above-ground biomass estimates often did not follow the patterns of grain yield estimates and, thus, harvest indices were also different. Estimates of soil moisture dynamics varied greatly. In comparison, even though the growing cycle for winter wheat is several months longer than for spring barley, using RMSE and IA as indicators, models performed slightly, but not significantly, better in predicting wheat yields. Errors in reproducing crop phenology were similar, which in conjunction with the shorter growth cycle of barley has higher effects on accuracy in yield prediction. (C) 2012 Elsevier B.V. All rights reserved.