Abstract: Local-scale daily climate scenarios are required for assessment of climate change impacts. ELPIS is a repository of local-scale climate scenarios for Europe, which are based on the LARS-WG weather generator and future projections from 2 multi-model ensembles, CMIP3 and EU-ENSEMBLES. In ELPIS, the site parameters for the 1980-2010 baseline scenarios were estimated by LARS-WG using daily weather from the European Crop Growth Monitoring System (CGMS) used in many European agricultural assessment studies. The objective of this paper was to compare ELPIS baseline scenarios with observed daily weather obtained independently from the European Climate Assessment (ECA) data set. Several statistical tests were used to compare distributions of climatic variables derived from ECA-observed daily weather and ELPIS-generated baseline scenarios. About 30% of selected sites have a difference in altitude of > 50 m compared with the CGMS grid-cell altitude that was selected to represent agricultural land within a grid-cell. Differences in altitude can explain significant Kolmogorov-Smirnov test (KS-test) results for distribution of daily temperature and in t-tests for temperature monthly means, because of the well-known negative correlation between temperature and elevation. For daily precipitation, the KS-test showed little difference between generated and observed data; however, the more sensitive t-test showed significant results for the sites where altitude differences were large. Approximately 11% of sites showed small positive or negative bias in monthly solar radiation, although 86% sites showed > 3 significant t-test results for monthly means. These results can be explained by differences in conversion of sunshine hours to solar radiation used in CGMS and LARS-WG. We conclude that, considering the limitations above, ELPIS baseline scenarios are suitable for agricultural impact assessments in Europe.

Abstract: AimTo derive location-specific parameters that reflect the geographic differences among cultivars in vernalization requirements, sensitivity to day length (photoperiod) and temperature, which can be used to simulate the phenological development of wheat and maize at the global scale. LocationGlobal. Methods Based on crop calendar observations and literature describing the large-scale patterns of phenological characteristics of cultivars, we developed algorithms to compute location-specific parameters to represent this large-scale pattern. Vernalization requirements were related to the duration and coldness of winter, sensitivity to day length was assumed to be represented by the minimum and maximum day lengths occurring at a location, and sensitivity to temperature was related to temperature conditions during the vegetative development phase of the crop. Results Application of the derived location-specific parameters resulted in high agreement between simulated and observed lengths of the cropping period. Agreement was especially high for wheat, with mean absolute errors of less than 3 weeks. In the main maize cropping regions, cropping periods were over- and underestimated by 0.5-1.5 months. We also found that interannual variability in simulated wheat harvest dates was more realistic when accounting for photoperiod effects. Main conclusions The methodology presented here provides a good basis for modelling the phenological characteristics of cultivars at the global scale. We show that current global patterns of growing season length as described in cropping calendars can be largely reproduced by phenology models if location-specific parameters are derived from temperature and day length indicators. Growing seasons can be modelled more accurately for wheat than for maize, especially in warm regions. Our method for computing parameters for phenology models from temperature and day length offers opportunities to improve the simulation of crop productivity by crop simulation models developed for large spatial areas and for long-term climate impact projections that account for adaptation in the selection of varieties

Abstract: Regional agricultural yield assessments allowing for weather effect quantifications are a valuable basis for deriving scenarios of climate change effects and developing adaptation strategies. Assessing weather effects by statistical methods is a classical approach, but for obtaining robust results many details deserve attention and require individual decisions as is demonstrated in this paper. We evaluated regression models for annual yield changes of winter wheat and silage maize in more than 300 German counties and revised them to increase their predictive power. A major effort of this study was, however, aggregating separately estimated time series models (STSM) into panel data models (PDM) based on cluster analyses. The cluster analyses were based on the per-county estimates of STSM parameters. The original STSM formulations (adopted from a parallel study) contained also the non-meteorological input variables acreage and fertilizer price. The models were revised to use only weather variables as estimation basis. These consisted of time aggregates of radiation, precipitation, temperature, and potential evapotranspiration. Altering the input variables generally increased the predictive power of the models as did their clustering into PDM. For each crop, five alternative clusterings were produced by three different methods, and similarities between their spatial structures seem to confirm the existence of objective clusters about common model parameters. Observed smooth transitions of STSM parameter values in space suggest, however, spatial autocorrelation effects that could also be modeled explicitly. Both clustering and autocorrelation approaches can effectively reduce the noise in parameter estimation through targeted aggregation of input data. (C) 2015 Elsevier B.V. All rights reserved.