Abstract: The Northeast Farming Region (NFR) is the most important and the largest rain-fed maize production region in China, accounting for 30% of China’s maize. We investigated the effects of climatic factors, drought risk and irrigation requirement on maize yield in different maize growth phases during 1961 to 2010 using a statistical analysis of observed yield from 44 stations in NFR. We divided the maize growing season into four growth phases, comprising seeding, vegetative, flowering and maturity. The dual crop coefficient was used to calculate crop evapotranspiration and soil water balance during the maize growing season. The effects of mean temperature, radiation, effective rainfall, water deficit, drought stress days, actual crop evapotranspiration (ETa) and irrigation requirement in different growth phases were included in the statistical model to predict maize yield. During the period 1961 to 2010, mean temperature increased significantly in all growth phases in NFR, while radiation decreased significantly in southern NFR in the seeding, vegetative and flowering phases. Effective rainfall increased in the seeding and vegetative phases leading to less water deficit, whereas decreased effective rainfall in the flowering and maturity phases enhanced water deficit. More days with drought stress were concentrated in western NFR where larger volumes of irrigation were needed. Our results indicate that the increase of mean temperature in the seeding and maturity phases was beneficial for maize yield, higher ETa in each growth phase would lead to yield increase, but too high rainfall would damage maize yield. The results also show that water deficit and drought stress days had significant negative effects on maize yield, and the absence of irrigation would manifest such effects on maize production in NFR. Therefore, the development of irrigation and drainage systems is highly needed for ensuring the stability of maize production in NFR. In addition, other adaptation measures like introducing new cultivars and optimizing soil and crop management to better conserve soil water would be beneficial for future maize production. No Label

Abstract: Climate and soil data at coarse resolution are often used as input for crop models in order to simulate crop yields at larger scales, e.g. at regional or national level, potentially leading to biased yield estimates. While the response to data resolution differs between crop models, it is unknown how the spatial aggregation of different types of input data interacts and contributes to this so-called aggregation effect. An ensemble of crop models was run with soil and climate input data at different spatial resolutions from 1 to 100 km for the state of North Rhine-Westphalia, Germany. For this purpose, climate time series were averaged spatially and soil data was aggregated by selecting the dominant soil type with a representative soil profile based on a soil map at the scale of 1:50,000. Yields of winter wheat and silage maize were simulated under potential, water-limited and water-nitrogen-limited production conditions. Crop yields from soil and climate aggregation were evaluated separately.Mean of crop yields of the region and over the simulation period were reasonably reproduced by most models regardless of input data resolution, either using aggregated soil or climate as input. However, larger aggregation effects were observed at higher temporal resolution (e.g. annual yields). Models revealed similar spatial patterns in yield. Being distinct for soil and climate aggregation, these patterns indicate a larger impact of soil aggregation on the spatial distribution of simulated crop yield for this region. Additionally, models differed considerably in their susceptibility to input data aggregation. The results reveal the importance of model ensemble assessments and the relevance of data aggregation when short simulation periods are considered. No Label

Abstract: There have been several studies, where process-based crop models are developed, used and compared in order to project crop production and corresponding model uncertainties under climate change. Despite many advances in this field, there are some correlations between climate variables and crop growth, such as pest and diseases, that is often absent in process-based models. Such relationships can be simulated using empirical models. In this study, several statistical techniques were applied on winter oilseed rape data collected in some European countries. The empirical models were then used to predict yield of winter oilseed rape in the field experiments during more than 20 years, up to 2013. Results suggest that newly developed regression techniques such as shrinkage methods work well both in yield projections and finding the influential climatic variables. Many of regression techniques agree in terms of yield prediction; however, choice of significant climate variables is rather sensitive to the choice of regression technique. No Label