Sensitivity of field-scale winter wheat production in Denmark to climate variability and climate change

A simulation model of the direct effects of climate on winter wheat production and grain yield is presented. The model was calibrated using data from field experiments in Denmark. The model was validated using data from near optimally managed experimental plots with winter wheat from The Netherlands and Denmark. The model was further evaluated using data from 1971 to 1997 for 7 sites in Denmark. The model explained from 0 to 20% of the variation in detrended observed yields, depending on soil type. A regression analysis of observed yields against monthly climate data showed a positive effect of temperature in October, November and January on grain yield, a positive effect of radiation in April and a strongly negative effect of precipitation in July. Only the positive effect of radiation in April was predicted by the simulation model, probably because the indirect effects of climate are not taken into account by the model (e.g. effects of rainfall on lodging or Septoria disease). The sensitivity of simulated grain yield to changes in mean temperature, temperature variability, precipitation, length of dry spells and CO2 concentration was analysed for 4 soil types using generated climate data from 1 site in Denmark. Yield decreased with increasing temperature. This decrease was strongly non-linear with temperature: change when using a fixed sowing date, but almost linear for the optimal sowing date. There was only a very small response to changes in temperature variability. Increasing precipitation increased yields with the largest response on the sandy soils. Large changes in grain yield were also seen on sandy soils with changes in the length of dry spells. A comparison of the simulated responses to the direct effects of temperature and rainfall with those to the indirect effects of these variables as estimated from the regression analysis showed that the indirect and the direct effects had opposite effects and that they may almost cancel each other out. The simulated increase in grain yield due to increasing CO2 concentration in most cases exceeded the simulated responses to changes in climate variables.