Definition of agroclimatic regions in Ireland using hydro-thermal and crop yield data

When compared to the agronomic aspects of agriculture the contribution of climate is poorly integrated into the body of scientific knowledge in Ireland. It is known that specific crops grow well in specific regions and that success of a crop can be related to climate factors. The aim of this paper was to derive agroclimatic regions of Ireland using hydro-thermal climate (as expressed by climographs using mean monthly rainfall and temperature) in conjunction with a statistical clustering technique ( k-means clustering) to relate crop yield, estimated using mathematical simulation models for grass, barley, maize, potato and soybean to hydro-thermal climate data. A dataset, on a 10 km×10 km grid, consisting of monthly radiation, rainfall, maximum temperature and minimum temperature was used to drive crop simulation models to predict average yields, for each grid square. Results showed that the crops simulated were sensitive to either available water or length of growing season. As there are few objective measures of the ideal number of clusters to use in order to create agroclimatic regions, clusterings to 3, 4 and 7 regions were evaluated against known properties of the Irish climate and crop responses. A seven cluster agroclimate region map was thought to be a good basis for describing Ireland’s agroclimates and was compared with climate data for the national synoptic station network. Results suggested a reasonably good correspondence between clusters and their associated synoptic observation stations. Some anomalies were observed. The resulting agroclimatic map could be used for extrapolation of empirical research findings, for agri-environmental experimental design and as a framework for assessment of the impact of climate change on Irish agriculture.