Start Date
6-7-2022 7:20 AM
End Date
6-7-2022 7:40 AM
Abstract
We analyse the relationship between variations in crop yields across Belgium and dry (wet) spell metrics during the crop growing season. More specifically, we consider the number of consecutive dry (wet) days, and the total amount of rainfall during consecutive wet days. Dry (wet) days were defined as days with less (more) than 0.2 mm. A two-way ANOVA is employed to indicate statistically significant relationships between dry/wet spell metrics and regional crop yields. We make an extreme value analysis of the dry and wet spell metrics and estimate 20-year return levels, relevant to agricultural insurances and risk management strategies. We produce return level maps on the basis of the theory of spatial extremes, which naturally extends classical univariate extreme value distributions and encompasses key elements of conventional Gaussian geostatistics for non-extremes. As an application, we examine the extreme wet year 2016 and the extreme dry year 2018. Yield reductions for the year 2016 amounts to 39% for winter cereals, 45% for maize and 53% for potato, while 2018 resulted in losses of 74% for maize and 42% for potato. The exceptionality of both years is confirmed by spatial return periods of the order of several hundred years. Our framework offers perspectives for agricultural risk management and weather-based insurances.
Spatio-temporal variability of dry and wet spells and their influence on crop yields
We analyse the relationship between variations in crop yields across Belgium and dry (wet) spell metrics during the crop growing season. More specifically, we consider the number of consecutive dry (wet) days, and the total amount of rainfall during consecutive wet days. Dry (wet) days were defined as days with less (more) than 0.2 mm. A two-way ANOVA is employed to indicate statistically significant relationships between dry/wet spell metrics and regional crop yields. We make an extreme value analysis of the dry and wet spell metrics and estimate 20-year return levels, relevant to agricultural insurances and risk management strategies. We produce return level maps on the basis of the theory of spatial extremes, which naturally extends classical univariate extreme value distributions and encompasses key elements of conventional Gaussian geostatistics for non-extremes. As an application, we examine the extreme wet year 2016 and the extreme dry year 2018. Yield reductions for the year 2016 amounts to 39% for winter cereals, 45% for maize and 53% for potato, while 2018 resulted in losses of 74% for maize and 42% for potato. The exceptionality of both years is confirmed by spatial return periods of the order of several hundred years. Our framework offers perspectives for agricultural risk management and weather-based insurances.
Stream and Session
false