Risk of crop-failure due to compound hot and dry extremes in the Iberian Peninsula

Crop health and favourable yields depend strongly on precipitation and temperature patterns during the crop’s growing season. Compound events, such as co-occurring drought and heat can lead to extreme crop failure and cause larger damages than the impacts of the individual drought or heat alone.

Here we assess the relative role of hot and dry conditions (HDC) in crop yields and evaluate in what manner compound HDC enhance the probability of failure in rainfed cropping systems in the Iberian Peninsula. We use annual wheat yield data at the province level and cluster provinces with similar sensitivities of yields to climate conditions. Copula theory was applied to model the trivariate dependence between 3-monthly means of maximum temperature, 3-monthly means of precipitation and wheat yields. The climate variables and averaging periods have been chosen to maximize the dependence between the driver climate conditions during growing season and the annual yields. Copulas enable for the estimation of conditional probabilities of crop-loss under different hot and dry severity levels based on their trivariate joint distribution.

Our results demonstrate that the probability of wheat loss increases with the severity of the compound HDC and that losses are significantly larger during co-occurring drought and heat compared to individual water- or heat-stress. Moreover, the difference between heat impacts and compound heat and drought related impacts is larger than the difference between drought impacts and compound heat and drought related impacts, suggesting that water-stress is the major driver of wheat losses. These findings can help contribute to design management options to mitigate climate-related crop impacts and guide the decision-making process in agricultural practices.

Acknowledgements: A.F.S.Ribeiro would like to acknowledge the financial support through FCT (Fundação para a Ciência e a Tecnologia, Portugal) under the projects UIDB/50019/2020 – IDL and PTDC/CTA-CLI/28902/201 (IMPECAF). A.F.S.Ribeiro is also thankful to FCT for the grant PD/BD/114481/2016 and to the COST Action CA17109 for a Short Term Scientific Mission (STSM) grant to develop the present work.