Divergent effectiveness of irrigation in enhancing food security in droughts under future climates with various emission scenarios

Abstract Food security is a key target in the UN Sustainable Development Goals (SDGs) and is also one of the biggest challenges for China, the largest developing country in population in the world. Massive attention has been directed to the future impacts of hydrometeorological extremes on crop yield. However, knowledge gaps still stand concerning the effectiveness of irrigation, as the largest water consumption sector, on agricultural production under different climate scenarios. Here we showed the drought based on the modified Palmer Drought Severity Index and analyzed the drought-induced wheat yield losses in irrigated and non-irrigated agriculture across China according to three methods including the Multiple Linear Regression method, Deep Learning algorithm, and Erosion-Productivity Impact Calculator model. We found that the droughts become more intensive in the future, and drought-induced wheat yield loss under RCP8.5 scenario was expected to reach 32–49%. Intercomparison of drought-induced wheat yield changes between irrigated and non-irrigated agriculture indicated significantly less drought-induced crop-yield losses given sufficient irrigation under the RCP2.6 and RCP6.0 scenarios. However, our results under RCP8.5 suggested that the effectiveness of irrigation in enhancing food security is minor under this high-emission future climate scenario. These findings allowed us to revisit the effectiveness of irrigation in a warming climate and highlighted the importance of climate change mitigation in food security.