<p>Multiple climate extremes can coincide in time or happen sequentially and become a compound hazard. Recently, frequent occurrences of successive hot-pluvial extremes (SHPEs) have been presented, yet the risk of SHPEs on crop yield has not been investigated. Here we reveal peril of recurrent occurrences of SHPEs during intra-growing season and their connection with subsequent yield loss over breadbasket regions. Our results show an increasing trend in evolution of recurrent SHPEs during intra-growing seasons from 1979 to 2024. A significant risk of synchronized low crop yields is found in breadbasket regions, as shown by negative yield percentage changes and linear regression analysis. XGBoost classifier model predicts a negative likelihood of 49%, 50%, 49%, and 50% for maize, rice, soybean, and wheat yield responses to the emergence of frequent SHPE events, respectively. Alternatively, XGBoost regression surpasses and explains 36% of yield variability caused by recurrent occurrences of SHPEs globally for the studied crops, although it exhibits heterogeneities across different regions. The insights provide a foundation for considering the actual risk, which ultimately contributes to improving crop yield.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Risk of successive hot-pluvial extremes on crop yield loss over global breadbasket regions

  • Abebe K. Kabtih,
  • Cheng Qian

摘要

Multiple climate extremes can coincide in time or happen sequentially and become a compound hazard. Recently, frequent occurrences of successive hot-pluvial extremes (SHPEs) have been presented, yet the risk of SHPEs on crop yield has not been investigated. Here we reveal peril of recurrent occurrences of SHPEs during intra-growing season and their connection with subsequent yield loss over breadbasket regions. Our results show an increasing trend in evolution of recurrent SHPEs during intra-growing seasons from 1979 to 2024. A significant risk of synchronized low crop yields is found in breadbasket regions, as shown by negative yield percentage changes and linear regression analysis. XGBoost classifier model predicts a negative likelihood of 49%, 50%, 49%, and 50% for maize, rice, soybean, and wheat yield responses to the emergence of frequent SHPE events, respectively. Alternatively, XGBoost regression surpasses and explains 36% of yield variability caused by recurrent occurrences of SHPEs globally for the studied crops, although it exhibits heterogeneities across different regions. The insights provide a foundation for considering the actual risk, which ultimately contributes to improving crop yield.