Interpretable AI predicts a 2026 summer dry anomaly in central China
2026-08-19 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors use a deep learning model to convert atmospheric circulation predictions into forecasts of rainfall. Their model predicted less rain over central China in summer 2026, starting from weather data in spring. They found that certain patterns in the Pacific Ocean, especially warm waters, cause winds that reduce moisture and rain in that area. They also used a method to explain the model’s decisions and confirmed that these winds are the key factor in causing less rain. This approach helps understand AI climate predictions before real observations come in.
atmospheric circulationprecipitation anomaliesdeep learningcentral Chinaequatorial Pacific warmingcyclonic circulationnortherly windsmoisture divergencelayer-wise relevance propagationclimate prediction
Authors
Anran Wang, Wen Shi, Yong Luo, Jianbin Huang, Lijuan Chen, Junhu Zhao, Weixin Jin, Huihui Yuan
Abstract
Seasonal precipitation anomalies are largely regulated by atmospheric circulation, which dynamical models predict with greater reliability than precipitation itself. Here, we employ a deep learning model that translates dynamical circulation predictions into precipitation estimates. Predictions initialized from March to May consistently indicate a dry anomaly over central China in summer 2026. Retrospective evaluations revealed higher predictive skill in the analogue years, which also tended to feature central equatorial Pacific warming persisting from the preceding winter into summer. This warming favors an anomalous cyclonic circulation over the western North Pacific-South China Sea-South China region, which induces northerly winds and moisture divergence that jointly suppress rainfall over central China. Supporting this mechanism, layer-wise relevance propagation (LRP) independently identifies these northerly winds as the dominant driver of the prediction among all model inputs. Perturbation tests supported this attribution: removing LRP-identified features effectively eliminates the dry anomaly. Our framework thus provides physically interpretable explanations for AI-derived regional climate projections, facilitating evidence-based assessment before observational data become available.