Scholarlink

<p>The East Asian summer monsoon (EASM) rainfall often causes casualties and property damage in the region. Although there is a long history of studying and predicting EASM rainfall, prediction skill remains low. In this work, we identify the leading coupled modes between tropical sea surface temperature (SST) and EASM rainfall variations in observations and compare them with the corresponding modes in model predictions. Lastly, we propose a hybrid approach from a coupled-mode perspective to enhance the prediction skill of EASM rainfall. Two leading modes are dominant in the observations. One is associated with warming trends in tropical oceans and rainfall increases over eastern China and Japan. The other represents the influence of the El Niño-Southern Oscillation (ENSO). A climate model (CFSv2) only partially captures the temporal evolution and spatial patterns of the observed two leading modes with noticeable deficiencies. These biases may be part of the reason for the low prediction skill. We combine the trends and ENSO impacts in the observations with model-predicted ENSO index and construct a hybrid approach that exhibits skill exceeding the CFSv2 predictions initialized in May. This encouraging result implies potential value in the operational application of this hybrid approach.</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

Challenges and a hybrid approach in predicting East Asian summer rainfall: a coupled-mode perspective

  • 期刊论文

  • 发表日期：

    2026年7月2日

查看全文
  • Wei Tan,
  • Zeng-Zhen Hu,
  • Yunyun Liu,
  • Ping Liang,
  • Renguang Wu

摘要

The East Asian summer monsoon (EASM) rainfall often causes casualties and property damage in the region. Although there is a long history of studying and predicting EASM rainfall, prediction skill remains low. In this work, we identify the leading coupled modes between tropical sea surface temperature (SST) and EASM rainfall variations in observations and compare them with the corresponding modes in model predictions. Lastly, we propose a hybrid approach from a coupled-mode perspective to enhance the prediction skill of EASM rainfall. Two leading modes are dominant in the observations. One is associated with warming trends in tropical oceans and rainfall increases over eastern China and Japan. The other represents the influence of the El Niño-Southern Oscillation (ENSO). A climate model (CFSv2) only partially captures the temporal evolution and spatial patterns of the observed two leading modes with noticeable deficiencies. These biases may be part of the reason for the low prediction skill. We combine the trends and ENSO impacts in the observations with model-predicted ENSO index and construct a hybrid approach that exhibits skill exceeding the CFSv2 predictions initialized in May. This encouraging result implies potential value in the operational application of this hybrid approach.