Attribution of Changes in Key Circulation Patterns Driving Summer Compound Hot–Dry Events over the Middle and Lower Reaches of the Yangtze River

Under the background of global warming, compound hot-dry events (CHDEs) in the middle and lower reaches of the Yangtze River have become increasingly severe. However, the respective contributions of atmospheric circulation and human activities to these events have not been systematically quantified. Based on CN05.1 observations, ERA5 reanalysis data, and CMIP6 model outputs, this study uses a Copula-based compound hot-dry index to describe the severity of CHDEs. The impacts of atmospheric circulation and background warming are quantified using a circulation-analogue approach. In addition, the Self-Organizing Map (SOM) method is applied to identify key circulation patterns associated with CHDEs and to explore the influence of human activities on their frequency. The main results are as follows: (1) Since 1961, the severity of summer CHDEs in the middle and lower reaches of the Yangtze River has increased, showing a significant upward trend since the 1980s as climate warming has intensified. (2) The “East Asia double-blocking-dipole” circulation pattern—characterized by the joint strengthening of the mid–high-latitude blocking ridge and the westward extension and northward shift of the Western Pacific Subtropical High (WPSH)—enhances subsidence, warming, and moisture deficit over the region, favoring the occurrence of CHDEs. (3) The circulation-analogue analysis reveals that such typical circulation patterns dynamically cause higher temperatures and reduced precipitation, while background warming thermodynamically amplifies the temperature anomalies. (4) The SOM clustering identifies 16 circulation patterns (A1–D4), among which D2–D4 represent typical hot-dry circulation types. These have the strongest influence on CHDEs and show an overall increasing frequency. Since the 1980s, human activities have significantly increased the occurrence probability of these hot-dry circulation types.