摘要:Extreme weather events in Asia have been occurring with increasing frequency as the globe warms in response to rising concentrations of greenhouse gases. Many of these events arise from weather regimes that persist over a region for days or even weeks, resulting in disruptive heatwaves, droughts, flooding, snowfalls, and cold spells. We investigate changes in the persistence of large-scale weather systems through a pattern-recognition approach based on daily 500 hPa geopotential height anomalies over the Asian continent. By tracking consecutive days that the atmosphere resides in a particular pattern, we identify long-duration events (LDEs), defined as lasting longer than three days, and measure their frequency of occurrence over time in each pattern. We find that regimes featuring positive height anomalies in high latitudes are occurring more often as the Arctic warms faster than mid-latitudes, both in the recent past and in model projections for the twenty-first century assuming unabated greenhouse gas emissions. The increased dominance of these patterns corresponds to a higher likelihood of LDEs, suggesting that persistent weather conditions will occur more frequently. By mapping observed temperature and precipitation extremes onto each atmospheric regime, we gain insight into the types of disruptive weather events that will become more prevalent as particular patterns become more common.