The relationship between atmospheric circulation patterns and extreme temperature events in North America

Extreme temperature events (ETEs) pose a significant risk to society, especially vulnerable populations with limited access to shelter and water and those with pre-existing respiratory and cardiovascular ailments. This research examines the relationship of atmospheric circulation with a myriad of metrics related to ETEs to better understand which synoptic-scale circulations are likely to have negative health/thermal comfort outcomes. Daily sea-level pressure (SLP) and 500-hPa geopotential height (z500) data from the North American Regional Reanalysis (NARR) were used to identify circulation patterns over North America. Self-organizing maps were used to partition the variability in circulation patterns over five distinct domains covering North America for both variables. Daily 2-m temperature, 2-m dewpoint temperature, and 10-m wind data from the NARR were used to derive five major categories of ETEs based on 95th percentiles: temperature events, apparent temperature events, dew point events, and excess heat and excess cold temperature events. The relationship of circulation pattern frequencies (SOM nodes) leading up to ETEs were assessed using point biserial correlations, accounting for spatial and temporal autocorrelation. The results show that z500 has a stronger association with ETEs than does SLP. A great deal of spatial variability exists in the strength of relationship for many ETE variables with circulation patterns likely due to the local geographical influence (e.g., leeside mountain adiabatic warming and low-level maritime flow). Generally, high extremes are associated with broad ridging and anticyclonic flow and cold extremes are associated with high amplitude trough patterns with low-level flow originating from the continental interior. The use of self-organizing maps presents a unique way of examining the potential for human health risks related ETEs and may be an effective method for statistically downscaling climate model data to assess the potential for ETEs in the future.