Nonstationary influence of the North Atlantic tropical cyclones on the spatio‐temporal variability of the eastern United States precipitation extremes

• Published in: International journal of climatology 2020-06, Vol.40 (7), p.3486-3499
• Main Authors: Dhakal, Nirajan, Jain, Shaleen
• Format: Article
• Language: English
• Subjects: Annual precipitation; Atmospheric precipitations; cluster analysis; Cyclones; Daily precipitation; Empirical analysis; Environmental risk; extreme daily precipitation; Extreme values; Extreme weather; Flood management; Flood protection; Flood risk; Floods; Heavy precipitation; Heavy rainfall; Hurricanes; Maximum precipitation; nonstationary; Precipitation; principal component analysis; Probability distribution; Probability theory; Rain; Rainfall; Risk management; spatial and temporal variability; Statistical analysis; Statistical methods; Temporal variability; Temporal variations; Tropical climate; Tropical cyclones
• ISSN: 0899-8418; 1097-0088
• DOI: 10.1002/joc.6409

In the eastern United States, the empirical probability distribution of extreme daily precipitation comprises heavy rainfall events stemming from North Atlantic tropical cyclones (TCs). At many locations, these events influence estimates of extreme value statistics (e.g., 100‐year event), thus have important bearing on the sizing of flood protection infrastructure and, in general, flood risk management and preparedness. Consequently, a characterization of location specific and regional patterns in precipitation extremes and changes therein has salience for both scientific and engineering concerns. To this end, analysis of seasonal and annual maximum daily precipitation at 667 long‐term stations across the eastern United States was pursued to ascertain recent changes in the extreme events over the 1950–2011 period. Three key results from this study illuminate less understood facets of recent changes in precipitation extremes: (a) an overall increase in the fraction of seasonal and annual maximum precipitation events linked to TCs, (b) a dramatic increase in the correlation between Accumulated Cyclone Energy Index and the leading principal component of extreme precipitation, and (c) changes in the spatial patterns of regions with highest TC‐related risk for heavy precipitation. In the eastern United States, the spatio‐temporal pattern of seasonal and annual maximum precipitation during the recent time period show high temporal coherence with accumulated cyclone energy (ACE) index indicating that the impact of North Atlantic tropical cyclones on the extreme precipitation distribution is increasing covariations in the recent decades. This also shows that ACE as a metric of hurricane activity is useful for the analysis of extreme rainfall pattern along the U.S. east coast and the Gulf coast.