May 22 - July 28
An East Coast Winter Storm Climatology and Projected Future Trends
Lena Dziechowski, Elinor Martin, and Liz DiGangi
What is already known:
- East coast winter storms coincide with increasing extreme precipitation, wind, and storm surge events.
- Large observed annual and decadal variability of frequency and intensity of storms.
- Most models project a decrease in frequency, however changes in intensity are less certain.
- Studies use only mean sea level pressure or 850-hPa vorticity to track cyclones.
What this study adds:
- Historical observations show increase in maximum wind and no change in minimum pressure. However, models underpredict the maximum wind in the historical period.
- Models project a decrease in frequency, no change in intensity.
- A tracking algorithm with a model dependent wind threshold should be tested for cyclone identification.
Abstract:
Densely populated major cities along the U.S. East Coast suffer substantial societal impacts due to extratropical cyclones occurring within the winter months of October-April. To mitigate damages and assist the nation in preparing for extreme winter weather in a changing climate, a full climatology of East Coast Winter Storms (ECWS) has been created using NCEP/NCAR reanalysis data from 1950-present day utilizing a cyclone-tracking algorithm that uses sea level pressure and a maximum wind threshold. The observations show an increase in maximum winds and no change in minimum pressure since 1950. After assessing changes in frequency and intensity based on minimum pressure, maximum winds and geographical location in the historically observed period, the cyclone-tracking algorithm was applied to historical and high emissions future scenarios utilizing data from seven models from the Coupled Model Intercomparison Project Phase 5. The models accurately represent historically observed minimum pressure, but under represent maximum winds and storm counts. Future models project a decrease in frequency but no change in intensity. Investigation of storm structure and lifecycle within the models is needed, in addition to a model dependent wind threshold.