Predictability of a Succession of Three Extreme Precipitation Regimes (EPRs) in eastern North America in February 2019

Long-lived and widespread extreme precipitation periods often lead to flooding, causing loss of life and property. Forecasting such periods is still a significant challenge. As an example, a succession of three extreme precipitation regimes (EPRs) in eastern North America in February 2019 contributed to major flooding in the Ohio and Mississippi Valleys, a record wet winter in the contiguous United States, and unusually snowy winter in Ontario and Quebec, especially north of the St. Lawrence River. This exceptionally wet period was not predicted by numerical weather models or the Climate Prediction Center (CPC) beyond a two-week lead time. We assess the farthest lead time at which the numerical weather models, specifically the NCEP Global Ensemble Forecast System (GEFS) and ECMWF Ensemble Prediction System (EPS), finally skillfully predicted the wet period. From the model data, I hypothesize relevant synoptic-scale features in the North Pacific and/or North America that are especially important to simulate to predict EPRs skillfully.