Item

Abstract

The February 2021 North American Cold Air Outbreak (CAO) featured historic cold and other winter hazards over a two-week period across the Central United States (US), leading to major socioeconomic impacts. Leading up to the event, model forecasts diverged on the evolution of critical features that ultimately affected their forecast accuracy. In particular, the CAO event was preceded by two wave breaks at high latitudes -- one in the East Siberian Sea (2-4 February 2021) and another in the Labrador Sea (9-11 February 2021). In this talk, we illustrate how these wave breaks were important for the ultimate evolution and intensity of the CAO in subseasonal model forecasts. We use a combination of ERA5 and realtime forecasts from two models from the Subseasonal-to-Seasonal (S2S) Prediction Project Database for the analysis: ECMWF and NCEP. We select model initialization dates with 2-3 weeks lead time to isolate ensemble members that simulated the wave breaks well and poorly and then evaluate their subsequent forecasts for the event. In both models, ensemble members that successfully simulated these wave-break features produce higher chances for widespread negative temperature anomalies across the Central US, corresponding to strong anomalous Arctic ridging, as seen in reanalysis. By contrast, ensemble members with either no or weak wave breaking forecast neutral to even positive temperature anomalies for the Central US over the target period, with less amplified flow. These results reinforce the need to simulate better blocking regimes in S2S operational models for improved accuracy of extreme winter weather events.