Regime-Dependent Predictability of Cold Season Precipitation Events in the St. Lawrence River Valley

The St. Lawrence River Valley in southern Quebec is characterized by a variety of precipitation types (p-types) during the cold season (October–May), including rain, freezing rain, ice pellets, and snow. These varied precipitation types exert considerable impacts on aviation, road transportation, power generation and distribution, and winter recreational activities, and are shaped by diverse synoptic and mesoscale processes that are influenced by the region’s complex topography. These characteristics motivated the Winter Precipitation Type Research Multi-Scale Experiment (WINTRE-MIX), which was carried out across the St. Lawrence River Valley in February–March 2022 with the goal to better understand how multi-scale processes influence the variability and predictability of p-type and amount under near-freezing surface conditions. During the experiment, a variety of synoptic-scale environments were sampled, including moisture-starved clipper systems embedded in northwesterly flow and strong warm-air advection regimes that featured several p-type transitions.

This study utilizes ERA5 reanalysis data and surface observations from Montreal, Quebec (CYUL) and Burlington, VT (KBTV) between 2000–2018 to investigate the spectrum of synoptic-scale weather regimes that induce cold season precipitation across the St. Lawrence River Valley. In particular, k-means clustering is used to classify cyclone tracks passing near the St. Lawrence River Valley into a set of weather regimes that include a U.S. East Coast track, a Central U.S. track, and two Canadian clipper tracks. Composite analyses are performed on each regime to reveal the synoptic-scale environments and the characteristic p-types that most frequently accompany each regime. GEFSv12 reforecasts are then used to examine the extent to which cyclone characteristics and local thermodynamic profile associated with each track type are skillfully predicted at 0–5-day forecast lead times. Preliminary results demonstrate that forecast cyclone tracks near the St. Lawrence Valley are generally too slow and left-of-track, which has implications for accurate forecasts of the local thermodynamic profile and p-type across the region under near-freezing conditions.