Multiscale Frontal Circulations during the 6–8 January 2014 Lake-Effect Snow Event: Understanding their Superposition and Predictability using Convection-Permitting Ensemble Forecasts

Lake-effect snow (LeS) is a high-impact weather phenomenon that is difficult to forecast, due in part to sensitivities of numerical simulations of LeS to uncertainties in the synoptic and mesoscale initial conditions/boundary conditions (ICs/BCs) that influence the position, intensity, structure, and evolution of individual LeS bands. Deterministic, limited-area, convection-permitting model guidance often fails to identify salient LeS forecast uncertainties, namely, the location and persistence (or lack thereof) of individual LeS bands in response to mesoscale frontal circulations that interact with shifts in the direction of the larger-scale wind flow. To better understand the predictability of these frontal circulations, such as those associated with cold fronts and land-breeze fronts, we focus on convection-permitting ensemble forecasts of the 6–8 January 2014 storm during the Ontario Winter Lake-effect Systems (OWLeS) field campaign. Throughout this extended three-day LeS event, long-lake-axis-parallel snowbands persisted downwind of the eastern shore of Lake Ontario, oscillating north and south during multiple cold surges in the wake of an intensifying extratropical cyclone. Snow accumulations of as much as 152 cm on the northern part of the Tug Hill Plateau accompanied the outbreak of Arctic air and strong winds.

To understand the evolution and predictability of cold frontal and land-breeze frontal circulations during this LeS event, we run limited-area, 20-member Weather Research and Forecasting (WRF) ensemble simulations at 1.33-km horizontal grid spacing for three different initialization times. While the ensembles use BCs from the Global Ensemble Forecast System (GEFS), ICs are generated using continuously cycled, ensemble Kalman filter (EnKF) data assimilation based on the NCAR Data Assimilation Research Testbed system.

From the EnKF IC/GEFS BC ensembles, we quantify the extent, orientation, and intensity of LeS and its interaction with the cold frontal and land-breeze frontal circulations. Furthermore, we assess the precipitation and frontal forecast uncertainty as a function of lead time. By quantifying the spread in intensity and position of both the fronts and LeS bands in these IC/BC ensemble simulations, we analyze the predictability horizon of the mesoscale details for this event. The WRF forecasts are also evaluated against OWLeS snowfall, rawinsonde, and radar observations to understand how the ensemble distributions of snowband morphologies, snowfall amounts, and frontal locations compare to observations. Ultimately, we are working to understand how the predictability of this LeS event is modulated by uncertainties in the interaction of multiscale frontal circulations with the complex shoreline geometry of Lake Ontario.