The Sensitivity of the December 2014 New York Mixed Precipitation Event to Initial Condition and Microphysics Uncertainty

During Dec. 9^th-10^th, 2014, a mixed precipitation event occurred in the upper Hudson River Valley. While rain was forecasted, this storm produced an unexpected heavy mesoscale snowband during rush hour in the Capitol region. These types of events continue to be difficult to forecast for operational meteorologists possibly due to the sensitivity of model simulations to both initial condition uncertainty and microphysical parameters. To assess the sensitivity, two types of 3-km Weather Research Forecasting (WRF) model ensemble forecasts are generated. One ensemble is characterized by different initial conditions (taken from the 21 members of the Global Ensemble Forecasting System) for each member, but the same physics package, while the other has the members with the same initial conditions, but different physics, similar to the Short Range Ensemble Forecast – Advanced Research WRF (SREF-ARW) members. “Good” and “bad” members were determined via comparisons of model soundings and radar to KALB observations. It was found that each member simulated different spatial coverage and intensity for the snow band that was observed in the upper Hudson River Valley. No member reproduced the actual mesoscale band during its peak, however, 5 of the 21 initial condition members produced a spatially similar meridional band. Furthermore, 6 of the 21 members accurately depicted the 975-800 hPa isothermal layer. This presentation will conclude by comparing members of the physics ensemble to members of the initial condition ensemble to assess the role of the initial condition uncertainties to the uncertainties in the physics for this event.