Date of Award

August 2016

Degree Type


Degree Name

Master of Science



First Advisor

Paul Roebber

Committee Members

Clark Evans, Kyle Swanson


On 8-9 February 2013, a strong extratropical cyclone brought historic winter storm conditions to the Northeast United States with a swath of one to three feet of snow falling across much of New England, with isolated pockets exceeding three feet across south-central Connecticut. Given the high socioeconomic impact that resulted from this blizzard, this study focuses on the run-to-run variability in operational model (North American Mesoscale model; NAM) forecasts leading up to the event. These forecasts, initialized forty-eight hours or less from the start of the event, showed two instances indicating a substantial shift in the expected impact. The first occurred between the runs of 0600 UTC and 1200 UTC 7 February 2013 and the second between 0000 UTC and 0600 UTC 8 February 2013. These shifts were discovered to have primarily resulted from large differences in the forecast strength of the warm air advection and the accumulated effects of those differences over time. Analysis of the mesoscale patterns show that the typical mesoscale snowband ingredients of mid-level frontogenesis and conditional symmetric instability (CSI) were not significant contributors to this high impact case. Instead, strong and focused warm air advection, with a secondary contribution from cyclonic vorticity advection through the Sutcliffe self-development process, appears to be the main source of cyclone track variations and the attendant intense, localized vertical motion that lead to the development of the snowbands.