Wintertime storms along the east coast of North America can produce impactful weather hazards such as heavy snow, wind, and storm surge. Much of the previous research regarding these storms in a climatic sense has centered around case studies, gridpoint distributions, or more large-scale metrics, such as seasonal means. However, societally-relevant statistics at the storm level have been lacking, particularly in studies aimed at future projections on decadal timescales in global models.

This presentation defines a storm tracking metric that collocates snowfall totals with population density and applies it to a large ensemble of climate simulations to project a decrease in northeastern United States snowstorm events by the end of the 21st century. We find that this decrease is not proportional across all storm intensity classes, with reductions being more heavily weighted towards weaker events as opposed to stronger ones. This arises due to offsetting impacts of increased precipitation associated with future coastal  low pressure systems and decreases in thermal support for precipitation to fall as snow. We also discuss the potential impact of atmospheric model resolution moving forward and highlight the importance of understanding higher-order moments of storm distributions and corresponding metrics that better link their impacts to society.