Public preferences for and understanding of probabilistic snowfall ranges

Communication of probabilistic forecast information has been a focused topic of interest in the Weather Enterprise for decades (National Research Council 2006, Morss et al. 2008, Hirschberg et al. 2011, Novak et al. 2023). The goal of such probabilistic messaging is to characterize and explicitly communicate the uncertainty inherent to forecasts due to predictability limitations of weather hazards, especially at local spatial scales and at longer lead-times. Empirical research on probabilistic messaging, for winter weather communication and in other hazard contexts, has demonstrated that members of the public and key expert stakeholders can understand probabilistic information and that such information can improve decision outcomes (Joslyn & LeClerc 2012, Ripberger et al. 2022, Rogers et al. 2023).

However, questions remain about when and how to effectively use probabilistic messaging (NOAA Science Advisory Board, 2021). This is true across weather hazards, including for winter storms as they evolve. For instance, it is unclear whether communicators should use quantitative (i.e., 50%) or qualitative descriptions (i.e., “medium”) to describe probabilities; or whether the public prefers “wide” (i.e., 2-8 inches) or “narrow” (i.e. 4-6 inches) ranges, or hybrid “goalpost” ranges that present a narrower range as well as the reasonable best and worst-case scenarios. Such open questions are further complicated by the real-world nature of snowfall forecasting where the distribution of forecast snowfall may not be normally distributed and where zero might be a plausible low-end for a snowfall range forecast.

In this presentation, we report on our efforts to empirically test how public preferences for and understanding of probabilistic snowfall ranges varies based on these messaging and meteorological dimensions of probabilistic snowfall ranges. We investigated these dimensions as part of the Winter Weather and Society survey (N=1459), which is a demographically representative sample of U.S. residents focused on winter weather perceptions and information usage. Our results augment existing empirical research to further inform the design and operationalization of probabilistic messaging for winter weather hazards.