Effective uncertainty visualization for aftershock forecast maps

Abstract. Earthquake models can produce aftershock forecasts, which have recently been released to lay audiences. While visualization literature suggests that displaying forecast uncertainty improve how maps are used, research on is missing from earthquake science. We designed a pre-registered online experiment test the effectiveness of three techniques for and their uncertainty. These showed forecasted number aftershocks at each location week following hypothetical mainshock, along with around location's forecast. Three different visualizations were produced: (1) adjacent one another; (2) map depicted in color scheme, shown by transparency color; (3) two lower upper bounds distribution location. compared using tasks specifically address broadly applicable user-generated communication goals. task responses between participants without its (the current practice). Participants completed map-reading targeted several dimensions readability visualizations. then performed Comparative Judgment task, demonstrated whether was successful reaching key goals: indicating where many no likely (sure bets) low but high enough imply potential risk (surprises). All equally well goal communicating sure bet situations. But substantially better than other designs surprises. results implications visual both within beyond