Should Severe Weather Graphics Wear Uniforms? Understanding the Effects of Inconsistent Convective Outlook Graphics on Members of the Public

Within the last decade, the Weather Enterprise has become increasingly concerned that visual graphics shared with members of the public can be inconsistent, and in turn, may have a negative effect on public risk perception. In particular, meteorologists often vocalize their concerns when members of the Weather Enterprise share the Storm Prediction Center’s (SPC) Convective Outlook graphic and alter its risk-related content (e.g., use different colors, risk language, different risk areas, etc.). But, does this visual variety have any negative consequences? The current literature fails to address the effects of inconsistent visual information and, consequently, is unequipped to offer advice on how and why practitioners and operational meteorologists should achieve message consistency – especially in this era of visual communication. To better understand and substantiate the widely assumed effects of inconsistent or conflicting graphical information, three experimental studies were conducted. Drawing on previous qualitative research with members of the public, the first experimental study manipulated five graphical variables (i.e., different risk areas, different colors, geographic scale, number of risk categories, and risk category language) to determine which one(s) resulted in the most perceived inconsistency. After examining the results of the first experimental study, two graphical variables emerged as prominent drivers of graphical inconsistency: (1) graphics that move a risk boundary and depict a location in two different risk areas and (2) graphics that use different colors to describe a location’s severe weather risk. As a result, these two variables were manipulated in a repeated-measure experimental design with both a student sample (experimental study #2) and a representative sample of the general public (experimental study #3). Not only that, but this experimental design also measured participants’ actual information seeking behaviors in real-time. This provided a rare glimpse into the weather information seeking habits of participants and the types of information they sought after seeing inconsistent Convective Outlook graphics. Therefore, these studies offer rich quantitative data that describes the effects of inconsistent weather-related graphical information on members of the public. More importantly, however, these results will likely have policy implications and provide operational best practices to promote Convective Outlook consistency across the Weather Enterprise.