The scope of this project was quite broad in covering a variety of factors which may contribute to the public's interpretation of weather forecasts. The goals of this project, however, were rather concise. In the interest of writing the most informative and unambiguous forecasts possible under NWS guidelines, the authors attempted to uncover and understand where communication problems may exist between the forecaster and the general public in Juneau, Alaska. During the course of the study, the authors' primary objective was to obtain useful information for the forecasters at WSFO Juneau pertaining to the public's interpretation, opinion, and utilization of the weather forecasts while using the questionnaires to help educate the public about the forecasts and the NWS.
After asking a number of questions pertaining to public interpretation of both numerical and verbal expressions of probability, it can be concluded that often there is a gap between how a user interprets the probability of precipitation and how the forecaster intended to convey it. This gap exists in the perceived meaning of some of the qualifying terms used in the precipitation forecast. Results from this series of questionnaires indicates that the verbal qualifiers often may not invoke the same expectation as the numerical percent probability expression does. The terms which the NWS may use to describe a categorical precipitation event (80-100% POP) only received mean percent probabilities of near 50% when respondents were asked to assign POPs to them. Additionally, users were asked to rate "missed" precipitation forecasts, i.e. forecasts for rain or snow for which the precipitation did not occur at their house, and respondents were more critical of the numerically expressed probabilities than the probabilities expressed with verbal qualifiers. Based on the results, the authors have concluded that forecasters should use care in applying supplemental verbal qualifiers to the forecast, such as descriptions of duration or spatial extent. These terms may be very effective at providing useful information to the user, but they may also be misapplied by the user in the wrong context.
The results of the project also showed that the public may be slightly more critical of snow forecasts than forecasts for rain. When respondents were given similar questions pertaining to accuracy with rain and snow independently used as the precipitation type, opinions were more critical of the snow forecasts. Respondents were also given identical questions with rain and snow used in separate questions, but with one question right after the other, and results still showed a slight heightened sensitivity to snow forecasts.
A variety of questions pertaining to terminology used in WSFO Juneau forecasts were asked, as well as a few questions concerning adverse weather conditions and natural disaster preparedness applicable to the Juneau area. Overall, the results to these questions were favorable in terms of public understanding. It is evident that further education by WSFO Juneau is necessary to create better understanding of NWS terminology and to increase safety awareness in the community.
Many of the respondents wrote personal comments on their questionnaires, and some of these were particularly meaningful to the authors. It is apparent that a number of people in the community do not have a good understanding of what the NWS does, and many do not differentiate between the radio announcers, the Weather Channel, and other local sources of weather information as being different from the NWS. These are issues that should be addressed in order to increase public confidence in the NWS and the weather products issued.
While a project such as this one does have a positive impact on its community, in addition to providing useful information to the forecasters about their audience, a properly conducted national public survey would make unparalleled progress in resolving these issues.