Estimating potential severe weather societal impacts using probabilistic forecasts issued by the NWS Storm Prediction Center

In addition to expressing the relative uncertainty of high impact weather forecasts through the use of probabilities, community efforts are also focused on effectively communicating the uncertainty to the public and describing the potential impacts to society. Collectively these efforts are sometimes described as “completing the forecast”. The NWS Storm Prediction Center (SPC) has been issuing a variety of its severe weather forecasts in probabilistic format since 2000. These forecasts include information on the individual severe weather hazards (tornado, severe wind, large hail) and their expected intensity and areal coverage. The SPC is beginning an initial effort to estimate potential severe weather societal impacts based on a combination of probabilistic forecasts and high resolution population data. For equal severe weather threat, events that occur over densely populated areas of the country are likely to have greater societal impact than events over less populated areas. The current effort is focused on development of quantitative information on the estimated societal impact, but future research will be needed to examine the optimal communication of this information to interested partners in the weather enterprise and ultimately to the general public.

The current work will examine SPC severe weather forecasts from 2000 to present and the potential for estimating likely societal impacts when combined with high resolution population data derived from the 2000 United States Census. One facet of the study is to examine the combination of severe hazard probability and population density as an integral measure of the likely societal impact on a given day. In addition, identification of key population density thresholds for dense urban, urban and rural areas will allow additional dimensions of the potential threat to be quantified. Specifically, the potential for catastrophic impacts due to a tornado in a dense urban area can be estimated through the combination of information on the threshold areal coverage and the forecast probability for tornadoes and strong tornadoes. Damage, injury and fatality data for historic severe weather events will be used to examine the statistical effectiveness of the societal impact estimates from the forecast sample.