2.5 An Overview of the NOAA/NWS Storm Prediction Center’s Integrated Machine-Based Predictive Analytics for Convective Threats to Society (IMPACTS) Statistical Model

Tuesday, 9 January 2018: 11:30 AM
Salon K (Hilton) (Austin, Texas)
Patrick T. Marsh, NOAA/NWS/Storm Prediction Center, Norman, OK; and R. Clark III, R. S. Schneider, H. E. Brooks, S. A. Erickson, and A. R. Dean

NOAA’s Weather-Ready Nation is about readying communities for extreme weather, water, and climate events. This effort requires not only accurate and timely forecasts and warnings but also building the capacity, by working with key public and private partners, to respond effectively to these high-impact events. The NWS Storm Prediction Center (SPC) has built a statistical framework, Integrated Machine-based Predictive Analytics for Convective Threats to Society (IMPACTS), to simulate potential societal impacts of tornadoes. IMPACTS consists of two separate components: a weather generator, based upon SPC’s Convective Outlooks, that simulates the number, location, and strength/intensity of potential tornadoes, and a method of assessing the potential societal impacts from these simulation(s).

IMPACTS will improve our nation’s tornado response capabilities by providing reliable, probabilistic predictions of potential societal impacts to decision-makers at the local, regional (FEMA and states), and national (FEMA) level. Additionally, the methods underlying IMPACTS will serve as the foundation for societal-impact prediction involving other hazardous weather phenomena at a multitude of temporal and spatial scales. These methods are therefore uniquely positioned to capitalize upon NOAA’s FACETs initiative.

A primary outcome of IMPACTS is the creation, sharing, and use of transformational datasets that enhance emergency managers’ abilities to respond to high-impact tornado events. This presentation will focus on the creation of IMPACTS and its initial operating capacity of estimating the number of people exposed to tornadoes. A companion presentation will focus on efforts to extend this model to estimating potential tornadic fatalities and damage.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner