NWS ensemble model to overcome rainfall data gap

Despite (U.S.) National Weather Service predictions of heavy rain and moderate flooding risk, as well as the issuance of repeated alerts, many rural Kentucky residents received no warnings during a severe weather event in July, 2022.

Limited cellular and Internet access are a major factor in reliable delivery of emergency alerts to communities. Kentucky state officials estimate “at least four years” until most rural counties may expect broadband infrastructure.

The lack of telecommunications infrastructure also results in “imprecise weather forecasting” as weather monitoring stations cannot reliably transmit observational data to accurately measure the severity of a present storm and predict future weather events.

In July, 2022 the NWS model predicted “moderate risk” but in hindsight, federal meteorologists may “argue that a ‘high risk’ designation would have been more appropriate.”

“Historically, many weather observations were developed around aviation, so a lot of weather radars are located at major airports in highly populated cities,” Jerry Brotzge, Kentucky state climatologist and director of the Kentucky Climate Center, explained to Claire Carlson and Anya Slepyan of The Daily Yonder. “That leaves a lot of rural areas with less data.”

With much less climatology data from historical rain events, weather prediction models will continue to fail rural communities – unless weather monitoring networks expand to increase data density and fill in the missing climatological data gap.

The NWS Environmental Modeling Center is developing the Unified Forecast System with a group of academic and community partners. Set to launch in 2024, the one-model system so far has shown promising results in the prediction of “short-range weather and associated extremes.”

The 10-model Rapid Refresh Forecast System is set to supply an ensemble of predictions to take into account more statistical uncertainty and produce a “broader, and more accurate, range of results.” The future of all weather prediction will depend more on “statistical probabilities and less on current and historical observational data.”

NWS anticipates that the Rapid Refresh Forecast System will “mitigate the disparity between urban and rural forecasting.”