College of Education and Behavioral Sciences News

Advances in weather forecast models have enabled forecasters to improve the accuracy of forecasts, including severe weather warnings, helping to save lives and property. The Kentucky Mesonet at WKU, a division of the Kentucky Climate Center, will play a key role in a $5 million National Science Foundation project aimed at making further advances in near-term forecasts that can be critical in active and severe weather situations.

The project, titled Precise Regional Forecasting via Intelligent and Rapid Harnessing of National Scale Hydrometeorological Big Data, is being led by Dr. Nian-Feng Tzeng of the Center for Advanced Computer Studies at the University of Louisiana at Lafayette (ULL). A team of 14 researchers at ULL, Louisiana State University, Southern University, the University of South Alabama and Western Kentucky University will work collaboratively. “This four-year research project addresses better regional weather forecasts via big data analytics, especially critical nowadays in light of climate change that has emerged as a stark problem of national importance,” Dr. Tzeng said.

He further noted that researchers “will undertake collaborative research with technical innovations and to enhance workforce development in [the] five participating institutions.”

At WKU, Dr. Stuart Foster, state climatologist and director of the Kentucky Mesonet, and Dr. Eric Rappin, research scientist, will lead efforts to develop two-dimensional mini-models, called modelets. Modelets will leverage meteorological data from the Kentucky Mesonet and high-resolution forecast model outputs generated by the National Center for Environmental Prediction using machine learning algorithms to generate high-resolution near-term weather forecasts. A key component of the research will involve the development of algorithms that account for the effects of topography, complex terrain, and other land surface features on weather.

 “Our participation in this project is a testimony to the value of the Kentucky Mesonet both to WKU and to the Commonwealth of Kentucky,” said Dr. Foster, the WKU project lead. “Kentucky’s diverse terrain and land use make the state an ideal natural laboratory for meteorological research. The prospect of using machine learning algorithms with Kentucky Mesonet observations has the potential to bring significant value to agriculture, energy, transportation, and other sectors of Kentucky’s weather-sensitive economy.”

The project will fund up to 16 student researchers at WKU over the course of four years. In addition, students will participate in Research Experience for Undergraduates (REU) summer camps at the University of Louisiana at Lafayette and the University of South Alabama focusing on computational science and meteorology, respectively.

About the Kentucky Mesonet at WKU: The Kentucky Mesonet at WKU is the Commonwealth’s official source for weather and climate data. The statewide network includes 73 stations in 68 counties. The Mesonet stations collect real-time data on temperature, precipitation, humidity, barometric pressure, solar radiation, wind speed and direction, soil moisture and soil temperature and transmit it to the Kentucky Mesonet at WKU Operations Center (housed at WKU’s Center for Research and Development) every five minutes, 24 hours per day, throughout the year. The data is available online at www.kymesonet.org.

Contact: Stuart Foster, (270) 745-5983; or Eric Rappin, (270) 745-4555