NSF Award
2435472
Lignite is a low rank of coal that is abundant in most Gulf Coastal plain aquifers and has been shown to cause increased occurrences of kidney disease on the Balkan Peninsula of eastern Europe and in Texas, Arkansas, and Louisiana. Mississippi led the nation in kidney disease mortality from 2015-2020 and was second in the nation in 2021. This early concept research project will collect water chemistry data throughout the state of Mississippi and will perform a suite of geospatial statistical analyses on the collected well water data and existing datasets from the US Geological Survey and from the medical sector and other sources to see if ground water from Gulf Coastal Plain aquifers can be linked to kidney disease in Mississippi. Broader impacts of this research is relevant for all residents of Mississippi and potentially residents of all Gulf Coastal Plain states who obtain their water from ground water wells. If a correlation is found, the results of this research can be used to develop mitigation strategies and could identify zones within specific aquifers or specific geographic regions that should be avoided as domestic water supply sources.<br/><br/>Climate-related drought and unpredictable rainfall patterns have led to increasing reliance on groundwater for agricultural irrigation in the Gulf Coastal Plain resulting in significant increases in annual withdrawals from aquifers that are also used for domestic water supplies via municipal and private drinking water wells. Most Gulf Coastal Plain aquifers have naturally occurring lignite, which has been shown to cause increased rates of kidney disease in other parts of the US and world. Mississippi’s mortality rates for kidney disease are among the highest in the nation; but it is unknown if this high rate may be related to the presence of lignite in Gulf Coastal Plain aquifers or due to some other cause. This project applies a variety of machine learning algorithms and techniques on datasets that include water well construction data (location, depth of screened interval, age), health data from the State of Mississippi Medicare database, U.S. Census data, and geochemistry results from a new reconnaissance groundwater survey this project will complete. Goals are to see if there is a statistical correlation between increased rates of kidney disease and the use of lignite-bearing ground water aquifers. Kidney health data will be filtered and classified by a medical doctor specializing in renal functions. Analyses will initially be done for each county in the state. The project funds a Masters-level student in geology at the University of Mississippi which is in an EPSCoR state. The student will learn multidisciplinary skills in geoscience and human health, including: (1) importing and managing data in a geographic information system, (2) performing geospatial analyses techniques; (3) learning how to integrate datasets into machine learning algorithms and how to train AI for neural network analyses on a supercomputer; (4) engaging and interacting with a large cross section of society through seeking permission to sample domestic wells throughout the state; and (5) sampling protocols and techniques for collection of data from water wells, including maintaining chain of custody documentation. Project results may have implications for millions of residents in states across the Mississippi Gulf Coastal Plain states (i.e., Alabama and Georgia) who use groundwater for drinking water. It many also have impact on states such as West Virginia, Kentucky, and Pennsylvania with significant coal-mining industries that have aquifers where there is an intersection of domestic water sources and coal mining activities.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
