Scope 2-Alan Fryar: 104B Water Resources Research Institute Annual Base Program 2021-2026

More than half of Kentucky is underlain by limestone and therefore prone to development of karst features. Integration of surface and subsurface drainage through sinkholes, conduits, caves, and springs results in rapid groundwater flow and minimal filtration of suspended particulate matter. Consequently, karst aquifers are prone to contamination by fecal pathogens, which can enter the subsurface via leaking or overflowing sanitary sewers, failing septic systems, wastewater discharge, and agricultural and urban runoff. Challenges in addressing contamination by pathogens include early detection of their presence and accurate assessment of human health risks resulting from exposure. E. coli bacteria are typically used to assess water quality with respect to enteric pathogens, but these simple fecal indicators require culturing and incubation prior to enumeration. A promising alternative is dissolved organic matter (DOM), notably tryptophan-like fluorescence (TLF), which has been positively correlated with E. coli counts in multiple studies and can be monitored with realtime sensors. However, few studies of TLF have been conducted in karst systems and none in Kentucky. Moreover, to assess exposure risks, E. coli must be related to actual pathogen concentrations. We propose to examine the utility of real-time TLF monitoring and conduct quantitative microbial risk assessment (QMRA) for the Cane Run/Royal Spring basin in Fayette and Scott counties and the Camden Creek basin in Woodford County. These basins encompass both urban and agricultural (pasture and row-crop) land uses. Georgetown, which draws water from Royal Spring, is the largest municipality in Kentucky using a karst aquifer for water supply. We plan to monitor Royal Spring and a site on Camden Creek for E. coli and DOM over one year via weekly sampling, event-based sampling during storms, and continuous monitoring of TLF and fluorescent DOM. This work will build on existing monitoring of turbidity, NO3, and other parameters being conducted by other UK researchers at both sites, as well as prior research by Fryar and students. By conducting laboratory analyses for DOM across a broad spectrum of fluorescent excitation and emission wavelengths, we will examine whether other DOM components besides TLF are potential indicators of fecal contamination. Non-parametric statistics will be used to assess correlations between water-quality parameters and differences between populations. Ratios relating E. coli concentrations to literature values of Campylobacter jejuni (bacteria) and Cryptosporidium parvum (protozoa) concentrations will be used in the QMRA model. Both these pathogens have been included in multiple QMRAs focused on ingestion through drinking water and recreational activities. Uncertainty and sensitivity analyses will be performed to quantify uncertainty in modeled health risks and highlight which model parameters are driving it. Our objectives are to provide state-of-the-art, robust, and readily usable techniques for stakeholders (e.g., public utilities and regulators) to assess the timing, extent, and health risk of microbial contamination. This project will underpin the MS thesis research of co-PI Ryan Dapkus.