Report of Investigations--KGS


Water quality in the Pleasant Grove Spring karst groundwater basin was monitored to determine the effectiveness of best management practices (BMP’s) implemented through the U.S. Department of Agriculture’s Water Quality Incentive Program (WQIP). The project was divided into three phases. Phase I, beginning in August 1990, was the initial reconnaissance of the hydrogeology and water quality of the basin. Phase II, beginning in October 1993, monitored the water quality for 1 year prior to BMP implementation. This phase was followed by a 1-year interim extension, which continued the monitoring. Phase III monitored the water quality during and following BMP implementation. The findings of phases I and II, along with extensive descriptions of the hydrogeology and groundwater quality, were reported in Currens (1999). This report covers the specific findings of the interim extension and phase III (October 1994–October 1998). It also summarizes the overall findings of the project and evaluates the outcome of the BMP’s, which began in 1995.

Pleasant Grove Spring discharges runoff from a 4,069-hectare (10,054-acre) karst groundwater basin in southern Logan County, southwestern Kentucky. The basin is characterized by mature karst topography developed on Mississippian carbonates mantled with residuum. Sinkholes and sinking streams dominate the landscape, and perennial surface-flowing streams occur only in the headwaters of the basin. Most of the area of the basin (about 90 percent) is used for agriculture. The principal crop grown is corn in rotation with winter wheat and soybeans. Other row crops include tobacco and other small grains. Livestock are dairy and beef cattle and swine. Over 68 percent of the area of the watershed was enrolled in the WQIP.

Analysis of samples collected since October 1994 at seven locations in the basin indicated the principal contaminants of probable agricultural origin were herbicides, nitrate-nitrogen, suspended sediment, orthophosphate, and bacteria (as was the case during the first two phases of the project). The maximum nitrate-nitrogen concentration measured in the basin between 1994 and 1998 was 13.1 mg/L, at Leslie Page karst window, and the average concentration was 5.05 mg/L. The maximum orthophosphate concentration was 1.4 mg/L, at Pleasant Grove Spring, and the median was 0.17 mg/L. The total suspended solids maximum was 3,267 mg/L, and the median concentration was 53 mg/L. The maximum triazine concentration measured by enzyme-linked immunosorbent assay (ELISA) was 393.0 µg/L, at Leslie Page karst window; median concentration was 1.15 µg/L. Maximum bacteria counts were 200,000 fecal coliform colony-forming units per 100 mL (col/100 mL) and 810,000 fecal streptococci col/100 mL; medians were 400 col/100 mL and 640 col/100 mL, respectively.

Water quality at Pleasant Grove Spring was monitored from May 1992 through the end of the project in October 1998. The maximum nitrate-nitrogen concentration measured at the spring was 8.11 mg/L, and the concentration never exceeded the maximum contaminant level (MCL) of 10 mg/L; average concentration was 4.8 mg/L. The maximum orthophosphate concentration was 1.4 mg/L and the median was 0.53 mg/L. The total suspended solids maximum was 3,073 mg/L, and median was 55 mg/L. The maximum triazine concentration (ELISA) was 62.2 µg/L. Triazine concentrations briefly exceed MCL’s during the spring each year. Peak concentrations of the other three frequently analyzed pesticides (alachlor, metolachlor, and carbofuran) were 12.0, 29.6, and 7.4 µg/L, respectively—the highest measured in the basin. Median concentrations of these pesticides, however, are near detection limits. Fecal coliform and fecal streptococci bacteria are always present at Pleasant Grove Spring, and counts occasionally exceed drinking-water supply limits (2,000 col/100 mL). Maximum bacteria counts were 60,000 col/100 mL of fecal coliform and 200,000 col/100 mL of fecal streptococci.

The quality of groundwater discharging at Pleasant Grove Spring before and after BMP implementation was evaluated by comparing the annual mass flux of nitrate-nitrogen, total suspended solids, and triazines (atrazine-equivalent). Annual descriptive statistics were compared for orthophosphate and bacteria, as well as for the other contaminants. The flux and annual statistics of nitrate-nitrogen were little changed over the course of the BMP program. Atrazine-equivalent flux and triazine geometric averages indicated an increase. Total suspended solids concentrations decreased slightly, whereas orthophosphate increased slightly. Fecal streptococci counts improved, but the improvement was not statistically significant.

The comparison of the pre- and post-BMP monitoring indicates that the WQIP was only partly successful. Although the program was fully implemented, the types of BMP’s funded and the rules for BMP participation resulted in less-effective BMP’s being chosen by producers. Future BMP programs for the protection of groundwater in karst aquifers should limit BMP’s to the installation of buffer strips around sinkholes, the exclusion of livestock from streams, and the removal of certain land from agricultural production.

Publication Date



Series XII

Report Number

Report of Investigations 11

Digital Object Identifier (DOI)


© 2005 University of Kentucky

Funding Information

Thanks to the University of Kentucky College of Agriculture and the U.S. Environmental Protection Agency, through the Kentucky Division of Water’s Nonpoint Source Pollution Control Program, for financial support. Field reconnaissance and sampling, and the purchase of major equipment, began in February 1991, when funding was received through the University of Kentucky College of Agriculture from Kentucky Senate Bill 271 (enacted in 1990). Funding for phase I was received from the U.S. Environmental Protection Agency’s Nonpoint-Source Program (Section 319 of the Clean Water Act) through the Kentucky Division of Water (Memorandum of Agreement 11399) in April 1992. Additional Section 319 funding was received in September 1993 for phase II (Memorandum of Agreement 12875) and for an interim continuance of phase II in September 1994 (Memorandum of Agreement 15424). Three years of funding for phase III was received in May 1995 (Memorandum of Agreement 16080), and a no-cost extension was granted to continue the monitoring through October 1998, the end of the 1997–98 water year.