Anne McFarland

Effects of Surface and Groundwater Interactions on Phosphorus Transport within Streambank Sediments

Understanding internal stream P transfers is important in controlling eutrophication. To determine the direction of groundwater and surface water interactions and evaluate P retention within streambank sediments, groundwater well pairs, about 5-m deep, were installed at three locations along a second-order, eutrophic stream in north-central Texas. Well cores were analyzed for P, and groundwater levels were monitored for about 2 yr. Water levels in wells furthest upstream always indicated a losing stream, while wells further downstream showed a gaining stream except during flow reversals with storm events and periods with reservoir backwater. Total-P from well cores ranged from 54 to 254 mg kg(-1) and was typically high near surface, decreased downward until redoximorphic features were encountered and then increased notably with depth to near or above surface concentrations. Very little extractable P occurred in sediments from the two upstream well sets; however, the set furthest downstream showed extractable P throughout with a high of 21 mg kg(-1) near the bottom. Repeated wetting-drying at sites A and B as noted by redoximorphic features may have shifted P into more stable sediment-bound forms. The decrease in extractable P at sites A and B compared to site C may be explained by conditions at C that were wetter and potentially anaerobic. Because the overall stream reach was more often losing than gaining, there appears to be a mass flow of P into streambank sediments. Streambank erosion may then transport this P downstream if not controlled.

Quantitative assessment of Naegleria fowleri and Escherichia coli concentrations within a Texas reservoir.

Previous presence/absence studies have indicated a correlation between the presence of the pathogenic amoeba Naegleria fowleri and the presence of bacteria, such as the fecal indicator Escherichia coli, in environmental surface waters. The objective of this study was to use quantitative real-time polymerase chain reaction (qPCR) methodologies to measure N. fowleri and E. coli concentrations within a Texas reservoir in late summer, and to determine if concentrations of N. fowleri and E. coli were statistically correlated. N. fowleri was detected in water samples from 67% of the reservoir sites tested, with concentrations ranging up to an estimated 26 CE (cell equivalents)/100 mL. E. coli was detected in water samples from 60% of the reservoir sites tested, with concentrations ranging up to 427 CE/100 mL. In this study, E. coli concentrations were not indicative of N. fowleri concentrations.

Preliminary studies on occurrence of monensin antibiotic in Bosque River Watershed

Water quality impact due to excessive nutrients has been extensively studied. In recent years, however, micro-pollutants such as pharmaceuticals and hormonal products used in animal agriculture have added an additional impact to overall water quality. Pharmaceuticals used in the poultry, swine, beef, and dairy industries have been detected in various environmental matrices such as, soil, groundwater and surface water. In this study, 26 surface water samples were collected throughout the Bosque River Watershed (BRW) with samples representing a range of land use conditions and locations of major dairy operations. Samples were analyzed using commercially available Enzyme-Linked Immunosorbent Assay test. Of the 26 samples, three samples consistently tested positive for monensin antibiotic with concentration ranging from 0.30 to 3.41 microg/L. These three samples were collected from sites that received varying amount of agriculture wastes (11.7% to 31.3%) and located downstream from sites associated with moderate levels of animal agriculture. The preliminary results suggest that there is a potential for monensin occurrence in the BRW, although initial findings indicate only very low levels.

ALTERNATIVE MONITORING STRATEGIES FOR ESTIMATING PHOSPHORUS CONCENTRATIONS AND LOADINGS FOR A TMDL

Data collection efforts generally represent a compromise between statistical confidence and available resources. Defining an appropriate data collection strategy for monitoring TMDL implementation involving nonpoint source contributions is, thus, a formidable task. Our objective was to define the sampling needed to adequately assess phosphorus (P) concentrations and loadings on an annual basis for sites along the North Bosque River in central Texas. Two TMDLs were conducted for segments on the North Bosque River to decrease soluble reactive P loadings and concentrations by about 50 percent. Our evaluation involved subsampling intensive storm and base flow monitoring data collected between 1995 and 1999. Preliminary results using routine sampling clearly show an increase in statistical confidence for annual average concentrations with increasing sampling frequency from quarterly to monthly to biweekly. Average annual P and TSS loadings were generally underestimated by biweekly sampling, using an integration approach, as compared to ‘true” loads as estimated from a full data set of storm and grab samples. Assuming that only a limited number of storm events could be monitored per year, storm sampling strategies evaluated included intensive monitoring of the first four storm events in a year or monitoring the first storm event in each quarter of a year. Flow-weighted event mean concentrations for storm events were associated with periods of elevated flow for loading calculations in conjunction with biweekly grab data representing base flow conditions. Intensive monitoring of the first four storm events in a given year appeared to give the most promising results for annual loading calculations for sites within the North Bosque River watershed. For sites where P or TSS is transported predominately by nonpoint source runoff, storm sampling using automated samplers is recommended to complement routine sampling.