In 1991, the U.S. Geological Survey (USGS) implemented the National Water-Quality Assessment (NAWQA) Program to describe the status and trends in water quality of a large, representative part of the Nation's surface- and ground-water resources. This program is based on a multidisciplinary approach using standard protocols to collect data in more than 50 study units (Hirsch and others, 1988; Leahy and others, 1990). The intensive fixed-site assessment is a component of the surface-water study design for the high intensity phase of sampling (Gilliom and others, 1995). The purpose of the intensive fixed-site assessment is to increase the sampling frequency and the number (relative to basic fixed-site sampling) of constituents analyzed at selected sites in the study unit. Dissolved pesticides and volatile organic compounds (VOC) were added to the list of constituents analyzed that already included major anions and cations, nutrients, suspended sediment, and suspended and dissolved organic carbon. The site selection and sampling strategy for the intensive fixed sites is centered on specific land use activities. The premise of the sampling strategy is that relatively frequent sampling at a few carefully chosen sites during key periods yields superior information about occurrence and seasonal patterns of constituents. This sampling strategy can provide information on seasonal and short-term temporal variability of general water quality and constituent transport and determine the occurrence and seasonal patterns of selected constituents such as nutrients, pesticides, and VOCs (Gilliom and others, 1995).
The major water-quality issue in the South-Central Texas (SCTX) NAWQA study unit (fig. 1) is the potential for contamination of the Edwards aquifer. The Edwards aquifer is the main aquifer in the study unit and is the sole source of water for the greater San Antonio area and for ranchers and farmers in the region. The intensive fixed-site assessment was designed to determine the effects of agriculture and urbanization on surface-water quality within the SCTX study unit that ultimately could affect the water quality of the Edwards aquifer. The Medina River watershed just west of San Antonio was used as an agricultural land use indicator, and the Salado Creek watershed within the San Antonio city limits was used as an urban land use indicator. The San Antonio River downstream of the San Antonio city limits, which is downstream of the inflows of the Medina River and Salado Creek, was used as an integrator of agricultural and urban land uses.
The purpose of this report is to describe and compare concentrations of nutrients, pesticides, and VOCs at three sites affected by different upstream land use in the San Antonio region of the study unit. Three intensive fixed sites with different upstream land use were selected for sampling on a weekly to monthly basis during April 1996-April 1998. Descriptions and comparisons of concentrations and temporal patterns of nutrients, pesticides, and VOCs at the three sites are based on graphs, boxplots, and tables.
The SCTX study unit in the San Antonio region comprises mainly the Edwards aquifer and its catchment area. The entire study unit extends beyond the San Antonio region to the Gulf of Mexico to include the complete watersheds of three major rivers (Nueces, San Antonio, and Guadalupe). Salado Creek and Medina River are in the San Antonio River system. For this report, the study area is defined as the San Antonio region of the SCTX study unit (fig. 1).
The Edwards aquifer is the sole source of water for about 1.3 million people in the San Antonio region. Water from the aquifer provides habitat for threatened and endangered species associated with major springs in the region. The Edwards aquifer is a dipping sequence of extensively faulted, fractured, and dissolutioned limestone and dolostone that yields large quantities of water to wells and springs. The aquifer crops out and is unconfined in the recharge zone (fig. 1). The aquifer is confined (artesian zone) beneath much less permeable rocks downdip from the recharge zone. Further downdip, where the rocks are virtually impermeable, they contain moderately saline to very saline water (saline-water zone) (U.S. Geological Survey, 1994).
The major process that can affect the quality of water from the Edwards aquifer in the study area is urban development in the greater San Antonio area. Land use in the city of San Antonio and surrounding area (fig. 2) is predominantly commercial and residential, including several large military installations and manufacturing and tourism facilities. Land use in the rest of the study area is mainly forest and rangeland with some agricultural land and small urban areas.
Three intensive fixed sites (fig. 1) were selected to determine the effects of agriculture and urbanization on surface-water quality in the study area. These sites are Medina River at LaCoste (08180640), Salado Creek (lower station) at San Antonio (08178800), and San Antonio River near Elmendorf (08181800). The drainage area of the Medina River watershed upstream of the sampling site, Medina River at LaCoste, is about 805 square miles (mi2) of which 634 mi2 is upstream of the dam forming Medina Lake west of San Antonio. Streamflow at this site is controlled by the dam at Medina Lake (Gandara and others, 1997). The predominant land use within the Medina watershed is forest and rangeland with small urban areas. Land use in about 10 percent of the watershed is agricultural. The Medina River sampling site was selected as an agriculture indicator site because of the proximity of agricultural activity. The predominant crops grown in the watershed are cotton, corn, and sorghum. Most of the cultivated fields are irrigated with Edwards aquifer water. Water quality at this site is affected by runoff from the fields into the river.
The drainage area of the Salado Creek watershed upstream of the sampling site, Salado Creek (lower station) at San Antonio, is about 189 mi2 and is mostly within the San Antonio city limits. Land use in this watershed is 80 percent urban and 20 percent agricultural and rangeland. Because of predominantly urban land use upstream, the Salado Creek site is considered an urban indicator site. The urban land use is predominantly commercial and residential with only a small percentage of industrial. Streamflow at this site might be affected during storm events by 11 floodwater-retarding structures in the upper part of the watershed (Gandara and others, 1997). Closer to the site, streamflow is maintained by precipitation and municipal wastewater discharges.
The drainage area of the San Antonio River watershed upstream of the sampling site, San Antonio River near Elmendorf, is about 1,740 mi2. Land use in this watershed is 14 percent urban, 22 percent agricultural, 61 percent rangeland and forest, and 3 percent water and barren land. Because of mixed land use upstream, the San Antonio River site is considered an integrator site. The city of San Antonio discharges wastewater effluent into the San Antonio River from three treatment plants located within the watershed (Gandara and others, 1997). The sampling site is downstream of Medina River and Salado Creek inflows to the San Antonio River.
The intensive fixed-site sampling strategy is to collect detailed data on the occurrence of contaminants such as pesticides and VOCs at fixed intervals and during high flows for approximately 1 year. The emphasis of the intensive sampling strategy is to collect samples when pesticides and VOCs are most likely to be detected. Most samples were collected monthly although pesticide sampling was more frequent during the spring when pesticides usually are applied, and VOC sampling was more frequent during the winter because VOC detections tend to be greatest during cold months (Lopes and Price, 1997). Pesticide and VOC samples also were collected during high-flow events. Tables 1-3 show when samples were collected at the three intensive fixed sites. Stream hydrographs for the three intensive fixed sites show discharge and sample collection (fig. 3).
Specific nutrients, pesticides, and VOCs analyzed in the samples are listed in table 4. Other constituents analyzed were major cations and anions, suspended and dissolved organic carbon, and suspended sediment, the concentrations of which are consistent among sites (Gandara and others, 1997, 1998).
Discrete fixed-interval and some high-flow samples were obtained for most constituents analyzed by collecting depth-integrated subsamples at equal-width increments (EWI) across the stream channel using either a US DH-81 or a US D-77 sampler (Edwards and Glysson, 1988; Shelton, 1994). Suspended and dissolved organic carbon samples were collected separately in a baked amber glass bottle using a weighted sampler at a single midstream vertical (Shelton, 1994). VOC samples were collected with a specially designed Wildco sampler, which has 40-milliliter (mL) vials placed directly in the sampler to collect a sample at a single point in the stream (Shelton, 1997). In addition, autosamplers were used to collect storm composite samples at a single point in the stream by programming the sampler to collect subsamples throughout the rise and fall of stream discharge (fig. 4). Autosamplers were used because of the short time period for the rise and fall. Field measurements of specific conductance, pH, water temperature, dissolved oxygen, and alkalinity were collected with the samples.
Samples were processed in the field immediately after collection to reduce the chance of chemical or biological alteration. After collection, all samples were composited in glass bottles using a cone splitter. Major cation and anion samples were filtered through a Gelman capsule filter of 0.45-micrometer (mm) pore size, and pesticide samples were filtered through a glass fiber filter of 0.45-mm pore size. Suspended and dissolved organic carbon samples were filtered through a silver filter of 0.45-mm pore size (Shelton, 1994). Major cation samples were preserved with nitric acid (HNO3), and VOC samples were preserved with concentrated hydrochloric acid (HCl).
Suspended and dissolved organic carbon filtering equipment was cleaned with organic-free deionized water. The pesticide filter assembly and VOC sampler were cleaned with dilute phosphate-free detergent solution, deionized water, and methanol. All other equipment used to collect and process samples was cleaned with a dilute phosphate-free detergent solution, rinsed with tap water, soaked in 5-percent HCl, rinsed with deionized water, rinsed with methanol, and then air dried (Shelton, 1994; 1997).
Quality-control samples comprised field blanks, equipment blanks, replicates, trip blanks, and field matrix spikes. These quality-control samples are described in Mueller and others (1997). One field blank and one replicate for major cations and anions, nutrients, suspended and dissolved organic carbon, and suspended sediment were collected each month during the study. In addition, four pesticide field blanks and three VOC field blanks were collected during the study. Two equipment blanks for trace elements and nutrients were collected. Four sets of split-concurrent replicates were collected during the study and analyzed for major cations and anions, nutrients, suspended sediment, and pesticides. One replicate and one trip blank were collected for VOC analysis. One field matrix spike was collected for pesticides and one for VOCs.
Quality-control samples were obtained for autosamplers by collecting an EWI sample at the same time that a discrete one-time sample was collected by the autosampler. This procedure was to help determine if there was bias in chemical concentrations from samples collected at a single point in the stream with an autosampler compared to samples collected using the EWI method, which might help determine a correction factor for chemical concentrations of samples collected with an autosampler. Field blank samples also were collected with the autosamplers.
Back to: Table of Contents
Forward to: Nutrients