Environmental setting, water quality, and ecological indicators of surface-water quality in the Mermentau River Basin, southwestern Louisiana, 1998-2001

Abstract

The U.S. Geological Survey collected data from 29 wells and 24 surface-water sites in the Mermentau River Basin, 1998-2001, to better understand ground-water and surface-water quality; aquatic invertebrate communities; and habitat conditions, in relation to land use. This study was a part of the National Water-Quality Assessment Program, which was designed to assess water quality as it relates to various land uses. Water-quality data were evaluated with criteria established for the protection of drinking water and aquatic life, and bed-sediment data were compared to aquatic-life criteria. Water-quality and ecological data were analyzed statistically in relation to drainage area and agricultural land-use intensity. Concentrations of nutrients and major inorganic ions in ground water and surface water generally were highest in the southeastern part of the study area where soils contain thick loess deposits. Peak concentrations of nutrients in surface water occurred March-May at two sites with high agricultural intensity; the lowest concentrations occurred August-January. The greatest potential for eutrophic conditions in surface water, based on nutrient concentrations, existed March-May, at about the same time or shortly after ricefields were drained. Secondary Maximum Contaminant Levels established by the U.S. Environmental Protection Agency (USEPA) were exceeded for sulfate, chloride, iron, or manganese in samples from 20 wells, and for iron or manganese in samples from all surface-water sites. Fewer pesticides were detected in ground water than in surface water. In 11 of the 29 wells sampled, at least one pesticide or pesticide degradation product was detected. The most frequently detected pesticides or pesticide degradation products in ground water were the herbicides bentazon and atrazine. Concentrations of 4 7 pesticides and degradation products were detected in surface water. At least 3 pesticides were detected in all surface-water samples. In 72 percent of the samples at least 5 hydrophylic pesticides were detected, and in more than 70 percent of the samples at least 3 hydrophobic pesticides were detected. Although atrazine concentrations in three samples collected in the spring exceeded 3 ~giL (micrograms per liter), the USEPA Maximum Contaminant Level of 3 ~giL was not exceeded because it is based on an annual average of quarterly samples. Concentrations larger than 3.0 ~giL were not detected in samples collected during other times of the year. Tebuthiuron was detected at all surface-water sites; the largest concentration (6.33 ~giL) was detected at a site on Bayou des Cannes, and was the only detection that exceeded the criterion (1.6 ~giL) for the protection of aquatic life. Malathion was detected at 16 surface-water sites; the largest concentration (0.113 ~giL) was detected at a site on Bayou Lacassine, and was the only detection that exceeded the criterion (0.1 ~giL) for the protection of aquatic life. Concentrations of fipronil exceeded numeric targets for acute total maximum daily loads (2.3 ~giL) at 3 sites and chronic total maximum daily loads (4.6 ~giL) at 14 sites. Maximum pesticide concentrations in surface water usually occurred in the spring at about the same time or shortly after ricefields were drained. Concentrations of DDE in bed sediment at two sites exceeded interim freshwater sediment quality guidelines for the protection of aquatic life. Fipronil sulfide and desulfinylfipronil were detected at all 17 sites from which bed-sediment samples were collected, but there are no current (2002) guidelines with which to evaluate the environmental effects of fipronil and degradation products. Two methods were used to group the ecological data-collection sites: (1) Sites were grouped before data collection (according to the study design) using drainage area and agricultural land-use intensity, and (2) sites were grouped statistically after data collection using canonical correspondence analysis (CCA) and classification (cluster analysis) techniques on surface-water quality, habitat, and aquatic invertebrate data. Aquatic invertebrate communities were used as ecological indicators of surface-water quality and habitat conditions at these sites. The CCA identified four significant environmental variables (instream cover score, percentage of open canopy, concentrations of dissolved oxygen, and maximum concentrations of dissolved fipronil) that described the distribution of aquatic invertebrate communities among ecological data-collection sites. Results from the CCA were used in a cluster analysis to identify four site groups that had similar water quality, habitat, and aquatic invertebrate characteristics. Environmental variables and biological metrics within the study-design (a priori, before sampling) and CCA-assigned (posteriori, after sampling) site groups were compared. Median values of 17 water-quality variables were lowest at sites in the northern part of the study area, where less than 45 percent of a drainage area is used for rice cultivation. Median values of 11 water-quality variables were highest at sites in the southeastern part of the study area, where the percentage of a drainage area used for rice cultivation varies. Median values of turbidity, and concentrations of total ammonia plus organic nitrogen, nitrate, total phosphorus, and dissolved fipronil, were highest at sites in the north-central part of the study area. Possible explanations for the differences in water quality among ecological data-collection sites may be the differences in (1) general soil composition and drainage characteristics, and (2) percentage of land used for agriculture in these basins. Habitat characteristics including channel size and morphology, water clarity, open canopy, and substrate differed between streams in the northern and southern parts of the study area. Stream habitat ratings were based on the total of 10 habitat parameter scores, using the Rapid Bioassessment Protocols habitat characterization. Scores increase as habitat quality increases. Ratings were suboptimal (102-154) to optimal (155-200) for 16 of the 19 ecological data-collection sites. Three sites were rated marginal (49-101). Differences in channel size, bank stability, and pool substrate may account for some differences in aquatic invertebrate communities between site groups distinguished by agricultural intensity. Organisms tolerant of turbidity, organic enrichment, and low dissolved-oxygen concentrations were common in the study area and dominated the aquatic invertebrate community. Metrics for aquatic invertebrate communities were significantly different (p ~ 0.05) among agri2 cultural land-use intensity site groups and CCA site groups in (1) percentage ofnoninsects, (2) abundance of midge taxa, (3) abundance of feeding groups, and (4) number of tolerant organisms. Dominance and diversity metrics were significantly different (p ~ 0.05) among CCA site groups. In this report, the maximum concentration of dissolved fipronil was the only significant environmental variable related to consistent decreases in relative abundance of many species, notably midges. Low species abundance in this report was associated with lower concentrations of fipronil degradation products than of the parent compound fipronil.