Detailed study of selenium and selected constituents in water, bottom sediment, soil, and biota associated with irrigation drainage in the San Juan River area, New Mexico, 1991-95

Abstract

In response to increasing concern about the quality of irrigation drainage and its potential effects on fish, wildlife, and human health, the U.S. Department of the Interior began the National Irrigation Water Quality Program (NIWQP) to investigate these concerns at irrigation projects sponsored by the Department. The San Juan River area in northwestern New Mexico was one of the areas designated for study. Study teams composed of scientists from the U.S. Geological Survey, the U.S. Fish and Wildlife Service, the Bureau of Reclamation, and the Bureau of Indian Affairs collected water, bottom-sediment, soil, and biological samples at 61 sites in the San Juan River area during 1993-94. Supplemental data collection conducted during 1991-95 by the Bureau of Indian Affairs and its contractor extended the time period and sampling sites available for analysis. Analytical chemistry performed on samples indicated that most potentially toxic elements other than selenium generally were not high enough to be of concern to fish, wildlife, and human health. Element concentrations in some water, bottom-sediment, soil, and biological samples exceeded applicable standards and criteria suggested by researchers in current literature. Selenium concentrations in water samples from 28 sites in the study area exceeded the 2-microgramper-liter (lg/L) wildlife-habitat standard. Vanadium concentrations in water exceeded the 100-Kg/L standard for livestock-drinking water at one site. In biota, selenium and aluminum concentrations regularly equaled or exceeded avian dietary threshold concentrations. In bottom sediment and soil, element concentrations above the upper limit of the baseline range for western soils were: selenium, 24 exceedances; lead, 2 exceedances; molybdenum, 2 exceedances;strontium, 4 exceedances; and zinc, 4 exceedances. Concentrations of total selenium in bottom-sediment and soil samples were significantly greater for Cretaceous than for non-Cretaceous soil types in the study area and were generally similar for habitats within and outside irrigation-affected areas. Mean and median total-selenium concentrations in samples from areas with Cretaceous soil types were 4.6 and 2.2 micrograms per gram (ps/g), respectively. Mean and median total-selenium concentrations in samples from areas with non-Cretaceous soil types were 0.6 and 0.15 pg/g, respectively. Samples from the study area had low concentrations of organic constituents. Organochlorine pesticides and polychlorinated biphenyls were detected in a few biological samples at low concentrations. Polycyclic aromatic hydrocarbon (PAH) compounds were not detected in whole-water samples collected using conventional water-sampling techniques. In tests involving the use of semipermeable-membrane devices to supplement conventional water assays for PAH's, low concentrations of PAH's were found at several locations in the Hammond Irrigation Supply Canal, but were not detected in the Hammond ponds at the downstream reach of the Hammond irrigation service area. PAH compounds do not appear to reach the San Juan River through the Hammond Canal. Data indicate that water samples from irrigation-drainage-affected habitats had increased mean selenium concentrations compared with samples from irrigation-delivery habitat. The mean selenium concentration in water was greatest at seeps and tributaries draining irrigated land (17 μg/L); less in irrigation drains and in ponds on irrigated land (61.tg/L); and least in backwater, the San Juan River, and irrigation-supply water (0.5 - 0.6 μg/L). Statistical tests imply that irrigation significantly increases selenium concentrations in water samples when a Department of the Interior irrigation project is developed on selenium-rich sediments. Water samples from sites with Cretaceous soils had significantly greater selenium concentrations than water samples from sites with non-Cretaceous soils. Water samples from Department of the Interior project irrigation-drainage sites developed on Cretaceous soils contained a mean selenium concentration about 10 times greater than those in samples from Department of the Interior project sites developed on non-Cretaceous soils. Selenium was much less concentrated in water than in bottom sediment, soil, or biota in the study area. The range in concentrations of dissolved selenium in water was less than 1 ptg/L to 37 1.1g/L (less than 1 to 37 parts per billion). The range in concentrations of total selenium in bottom sediment and soil was less than 0.1 to 23lig/g (less than 100 to 23,000 parts per billion). The range in concentration of selenium in biota was less than 0.1 to 24.0 fig/g (less than 100 to 24,000 parts per billion). Data indicated that bioaccumulation and leaching from soil were the important processes at the study area that lead to elevated levels of selenium. Other processes examined included: (1) evapoconcentration of selenium; (2) atmospheric deposition of aerosols containing selenium; and (3) contamination of surface water by point-source or non-point-source discharges. Selenium concentrations in biological samples were evaluated by a number of variables including: (1) media sampled (emergent and submergent plants, nektonic and benthic invertebrates, omnivore/herbivore and carnivore fish, and terrestrial and aquatic amphibians); (2) habitat (San Juan River main-stem reaches, backwaters, tributary reaches, irrigation delivery or drainage canals, and ponds); (3) irrigation project area and reference sites; and (4) soil type (non-Cretaceous or Cretaceous soils). Graphical techniques and nonparametric statistical tests were applied to determine the influence of selected physiographic variables on selenium concentrations in biological samples collected in the San Juan River area. Species of sucker and of smaller fish contained significantly higher selenium concentrations in the upstream portion of the river where a productive community of plants and animals is found that is associated with warming, nutrient-rich waters discharged from an upstream reservoir. Selenium concentrations in algae, odonates, and mosquitofish collected from both irrigation-drain and pond habitats underlain by Cretaceous soils were significantly greater than in those collected from similar habitats underlain by non-Cretaceous soils. Investigators conclude that the major factor affecting the variability of selenium accumulation in biota at aquatic habitats was the presence of underlying Cretaceous soils. Median selenium concentrations were less than 2 lAg/g for plant samples, less than 7 μg/g for invertebrate samples, and less than 6 lAg/g for whole-fish samples collected from aquatic habitats underlain by non-Cretaceous soils. Similar samples collected from aquatic habitats underlain by Cretaceous soils contained median selenium concentrations two to five times greater. Leaching of selenium from Cretaceous soils in the San Juan River area increases the accumulation of selenium concentrations in the biota and thereby increases the exposure and potential health risks associated with selenium to migratory birds, fish, and other wildlife that use these aquatic habitats extensively. Aquatic habitats presenting the greatest average exposure to excess selenium concentrations in the diets of resident wildlife are from consumption of plants, invertebrates, and fish at irrigation-drain habitats underlain by Cretaceous soils. Of the irrigation projects evaluated in the San Juan River area, the highest median selenium concentrations in algae, cattail leaves, odonate nymphs, mosquitofish, and leopard frog samples from the study area were collected from the east hogback irrigation drain.