Introduction

    Surface water and sediment were sampled from three regional reservoirs, Brantley Lake, Lake Carlsbad and Red Bluff Reservoir, from January to March 1998. As described in the WIPP EM project section of this report, Brantley Lake and Red Bluff Reservoir are impoundments located "upstream" and "downstream", respectively, relative to surface and ground water flows from the area immediately surrounding the WIPP site. Both reservoirs support a warm-water fishery and are used for irrigation, livestock watering, wildlife habitat and recreation. Lake Carlsbad is an impounded section of the Pecos River within the city of Carlsbad that is used extensively by the local population for recreational warm-water fishing, boating and swimming. In addition, it can be used for industrial water supply, livestock watering, and wildlife habitat (20 NMAC 6.1, 1995, State of New Mexico Standards for Interstate and Intrastate Streams).

    The analyses reported here represent an expansion of studies of sediment and surface water in Brantley Lake that were summarized in the CEMRC 1997 Report. The 1998 samples were analyzed for selected radionuclides, chemical elements and compounds as part of WIPP EM baseline characterization.

Methods

    Sediment samples were collected at four randomly selected locations within the deep basins of each reservoir (Fig. 25, 26 and 27). Deep basins were chosen for sampling to minimize the disturbance and particle mixing effects of current and wave action that occur at shallower depths. Also, many of the analytes of interest tend to concentrate in the fine sediments that settle in the deep reservoir basins; thus, measurements from these areas would typically represent the highest levels that might be expected for a given reservoir.

    Sediments were collected to depths of 5-10 cm using a grab sampler or Eckman dredge, to obtain > 6 L of sediment at each sampling site. In the field, a 1-L aliquot of each sample was sealed in a pre-cleaned plastic jar for inorganic analyses, and > 5 L of wet sediment were sealed in a plastic bucket and transported to CEMRC for preparation prior to radiological analyses.

    In the laboratory, samples destined for radiochemical analyses were dried at 105°C to a constant weight, pulverized and homogenized prior to analysis. A 250-g aliquot of each sample was sent to Duke Engineering & Services (Bolton, Massachusetts) where it was analyzed for the alpha-emitting radionuclides ^239,240Pu, ²²⁸Th, ²³⁰Th, ²³²Th, ²³⁴U, ²³⁵U and ²³⁸U. The remainder of the sediment samples was archived for future analyses.

    Samples analyzed for inorganic chemical constituents were dried at 60 ^oC, and pulverized prior to analysis. Following preparation, the samples were analyzed by CEMRC using ion chromatography (IC), inductively-coupled plasma emission spectrometry (ICP-ES), and atomic absorption spectrometry (AA), as described for soils.

    In addition to the above analyses, a 1.5-L aliquot of one sediment sample from each reservoir was sent to A&L Plains Laboratory (Lubbock, Texas) (A&L) for several analyses that could not be performed in CEMRC laboratories (i.e. organic analyses) and for inter-laboratory comparison.

    Surface water was sampled at two locations in the both Brantley Lake and Red Bluff Reservoir, and at one location in Lake Carlsbad. Within each reservoir, surface water samples were taken in the same general area as the sediment samples. At each sampling location, one sample was collected from the surface (~ 0.5 - 1 m depth) and a second sample was collected from approximately 0.5 - 1 m above the sediment bed. The sample volume collected for radiological analyses was approximately 105 L. In addition, two 1-L samples were collected for inorganic analyses by CEMRC. At Red Bluff Reservoir and Lake Carlsbad, additional 3-L surface samples were collected for analysis of substances regulated under the Safe Drinking Water Act (SDWA), plus selected compounds and elements identified as possible constituents of waste to be deposited in WIPP. These analyses were performed by the NMSU SWAT Laboratory. Samples collected in 1997 from Brantley Lake were analyzed for these substances and reported in the CEMRC 1997 Report. All water samples for analysis by the SWAT Laboratory were collected and handled in accordance with EPA guidelines.

    In the laboratory, the water collected for radiological analyses was filtered to 0.2 mm using a high-volume filtration unit, acidified with HNO₃ to a pH < 2, and a 3-L aliquot was removed for analysis of gamma-emitting radionuclides (⁴⁰K, ⁶⁰Co and ¹³⁷Cs). Surface water samples collected for inorganic analyses by CEMRC were analyzed following applicable EPA standard methods.

Results

    Although the sediment concentrations of the U and Th isotopes were variable within and between reservoirs, the isotopic ratios were very similar across all three reservoirs. The reservoirs appeared to be slightly enriched in ²³⁴U compared to ²³⁸U, with mean ²³⁴U/²³⁸U activity ratios ranging form 1.4 to 1.6. Mean ²²⁸Th/²³²Th ratios were close to unity, ranging from 1.2 to 1.0, indicating that the isotopes are in secular equilibrium within the sediments. In a review of U and Th sediment chemistry, Onishi, Y., et. al. (1981, Critical Review: Radionuclide Transport, Sediment Transport, and Water Quality Mathematical Modeling; and Radionuclide Adsorption/Desorption Mechanisms, NUREG/CR-1322, Pacific Northwest Laboratory, Richland, Washington), reported that U has been found to be much more soluble than Th in the water column of oxidizing and alkaline environments. Th is relatively insoluble under alkaline conditions and adsorption to suspended particulate is rapid. In addition, ²³⁸U decays to ²³⁴U via the relatively short-lived ²³⁴Th (T_1/2 = 24.1 days). These chemical and physical properties result in an enrichment of ²³⁴U in sediments relative to ²³⁸U because the intermediate Th isotope is scavenged from the water column and deposited on the sediments before the decay to ²³⁴U occurs.

    These concentration data of the individual radionuclides in sediments suggest that there may be differences between lakes. In particular, sediment concentrations in Lake Carlsbad appear to be lower than in the other two reservoirs. A multiple analysis of variance (MANOVA) was used to test the null hypothesis that there are no significant differences in radionuclide concentrations between reservoirs. The MANOVA results (Wilks’ Lambda = 0.061, F = 1.304, numerator df = 14, denominator df = 6, P > 0.392) suggest that there were no significant differences between the lakes. The relatively high variance in concentrations between samples within each reservoir combined with the small number of samples in each reservoir (N = 4 per reservoir) undoubtedly constrains the power of the MANOVA to detect differences.

    Comparing measurements made on sediment samples collected from Brantley Lake during January 1998 to measurements from samples collected in the same general area of the lake during March/April 1997, the mean value for ^239,240Pu was approximately 30% higher (0.39 ± 0.089 mBq g^-1) in 1997 analyses (Fig. 31). In contrast, the mean Th and U measurements were 30-120% higher in the 1998 analyses. The reasons for these differences are not clear. However the number of samples collected was relatively small and the analyses were performed by different laboratories using different analytical methods. Additional future sampling will help clarify whether differences are real or an analytical artifact.

    The sediment and soil concentrations were compared by calculating the mean activity concentration of all samples of a given type (pooled across all reservoirs and soil sampling sites in the vicinity of the WIPP site) and then taking the ratio of the means. Sediments had higher activity concentrations than soil for all radionuclides measured with concentration ratios ranging from 1.7 for ^239,240Pu to 7.2 for ²³⁴U (Fig. 32). The observed difference in concentrations is not surprising because reservoir sediments are often a sink or integrator for many contaminants as the soil in the surrounding watershed is leached and eroded. One of the primary factors that may influence contaminant concentrations in both sediment and soil is the particle size distribution. Particle size analyses on representative samples of sediment and soil revealed a substantial difference in the proportion of fine and coarse-grained particles in each media. Sediments in the three reservoirs studied contained between 38-52% silt and clay-sized (< 63 m m) particles, with clay-size particles dominating the size distribution. In contrast, soils collected in the vicinity of the WIPP contained 88-90% sand-size particles and only 10-12% small particle sizes. As noted in the methods, the sediment sampling focused on the profundal zones (deep, undisturbed basins) of the reservoirs, that are characterized as zones of accumulation for fine-grained sediments. It is well documented that many contaminants (including many radionuclides) are found in higher concentrations in the fine-grained particles (Mudroch, A. and J.M. Azcue, 1995, Manual of Aquatic Sediment Sampling, Lewis Publishers, Ann Arbor, Michigan; Hakanson, L. and M. Jansson, 1983, Principles of Lake Sedimentology, Springer-Verlag, New York.; Onishi, et. al. 1981). One method that is commonly used to correct the concentrations for differences in the amount of the fine-grained silts and clays, is to normalize the concentration data to the amount of Al present in the samples. In sediments and soils, Al occurs primarily as alumino-silicates which are primarily associated with the silt and clay fractions in both media (Mudroch and Azcue, 1995) and can be used to represent the amount of fine-grained material in these media. Using this technique to correct for particle size, the radionuclide activity: Al concentration ratios ranged from 0.2 (^239,240Pu) to 0.9 (²³⁴U) which suggests that Al-normalized soil concentrations are very similar, or in the case of ^239,240Pu even slightly higher, than in the sediments.

    Brantley Lake sediments were relatively high in Al, Co, Cr, and Fe while Red Bluff Reservoir sediments were high in Na and chloride (Fig. 35 and 36). Lake Carlsbad typically had the lowest sediment concentrations for many of the analytes, including Al, Ba, Be, Co, Fe, K, Mo, Mn, Na, Ni, V, chloride, and sulfate. The high Na and chloride concentrations observed in Red Bluff water and sediments was not unexpected given that the reservoir is located down-gradient from a number of highly saline lakes or playas and saline aquifer inputs. The ratio of the mean concentrations (all reservoirs combined) observed in sediment versus surface water samples (sediment/water) ranged between 1x10² and 1x10⁴. Al, Fe and Mn occurred in the highest ratios, with lower ratios for more soluble analytes (Na, chloride and sulfate) (Fig. 37).

    As was the case for radionuclides, inorganic analyte concentrations in sediments were typically 10 to 100 times higher than in soils. As previously noted, these differences may be due in large part to differences in the particle size distributions of the sediment and soil samples. When corrected for particle size by normalizing to Al, sediment and soil concentrations appear more similar, with the normalized sediment/soil ratios ranging from 0.5 to 3 for most of the analytes (Fig. 38). For all but four of the analytes, the normalized sediment and soil concentrations were within a factor of 10, with the concentration ratio for selenium at 12. Three of the analytes, Na (ratio = 71), sulfate (ratio = 98) and chloride (ratio = 240), appeared to be enriched in sediments relative to soil.

    In addition to the CEMRC analyses, the NMSU SWAT laboratory and A&L Plains laboratory performed several analyses on surface water and sediment samples. The SWAT laboratory analyzed one surface water sample from Red Bluff reservoir and Lake Carlsbad for a large number of volatile organics, semi-volatile organics, metals and secondary analytes that are either regulated under the Safe Drinking Water Act (SDWA) or that will be constituents of the waste going into WIPP. It should be noted that none of the reservoirs serve as primary drinking water sources and therefore are not subject to regulation under the SDWA.

    In Lake Carlsbad and Red Bluff Reservoir, all of the analyses of organic compounds in surface water samples were below method detection limits. In addition, all metals and secondary analytes were either below detection limits or below the regulatory reference levels. In 1997, surface water from Brantley Lake was analyzed for the same chemicalconstituents with similar results (presented in the CEMRC 1997 Report).

    One sediment sample from each of the three reservoirs was analyzed by A&L for BTEX (ethylbenzene, m,p-xylene, o-xylene, toluene) and TPH (total petroleum hydocarbons), compounds typically associated with contamination from oil and gas operations. BTEX concentrations in all of the sediment samples were below the method detection limits. The sediment concentrations of TPH were 10, 52 and 118 mg kg^-1 in Brantley Lake, Red Bluff Reservoir and Lake Carlsbad samples, respectively. For comparative purposes, in New Mexico, remediation of TPH contaminated soils is considered adequate when TPH concentrations drop below 1000 mg kg^-1 (20 NMAC 9.1, section 708).