Mercury concentrations in muscle tissue from sportfish in Lake Mead, Nevada

Abstract

The aim of this study was to determine the concentrations of mercury present in commonly consumed fish from Lake Mead and to identify if differences exist between the 4 major basins. To date, no formal study using US Environmental Protection Agency (USEPA) approved methodology has been conducted to quantify the amount of mercury present in fish tissue from Lake Mead. Largemouth bass (n = 49), striped bass (n = 94) and channel catfish (n = 78) were collected from selected sites in Boulder Basin, Overton Arm, Virgin Basin and Gregg Basin of Lake Mead. Muscle tissue was homogenized, digested and analyzed for mercury in accordance with USEPA Method 245.6. Mean total mercury concentrations were (± SD) 0.089 ± 0.065, 0.154 ± 0.127 and 0.098 ± 0.080 μg/g in largemouth bass, striped bass and channel catfish, respectively. There was a significant overall difference in mean mercury concentration between fish from the 4 major basins of Lake Mead (F_3,208= 20.541, p < 0.001). The mean mercury concentration in Boulder Basin was significantly lower than all other locations (p < 0.001). Of 221 samples analyzed, 10 samples were found to have mean mercury concentrations above the USEPA's 2001 tissue residue criterion of 0.3 μg methlymercury/g fish tissue.

Key words:

• aquatic toxicology
• Lake Mead
• mercury
• sportfish

Lake Mead, Nevada is currently the largest man–made reservoir in the United States. It is formed by water impounded from the Hoover Dam and spans approximately 110 miles over the states of Nevada and Arizona on the Colorado River. Inflows to the lake include the Colorado, Virgin and Muddy rivers, as well as the Las Vegas Wash, which carries discharge from municipal wastewater treatment plants and storm– and rainwater runoff from the Las Vegas Valley (LaBounty and Horn 1997, Gerstenberger and Eccleston 2002). Lake Mead distributes water to communities in Southern California and Nevada and is used for a variety of recreational activities including sportfishing. According to a survey conducted by Gerstenberger and Eccleston (2002), an average of 22.7 ± 3.6 fish meals per year are consumed from Lake Mead by sport fishermen and their families, although the maximum number reported reached more than 300 meals per year.

Sportfishing at Lake Mead provides both recreational and health benefits to residents and visitors alike. The health benefits of consuming fish are well established: fish are relatively low in fat and cholesterol, high in protein and often provide a good natural source of omega 3 fatty acids. However, these benefits can be negated by the health risks posed by the toxic effects of mercury (Hg) and other contaminants found in fish tissues. All bodies of water contain at least small amounts of mercury from both natural and anthropogenic sources. An increase in human activities that contribute to the environmental burden of mercury has raised public concern as to the safety of fish caught in these waters. To address this issue, in 2001 the US Environmental Protection Agency (USEPA) announced a tissue residue criterion of 0.3 μg methylmercury/g fish tissue wet weight (ppm). This value represents the concentration of methylmercury in freshwater and estuarine fish and shellfish that should not be exceeded to protect consumers in the general population (USEPA 2001). For sensitive populations such as pregnant and nursing women and their infants and children, the USEPA and United States Food and Drug Administration (USFDA) issued a joint advisory in 2004 with more stringent recommendations. The joint advisory recommends consuming no more than two 6 oz meals of fish that are low in mercury per week (USFDA and USEPA 2004).

In this study, total mercury concentration was measured in whole fish filets. While the USEPA tissue residue criterion utilizes methylmercury for its standard, previous research indicates that more than 95% of mercury present in fish muscle tissue is in the organic form of methylmercury (Bloom 1992). Therefore, for the purposes of this study, total mercury concentration is considered to be within 5% of methylmercury concentration in fish and will hereafter be simply referred to as mercury.

Previous studies have been conducted on the presence of mercury in fish in Lake Mead (Cizdziel et al. 2002); however, the methods of these studies have not been in accordance with the federal guidelines for quantifying mercury in edible fish tissues and therefore could not be used to determine the need for a fish consumption advisory. This study was proposed in response to the protocol developed by Gerstenberger and Eccleston (2002) for a fish contaminant monitoring program on Lake Mead. The primary objective was to assess and evaluate the concentrations of mercury present in 4 species of fish commonly caught and consumed from Lake Mead: largemouth bass, striped bass, channel catfish and blue tilapia. Samples were taken from each of the 4 major basins of the Lake: Boulder, Virgin, Gregg and Overton Arm (Fig. 1) and analyzed for mercury content in relation to K factor (a measure of fish condition, equaling (W/L³)* 100), trophic level and location.

Figure 1 Lake Mead sampling locations.

Materials and methods

Sample collection

Four species of sportfish were collected from Lake Mead during October of 2007 and 2008 including 49 largemouth bass (Micropterus salmoides), 94 striped bass (Morone saxatilis), 78 channel catfish (Ictalurus punctatus) and 31 blue tilapia (Oreochromis aurea). Samples of fish from each species were taken at selected sites from the Nevada Division of Wildlife (NDOW) annual fish survey within the 4 major basins of Lake Mead: Boulder Basin, Virgin Basin, Overton Arm and Gregg Basin (Fig. 1). Fish were collected by 1 of 2 methods established by NDOW and Arizona Game and Fish (AZGF): the suspension of vertical gill nets in water overnight or electrofishing with pulsed DC current for 900 s. Fish were weighed to the nearest gram and measured to the nearest 0.1 cm (total length) and were placed in labeled plastic bags or aluminum foil and stored on ice until return to the laboratory. The condition factor (K) of each sample was calculated using the formula (W/L³)*100 (Williams 2000).

Sample preparation

Fish were filleted and skin was removed from each sample using an electric fillet knife. Fillets were rinsed with deionized water and frozen individually in labeled plastic freezer bags. Upon thawing, fillets were homogenized in a clean glass beaker using a Kinematica® Polytron PT 6100 homogenizer (Lucerne, Switzerland) for approximately 2 min or until sufficiently homogenized. Alternatively, larger samples were homogenized by passing the fillet through a Cabela's® Pro450 professional meat grinder (Sidney, NE) and collecting in a clean beaker. This process was repeated approximately 3 times for each fillet or until sufficiently homogenized. Individual homogenized fillets were stored in clean, labeled Whirl–packs® at –20 degrees C. All labware used in the fillet and homogenization procedures were cleaned using 2% Citronox®, deionized water and 10% HNO₃ between samples.

A microwave oven, Anton–Parr Multiwave™3000 (Graz, Austria), with teflon vessels was used in a microwave–assisted acid digestion of tissues. Approximately 2 g of fish tissue was combined with 4 mL HNO₃ and 4 mL deionized H₂0 and microwaved with increasing power to approximately 1200 W and 160 C. After cooling, 4 mL 5% amidosulfonic acid was added to the solution to minimize nitric acid fumes, which can interfere with Hg analysis. The vessels were rinsed 3 times with deionizedwater for a total volume of 25 mL. For analysis, a 1 mL aliquot of the raw digested material was transferred into a separate clean and labeled centrifuge tube containing 5 mL 3% HCl. This 1:6 solution was used in most cases for analysis using the PerkinElmer® Flow–Injection Mercury System 100 (FIMS 100; Sheldon, CT). The ratio of digestate to HCl was adjusted as necessary to allow the concentration of mercury in the sample to fall within the range of the calibration curve.

Sample analysis

Total mercury was analyzed in accordance with USEPA Method 245.6 using a Perkin–Elmer FIMS 100 equipped with an AS–91 autosampler using the flow–injection mercury cold–vapor technique. The software program WinLab™32 for AA was used in conjunction. The instrument detection limit is reported to be 0.2 ng/g. The method detection limit was calculated to be 0.010 μg/g.

The FIMS 100 was calibrated at the beginning of each analysis day. Calibration standard solutions of 0.0, 0.5, 1.0, 2.5, 5.0 and 10.0 ug/L were prepared from 1000 ug/mL Hg in 5% HNO₃ JT Baker® stock reference solution (Phillipsburg, NJ) by serial dilution.

Fish samples were analyzed on a wet weight basis according to the following manufacturer issued conditions: a carrier gas of argon with an inlet pressure of 350 kPa, a carrier solution of 3% hydrochloric acid (v/v) and a 5.5% stannous chloride (w/v) reducing agent. The FIMS program used included a prefill time of 15 s followed by a sampling time of 10 s, during which the sample was injected into the carrier stream. The sample was mixed with the carrier gas and the reaction products entered into a gas liquid separator. Next, the gas phase was transferred into a glass cell in which absorption of mercury vapor was measured over 20 s. The absorbance was plotted versus time, and peak height was measured. Peak height of the sample was compared to the initial calibration curve, and the sample concentration of Hg was measured. Three replicates were performed on each sample and an average of the 3 measurements was reported.

Quality assurance/quality control

Quality assurance and quality control were ensured by performing a calibration with calibration blank each day prior to analysis. A 0.995 or higher correlation coefficient was considered acceptable for the calibration curve. Each microwave digestion tray contained 16 samples including a reagent blank; 2 certified standard reference materials: National Research Council Canada DORM–3 dogfish muscle tissue (Ontario, Canada) and National Institute of Standards and Technology Standard Reference Material® 1946 Lake Superior Fish Tissue (Gaithersburg, MD); and a replicate sample of fish muscle tissue spiked with a known amount of liquid Hg. For further assurance, 10% of samples were randomly selected for duplicate analysis. A recovery between 80 and 120% of expected value was accepted for the standard reference materials (SRM), duplicate samples and spiked samples. Samples were not recovery corrected.

Statistical approach

All data were analyzed using SPSS® version 16.0 (Chicago, IL). Mercury concentrations were compared among basin and species and were adjusted for length using an ANCOVA Type IV sum of squares error. The data were log₁₀transformed to meet the assumptions of the statistical test. The log₁₀ of the mean mercury concentration was the dependent variable, and trophic level and location were the independent variables. The effect of length on mercury concentrations was controlled through its use as a covariate. All data reported are length–adjusted unless otherwise indicated.

Results

K Factor

K factors in all species ranged from 0.40 to 1.67. Overall, largemouth bass had the highest mean K factor among the 3 species (± SD; 1.27 ± 0.15), channel catfish had the lowest mean K factor (0.79 ± 0.17) and striped bass had an intermediate mean K factor (0.81 ± 0.17). The data showed a consistent pattern with respect to K factor among basins. Fish of all 3 species had the highest K factor in Boulder Basin followed by Gregg Basin, Overton Arm and Virgin Basin (Fig. 2).

Figure 2 Mean K factors [(W/L³)*100] of fish from Lake Mead by species and location in Lake Mead. Error bars indicate standard deviation of the mean.

A bivariate correlation was used to evaluate K factor as a covariate. Overall, K factor was found to be significantly correlated at the 0.01 level (one–tailed) with the log₁₀ of mercury concentration (r =−0.272, p < 0.001). An ANCOVA using K factor as the covariate did not reduce the interaction (F = 2.171, p = 0.047) between the independent variables species and location. A species–specific analysis revealed that a significant correlation between K factor and log₁₀ of mercury concentration only existed for striped bass (r =−0.536, p < 0.001). No significant correlation was found for largemouth bass or channel catfish. Therefore, K factor was found not to be a suitable covariate for these data; however, K factor is a measure of fitness of a fish and not size. To better understand the relationship between mercury concentration and size of fish, length was evaluated independently from weight as a possible covariate.

Length

Length was evaluated as a covariate and found to be positively correlated overall at the 0.01 level (one–tailed) with the log₁₀ of mercury concentration (r = 0.234, p < 0.001). A simple linear regression (with outliers removed) was performed between the log₁₀ mercury concentration (μg/g) and length (cm) for each species as well as for all species combined (Fig. 3). Length accounted for approximately 10% of the variability in mercury concentrations overall. In largemouth bass, striped bass and channel catfish length accounted for about 35, 35 and 0.05% of the variability in mercury concentrations, respectively. An ANCOVA utilizing length as a covariate reduced interaction between the independent variables species and location so that it was no longer significant (F = 1.261, p = 0.277). For these reasons, length was determined to be an effective covariate for the data.

Figure 3 Simple linear regression plots of length (cm) vs. 5 % trimmed mercury concentration (μg/g) for largemouth bass, striped bass, channel catfish and all species combined.

Trophic level

Length–adjusted mean mercury concentrations for largemouth bass, striped bass and channel catfish were calculated (Table 1). Blue tilapia (n = 31) were also collected, but all mean mercury concentrations in tilapia were below our method limit of detection of 0.010 μg/g and therefore were not included in the statistical model. The ANCOVA indicated a significant difference between mercury concentrations among the 3 species (F_2,208= 22.488, p < 0.001). Contrasts revealed that mean mercury concentration of largemouth bass differed significantly (p = 0.001) from that of striped bass and channel catfish (p = 0.025). The means of striped bass and channel catfish also differed significantly (p < 0.001) from each other (Table 1). When blue tilapia were included in the model with mercury concentrations reported at the method detection limit of 0.010 μg/g, there was a significant difference (p < 0.001) between the mean mercury concentration of blue tilapia and the means of all other species.

Table 1 Length–adjusted mean mercury concentrations by species and location.

Species	Location	n	Mean Hg (g/g)	Std. Error	Mean Length (cm)	Std. Dev.	Mean Weight (g)	Std. Dev.
Largemouth bass* (n = 49)
	Boulder Basin†	12	0.054	0.081	34.9	6.3	647	242
	Overton Arm	9	0.094	0.061	34.8	7.1	574	302
	Virgin Basin	11	0.085	0.084	37.7	11.7	837	1125
	Gregg Basin	17	0.100	0.070	31.5	4.7	443	309
Striped bass* (n = 94)
	Boulder Basin†	22	0.061	0.060	41.1	4.0	692	222
	Overton Arm	23	0.132	0.059	36.0	8.5	372	408
	Virgin Basin	28	0.135	0.054	34.5	7.6	302	167
	Gregg Basin	21	0.185	0.061	39.5	8.1	551	218
Channel catfish* (n = 78)
	Boulder Basin†	19	0.028	0.066	44.4	7.7	924	428
	Overton Arm	23	0.093	0.061	45.4	9.0	845	811
	Virgin Basin	17	0.073	0.068	40.4	5.9	530	350
	Gregg Basin	19	0.073	0.064	37.5	10.4	496	782
*The mean mercury concentration of each species differed significantly (p < 0.05) from each other species.
†The mean mercury concentration in Boulder Basin differed significantly from the means in Overton Arm (p < 0.001), Virgin Basin (p = 0.002), and Gregg Basin (p < 0.001).

Because collection for the study took place over 2 consecutive years, mean mercury concentrations from each species were compared by year of collection to determine if the means differed over time. The means for each species did not significantly differ (F_1,208= 0.165, p = 0.685) from October of 2007 to October of 2008.

Location

Length–adjusted mean mercury concentrations for fish from Boulder Basin, Overton Arm, Virgin Basin and Gregg Basin were calculated (Table 2). The ANCOVA revealed a significant overall difference in mean mercury concentrations between the 4 major basins of Lake Mead (F_3,208= 20.541, p < 0.001). Contrasts showed that the mean mercury concentration of fish in Boulder Basin was significantly lower than that of the other 3 locations: Gregg Basin (p < 0.001), Virgin Basin (p < 0.001) and Overton Arm (p < 0.001). No significant difference was revealed between the mean mercury concentrations in Gregg Basin, Virgin Basin and Overton Arm.

For largemouth bass and striped bass, unadjusted mean mercury concentrations were lowest in Boulder Basin followed by Overton Arm, Virgin Basin and highest in Gregg Basin (Fig. 4). Channel catfish display a slightly different pattern with the lowest mercury concentration in Boulder Basin, followed by Virgin and Gregg Basins and the highest concentration in Overton Arm (Fig. 4).

Figure 4 Length– adjusted mean mercury concentrations (μg/g) in fish muscle tissue by species and location in Lake Mead. Error bars indicate standard deviation of the mean.

Discussion

Of the 221 fish collected for analysis, lengths ranged from 20.0 to 71.2 cm (mean 38.5 cm), weights from 95 to 4180 g (mean 584 g) and mercury concentrations (unadjusted) from 0.011 to 0.801 μg/g (mean 0.119 μg/g). Mercury concentrations were generally low relative to the USEPA tissue residue criterion for methylmercury in fish and shellfish of 0.3 μg/g (USEPA 2001). Of 221 samples collected, 10 exceeded the standard, 7 of which were striped bass nearing 20 in (50.8 cm) in length. Also exceeding the standard were 2 catfish and 1 largemouth bass.

Fish condition and body size

Mercury is a potent neurotoxin and exhibits the potential to bioaccumulate in food webs. Such bioaccumulation is the basis for exploring the relationship between mercury concentrations and measurements such as K Factor and fish length. The data from this study displayed an overall negative correlation between mercury concentration and K Factor (r =−0.272, p < 0.001). Prior studies from other bodies of water (Scott and Armstrong 1972, Scott 1974, Lange et al. 1994, Watras et al. 1998) have found fish condition to be positively correlated with mercury concentration. Upon analysis of individual species, the significant negative correlation only remained for striped bass. Channel catfish and largemouth bass showed no significant correlation between K factor and mercury concentration. One study of fish in Lake Mead from fall of 1998 also reported an inverse correlation between fish condition and mercury concentration in striped bass (Cizdziel et al. 2002). The authors suggested that because fall in Lake Mead is a lean–food season, the striped bass may have been experiencing a starvation period during which organs shrank and mercury became more concentrated in tissues (Cizdziel et al. 2002). A comparison of mercury concentrations of striped bass caught during spring and fall would be necessary to validate this theory (Dellinger et al. 1995).

Fish length was positively correlated with mercury concentrations overall (r = 0.234, p < 0.001). When species were evaluated independently, the relationship between length and mercury concentration remained for largemouth and striped bass. There was not a significant correlation between length and mercury concentration for channel catfish. Previous studies (Scott and Armstrong 1972, Scott 1974, Lange et al. 1994, Dellinger et al. 1995) have also reported fish length to be positively correlated with mercury concentration.

Although length was a predictor for mercury concentrations in largemouth bass and striped bass, some other contributing factors were not evaluated in this study. Some variables of interest include environmental factors such as temperature, seasonality, dissolved organic carbon (DOC), selenium and prey availability; and physiological factors such as age, growth rate, metabolism and diet. These parameters have the potential to affect mercury concentrations in fish; therefore, further research is necessary to elucidate this complex relationship.

Trophic level

Of the 3 species analyzed in this study, striped bass had the greatest mean mercury muscle concentrations (Table 1). Because striped bass are predominantly piscivores, one would expect mercury concentrations to be higher in striped bass than in species that consume a variety of prey at lower trophic levels. The mean mercury concentrations in largemouth bass were significantly higher than those in channel catfish (p = 0.025). This is consistent with trophic level expectations because higher level predators are subject to a greater risk of bioaccumulation through diet (Post et al. 1996, MacRury et al. 2002). Largemouth bass are carnivores, feeding on a mixture of shad, crayfish and certrarchids, while channel catfish are omnivores, feeding on a wide variety of fish, insects and detritus and therefore do not accumulate mercury as rapidly (Minckley 1973). Muscle mercury concentrations in blue tilapia were below the limit of detection for this method (0.010 μg/g); however, when included in the model at the method detection limit, mercury concentrations were still significantly lower than all other species. A diet consisting entirely of plants does not promote the bioaccumulation of mercury as quickly as a diet containing fish, so one would expect mercury concentrations in the herbivorous blue tilapia to be relatively low. The results of this study with respect to trophic level were somewhat consistent with those of Cizdziel et al. (2002) who reported striped bass to have the highest mercury concentrations followed by channel catfish, largemouth bass and blue tilapia. However, note that mercury concentrations in the aforementioned study were not adjusted for length or body size.

Location

The data from this study revealed that fish from Boulder Basin contained significantly lower mercury concentrations than all other locations analyzed (Table 2). There was no significant difference between the mean mercury concentrations in fish from Overton Arm, Virgin Basin and Gregg Basin. The sampling for this study was completed during October 2007 and 2008 with NDOW and AZGF during their annual fish survey of Lake Mead. Sampling sites were selected by these agencies to give an accurate representation of the fish populations found throughout the reservoir. Lake Mead consists of four large basins: Boulder, Virgin, Gregg and Overton Arm, separated by narrow canyons (LaBounty and Horn 1997; Fig. 1). Based on previous studies on the distribution of fish populations within the reservoir, the assumption is that the 4 basins sampled contain relatively distinct fish populations (Mueller and Horn 2004).

The results of this study with respect to basin are in concordance with those of Cizdziel et al. (2002) in that fish from Boulder Basin contained the lowest muscle mercury concentrations overall. Cizdziel et al. suggested that lower mercury concentrations could be due to a starvation concentration whereby fish with low K factors concentrate mercury to a higher degree during periods of low food availability. This hypothesis is not fully supported by the data from this study. For the starvation concentration theory to be supported, we would expect to see an inverse correlation between K factor and mercury concentration with respect to location. Gregg Basin had the second highest K factor for all species (Fig. 2) and the highest mercury concentrations for 2 of the 3 species, largemouth bass and striped bass (Fig. 4). From a locational perspective, there are some issues worthy of exploring to explain the disparity in mercury concentrations between Boulder Basin and the other locations, but these matters extend beyond the range of the data collected for this study.

Boulder Basin is the westernmost and most downstream basin of Lake Mead. This region receives inflow from the 2 main arms of the reservoir as well as all drainage and effluent from the Las Vegas Valley via the Las Vegas Wash, resulting in a great deal of nutrient loading relative to other basins (Paulson and Baker 1981, Prentki and Paulson 1983, LaBounty and Horn 1997). There is a known existing relationship between DOC and methylation of mercury within water and sediment. Gilmour and Henry (1990) reported that an increase of DOC in water results in a decreased rate of methylation; however, in sediment, increased DOC results in a greater rate of methylation of mercury. Due to inflow from the Las Vegas Wash, waters from Boulder Basin contain considerably elevated DOC levels, which decline with increasing distance from the Wash (LaBounty and Horn 1997). Rosario–Ortiz et al. (2007) completed a study of inflows to Lake Mead in 2005 and concluded that the Las Vegas Wash contained greater DOC than the upper and lower Colorado River as well as the Virgin and Muddy Rivers. Elevated DOC in waters from Boulder Basin with respect to all other locations could result in a decrease in methylation and thus lower concentrations of mercury in fish; however, a current calculation of DOC in water and sediment for each location would be necessary to draw any conclusions.

Another possibility for the disparity between mercury concentrations in Boulder Basin compared with all other locations is a possible increased concentration of selenium in Boulder Basin compared with other locations. Studies in freshwater lakes have demonstrated that the addition of selenium decreases the rate of methylmercury bioaccumulation (Pelletier 1986, Turner and Swick 1983, Chen et al. 2001). Because mercury and selenium experience a high affinity for one another, the formation of mercury–selenium complexes is thought to render both compounds biologically inactive (Moller–Madsen and Danscher 1991, Raymond and Ralston 2004). Selenium concentrations have been calculated recently in Boulder Basin because of the concern of selenium input from the Las Vegas Wash; however, data from other locations are not available at this time (Cizdziel and Zhou 2005; L. Blish, Southern Nevada Water Authority, December 2008, unpubl.). Further research in this area is advisable.

The limnology of Lake Mead has been altered in the recent past with the introduction of the quagga mussel (Dreissena bugensis), a nonindigenous invasive species of mussel. Quagga mussels were discovered on 6 January 2007 in Boulder Basin of Lake Mead (LaBounty 2007). After their discovery, their spread was monitored closely throughout all of Lake Mead. The mussels are currently present with varying densities in all basins of Lake Mead. Dreissena species are filter feeders, grazing on phytoplankton and seston in lakes and rivers. As such, Roditi and Fisher (1999) suggested they might be effectively utilized as bioindicators of freshwater contamination and have an impact on the cycling and residence times of metals such as mercury (1999). Based on recent estimates of quagga mussel density, Wong (2010) estimated that quagga mussels would effectively filter the entire volume of Lake Mead in 148.2 d (unpublished data). Although quagga mussels were discovered in January 2007, it is uncertain as to when they were first introduced into the reservoir. They were possibly present and filtering the water for some time before their discovery in Boulder Basin. If this were true, it might help to explain the disparity in mercury concentrations between Boulder Basin and the other locations within the reservoir, possibly due to the colonization of specific basins earlier than others.

Summary and conclusion

Lake Mead, Nevada is a sportfishing resource for residents of Nevada, Arizona, and visitors to these states. Fish caught from the 4 major basins of Lake Mead are often consumed by anglers and their families, with an average of approximately 23 fish meals per year consumed (Gerstenberger and Eccleston 2002). According to angler interviews (n = 150) conducted during 1999–2002, striped bass is the most commonly consumed fish species from Lake Mead constituting 70% of all fish consumption, followed by largemouth bass (12%) and channel catfish (11%; Gerstenberger and Eccleston 2002). To date, there have not been analyses of fish from Lake Mead utilizing USEPA–approved methodologies for quantifying mercury concentrations in edible fish tissue. One aim of this study was to fill this knowledge gap and determine if a mercury advisory is necessary for consumers of any of the popular fish species of varying trophic levels caught in Lake Mead. The mean mercury concentrations of fish from all trophic levels and all locations were below the USEPA tissue residue criterion of 0.3 μg/g; however, 10 samples of 221 collected contained mercury concentrations that were greater than this value. To portray these results from a practical standpoint, we compared the mean mercury concentrations from this study to the USEPA “Monthly Fish Consumption Limits for Noncarcinogenic Health Endpoint–Methylmercury” (USEPA 2000). This comparison revealed that approximately 4 fish meals per month of striped bass, 8 meals per month of channel catfish and largemouth bass and more than 16 meals per month of blue tilapia can be safely consumed by an adult of 70 kg, with an assumed meal size of 8 oz (USEPA 2000).

Second, we compared mercury concentrations in fish among the 4 major basins of the reservoir to determine if mercury concentrations are higher in some areas than others. Fish from Boulder Basin contained muscle mercury concentrations significantly lower than all other basins. Fish collected from Gregg Basin contained the highest mean mercury concentrations but were still well below federal guidelines for issuing a fish consumption advisory.

The conclusions drawn from these results are that largemouth bass, striped bass, channel catfish and blue tilapia from Lake Mead can be safely consumed, but high risk consumers should abide by certain guidelines when consuming fish from Lake Mead. Pregnant women, infants and children should always exercise caution when consuming fish because of the increased susceptibility of the developing nervous system to mercury. Common and effective ways to reduce mercury consumption include: eating smaller fish (<20 in), limiting the amount of fish meals, and reducing portion sizes.

Acknowledgments

We would like to gratefully acknowledge those who donated their time and expertise in collection of samples for this project, especially Mike Burrell of the Nevada Division of Wildlife and his crew, as well as Matt Chmiel, Andy Clark, and their crew from Arizona Game and Fish. Sincere thanks go to Dr. Chad Cross for statistical consultation. Technical support in the laboratory and field was also provided by Ms. Sara Mueting, Ms. Ashley Phipps, Ms. Lanisa Pechacek and Mr. Chris Rendina.