NOAA Technical Memorandum NMFS-NWFSC-16

National Status and Trends Program National Benthic Surveillance Project: Pacific Coast Analyses of Elements in Sediment and Tissue Cycles I to V (1984-88) James P. Meador, Robert C. Clark, Jr., Paul A. Robisch, Donald W. Ernest, John T. Landahl, Usha Varanasi, Sin-Lam Chan, and Bruce B. McCain National Marine Fisheries Service Northwest Fisheries Science Center Coastal Zone and Estuarine Studies Division 2725 Montlake Blvd. E. Seattle WA 98112 August 1994 U.S. DEPARTMENT OF COMMERCE Ronald H. Brown, Secretary National Oceanic and Atmospheric Administration D. James Baker, Administrator National Marine Fisheries Service Rolland A. Schmitten, Assistant Administrator for Fisheries

EXECUTIVE SUMMARY

PREFACE

INTRODUCTION

METHODS

Site Selection and Descriptions

Field Sampling

Laboratory Analyses

Analysis of Elements in Sediment and Tissue

Extractable Elements

Statistical Analyses

Screening of Chemical Data for Accuracy

Display Techniques

Calculation of comparison intervals in graphs

Floating-bar plots

GT2 plots

Data Analysis

Determining background sediment concentrations

Methods for correlations

RESULTS AND DISCUSSION

Concentrations of Selected Elements in Sediments

Antimony, Arsenic, and Cadmium

Chromium, Copper, and Lead

Mercury, Nickel, and Selenium

Silver, Tin, and Zinc

Aluminum and Iron

Manganese and Silicon

General Discussion of Sediment Chemistry

Concentrations of Elements in Fish Livers

Fourhorn Sculpin, Spotted Sand Bass, Spotted Turbot, and Black Croaker

Flathead Sole

English Sole

Starry Flounder

White Croaker

Hornyhead Turbot

Barred Sand Bass

Concentrations of Elements in Fish Stomachs

Relationships Between Chemical Parameters

Sediment

Baseline metals and excess concentrations in urban sediments

Extractable metals

Correlations of elements in sediment

Tissue and Sediment

Liver tissue and stomach contents

Correlations of elements in liver tissue, stomach contents, and sediment

CONCLUSIONS

ACKNOWLEDGMENTS

CITATIONS

This technical memorandum summarizes and interprets the results for the metal and metalloid (elements) contaminants for the first 5 years (1984-88) of the Pacific coast portion of the National Benthic Surveillance Project (NBSP). As a component of NOAA's National Status and Trends Program, the NBSP determines the levels of contaminant chemicals and prevalence of pathological lesions in bottom fish and relates them to contaminant levels in associated sediment. The levels of organic contaminants and the prevalence of presumptive pollution-related liver and kidney lesions are documented in other technical memorandums and publications.

Employing uniform sampling protocols and state-of-the-art analytical methods, an extensive database has been developed, which includes detailed information on the distribution of selected elemental contaminants. These include the toxic elements antimony, arsenic, cadmium, chromium, copper, lead, mercury, nickel, selenium, silver, tin, and zinc as well as the major elements aluminum, iron, manganese, and silicon in surficial sediments, liver tissues, and in stomach contents of selected fish. Of the 51 sites sampled in Washington, Oregon, California (West Coast), and Alaska, 27 were located in or near urban centers. (Pacific coast sites include all West Coast and Alaskan sites, even though three of the Alaskan sites are in the Arctic.) The locations of the sites in urban embayments were selected to be as representative as possible to assess inputs to the area from multiple sources; hence, the results from individual sites should not be viewed as representative of entire embayments because of limited sampling.

The overall findings from the NBSP for the years 1984-88 indicate that the highest concentrations of most sediment-associated toxic elements were present in the highly urbanized areas. Results of measured concentrations in sediment and analysis of the nonurban sediment to derive natural background concentrations indicate that a number of urban sites contained elevated concentrations of metals and metalloids beyond what was expected for sediment that was relatively pristine. Based on the frequency of excess toxic element concentrations over background levels in sediment (excluding chromium and nickel), the sites most contaminated were Elliott Bay near Seattle; Oakland estuary in San Francisco Bay; San Pedro Outer Harbor, San Pedro Canyon, and west Santa Monica Bay in the Los Angeles area; and San Diego Bay (north and south), and National City in the San Diego area. Moderate contamination was found at Skagway (Alaska); Castro Creek, Oakland, Redwood City, and Hunters Point (all San Francisco Bay); and west Harbor Island (San Diego Bay). Often, sediment from a site that contained high concentrations of one toxic element also contained elevated concentrations of other toxic elements. The toxic elements found most often in excess of background concentrations in the sediments (>19 sites) were lead, nickel, and zinc. The next most frequent group of elevated metals consisted of copper, silver, and tin with excess values occurring for 11-13 of the sites. Another group of elements with less frequent overages (4-8 sites) consisted of antimony, arsenic, cadmium, chromium, mercury, and selenium. In terms of concentration, only copper, lead, nickel, and zinc were generally much higher (>2x) than their background or reference concentration. We conclude, by a weight of evidence argument, that those sites with concentrations of many elements elevated above background levels at all 3 stations for most years sampled are generally contaminated.

When correlations of elements in sediment were examined a few interesting patterns emerged. In West Coast urban sediments, copper, iron, lead, manganese, and zinc all correlated highly to each other. Iron and manganese are major components of sediment and their abundance is probably controlled in part by such factors as grain size and redox state. Copper, lead, and zinc were all correlated to each other probably due to common sources found in these industrialized areas. Based on correlations of iron and manganese, it appears that the background geochemical makeup of the Alaskan sites is generally the same as the West Coast sites (Washington, Oregon, and California). One difference between the Alaskan sites and the West Coast urban sites is a weak correlation between copper and zinc at Alaskan sites, which may be due to differences in the source of these metals. Alaskan sites do however display a strong correlation between zinc and lead, indicating a common source, possibly mining operations.

An examination of a few sites for extractable metals with 1N hydrochloric acid was informative in that this study indicated some elements may be more labile, and hence probably more bioavailable than others. By subjecting a sediment to a weak acid extraction, labile metals will be released from the surface of sediment particles indicating which portion of the total concentration may be available for uptake by organisms. Some elements (cadmium, lead, nickel, and silver) exhibited high levels of extractability, even at some reference sites (particularly Bodega Bay). The results are preliminary, but high levels of extractability of these elements at reference sites may explain the observed high stomach contents and liver concentrations of these same elements in fish from some of these sites.

In general, total concentrations of elements in sediment did not correlate with those found in liver tissue of fish. One metal, mercury, did show some association that was strongest in flatfish (English sole, flathead sole, hornyhead turbot, and starry flounder). Mercury in fish generally occurs in the organic (methyl) form which will accumulate in tissues and may be abundant in liver, whereas other elements may be regulated by the fish and not accumulated. Correlations of concentrations of elements between sediment and stomach contents were stronger than the correlations between liver and sediment. This was especially evident for chromium, copper, lead, mercury, and nickel. The predominant prey for flatfish are sediment-dwelling invertebrates, which because of their close association with the benthos, probably better reflect sediment concentrations than fish prey, especially if the predators were feeding recently in the area where sediment was collected.

The results of 5 years of monitoring the Pacific coast have demonstrated that concentrations of aromatic and chlorinated hydrocarbons in sediments were generally highly correlated with levels of these compounds or their derivatives in fish. This is generally not true for metals and metalloids for several reasons: 1) Most elements (essential and nonessential) are regulated by fish using various homeostatic control mechanisms; hence, even though in some cases these elements are taken up by the fish in proportion to their broad concentration range in the sediment, they may not accumulate in the tissues. 2) Many factors control the bioavailability of metals to organisms making determination of the form available for uptake difficult and correlations between tissue and sediment equivocal. Because sediments vary widely in these factors (e.g., grain size, organic (humic and fulvic acids) and inorganic (sulfides) ligands, redox state, cation exchange capacity, pH) and our understanding of their function is incomplete, it is not possible at this time to determine the portion of the total element concentration in sediment that is available for uptake. 3) Elements in sediment are commonly distributed in a nonhomogeneous fashion over small geographic areas, whereas fish are mobile over large areas; hence, accumulation of elements by fish within a site may be so variable that correlation analysis would be futile. 4) We examined only the liver, which may be the best single tissue to assess metal accumulation; however, other tissues such as kidney or brain may be more important for accumulation of certain elements. Future research will attempt to identify the various organismal compartments where metals and metalloids may accumulate.

We believe that some of the high concentrations of elements observed in the fish livers from reference areas may be due to the global spread of pollutants, especially at the West Coast sites. Because these reference sites are often sandy and have low total organic carbon (TOC), low manganese and iron oxides, and other properties associated with metal accumulation in sediments, the contaminant metals found there are more concentrated on a unit weight basis when normalized to one of these properties (e.g., TOC or percent fine sediment). Because these characteristics may control bioavailability (or are correlated to the actual controlling parameter), these stations having sandy sediment may have a higher proportion of the element that is bioavailable, even when compared to stations with higher concentrations of that element and a greater percentage of fine particles. The extraction study with hydrochloric acid to assess labile (bioavailable) concentrations supports this observation for many elements. Also, the stomach contents of fish from these reference sites contained elevated concentrations of elements possibly due to the increased bioavailability to sediment-dwelling invertebrate prey. Additionally, we have noticed that for many fish species, the elements arsenic, cadmium, lead, mercury, nickel, and silver were higher in fish livers from reference sites along the West Coast (Washington, Oregon, and California), which was generally not true for flathead sole from the Alaskan reference site when compared to Alaskan urban sites. All of these incidental observations taken together point to possible increased contamination at our reference and nonurban areas.

When all fish species are considered together, several elements exhibit elevated concentrations in liver tissue. Elements such as copper, mercury, selenium, and silver display significant differences between various fish species and are often associated with high sediment concentrations. Most concentrations show a progression, decreasing from sediment to stomach contents to liver. The elements cadmium, mercury, and selenium show the opposite pattern in that the liver concentration is much higher than that found in sediment. Because selenium is an essential element and basically constant over all species, its high concentration in tissue in relation to sediment concentration is generally not of concern, although the levels that distinguish normal from toxic are not well defined. Some elements (copper, silver, and zinc) show both patterns which may indicate inherent species differences or response to high sediment concentrations. Because cadmium and mercury are nonessential elements and they show a gradation of concentration, increasing from sediment to liver (with stomach contents intermediate in concentration), they appear to be bioaccumulated. Other nonessential elements which are lower in sediment than tissue, such as arsenic and silver, and which display species differences, may also be bioaccumulated.

Overall this survey has highlighted some important patterns of metal and metalloid abundance in sediment and fish along the Pacific Coast of the United States. Although this data set is relatively small for a monitoring program and covers only a few years, the results indicate some important trends which should be followed in subsequent studies. Many of the results, such as the association of mercury in flatfish, elevated sediment concentrations at urban sites, and the acid-extracted studies, generate new directions for research and testable hypotheses for clarification of mechanisms. Because of the inherent variability seen in sediment and tissue concentrations and the assumed complex nature of element speciation and control of bioavailability, future work highlighting the nature of chemical forms and their ability to be assimilated by organisms will go a long ways towards understanding the dynamic interactions between elements in water, sediment, and marine organisms.

• Measurement of concentrations of chemical contaminants in sediment and in species of bottom-dwelling fish at selected sites in urban and nonurban embayments.

• Determination of the prevalence of diseases in these same fish species.

• Exploration of associations between contaminant concentrations in tissue and sediment and between tissue, sediment, and fish disease.

• Evaluation of spatial and temporal trends of contaminant concentrations and fish diseases.

This technical memorandum summarizes the results of the toxic element contaminant analyses for the first 5 years of the Pacific coast portion of NBSP. It reports on the concentrations measured in sediment (1984-88) and tissues of fish (1984-87) from sites along the Pacific coast of the United States. The sites are from various bays and estuaries in Alaska, Washington, Oregon, and California. Our intent was to report, compare, and evaluate concentrations of elements in sediment and tissue from these sites in order to determine the current status, and if any spatial and temporal trends existed. We performed statistical analyses with sediment concentrations from nonurban sites to determine natural background levels and evaluated each urban site's concentration with respect to these baseline levels. In our study we also performed correlations in order to search for spatial patterns of association between elements in a single matrix (e.g., in sediment between sites) and elements in two matrices (e.g., one element in sediment vs. tissue). We could not pursue temporal trend analysis because of the high variability found at sites and the limited number of time points for each site. Future reports may attempt to take subsequent data from this program, in addition to historical data, and attempt temporal trend analysis.

While this technical memorandum is an overview of findings as well as a detailed presentation and treatment of these data, it is not meant to comprehensively review the related marine pollution literature on this subject. However, pertinent references on the most significant aspects are included in discussions.