Floating-bar plot presentations of the silver (Ag), tin (Sn), and zinc (Zn) in sediments are given in Figure 6.
Silver--The total silver concentration in Pacific coast sediments was significantly higher than the maximum reference site (Nisqually Reach UCI: 0.4 µg/g) at west Santa Monica Bay (5.6 µg/g), Moss Landing (1.8 µg/g), and three San Diego Bay sites: National City (1.4 µg/g), south bay (1.2 µg/g), and north bay (0.9 µg/g) (Figure 6). The west Santa Monica Bay silver concentration was significantly higher than all other Pacific coast sites. Moss Landing and National City values were also significantly higher than the lower 36 sites (Figure 6). The high silver concentrations reported for Moss Landing are perplexing but consistent over the three stations sampled (1.6, 1.7, and 2.0 µg/g)). Unfortunately, we sampled this site at only one time period (1986).
In the section on baseline metals and excess concentrations in urban sediments, we discuss the possibility that silver at Nisqually Reach was elevated beyond normal background concentrations; hence, this site was excluded in the determination of background silver concentrations. The reference site with the next lowest concentration (Dana Point) was used instead and only one additional site (San Pedro Outer Harbor) was significantly elevated above reference site upper comparison interval (0.33 µg/g).
Industries that have had potential for being sources of silver include jewelry, battery, porcelain, ink, antiseptic, and silverware manufacturing; photographic, ink, electroplating, and food processing, and cloud seeding and mining. About 40% of the U.S. supply of silver is for photographic film and chemicals (McComish and Ong 1988). Silver forms strong complexes with sulfides and will also complex with organic matter (McComish and Ong 1988).
One study has found variable concentrations of silver in surface sediment of Commencement Bay (0.07 to 0.34 µg/g) (Crecelius et al. 1985) which is similar to what we found at this site (0.28 (0.12) µg/g). Preindustrial silver concentrations in sediment for this area were found to be less than 0.05 µg/g (Crecelius et al., 1985).
According to Phillips (1987), silver in sediments from northern San Francisco Bay and the Sacramento/San Joaquin River Delta exhibit concentrations similar to background levels (<0.2 µg/g), indicating little anthropogenic contamination from upriver sources. Our San Pablo Bay and Castro Creek sites had mean (sd) sediment concentrations of 0.24 (0.20) and 0.16 (0.07) µg/g, respectively. However, within the central and southern reaches of the San Francisco Bay, silver is more abundant and concentrations in sediments are indicative of moderate enrichment in many areas with substantial enrichment in a few locations, such as Islais Creek (Phillips 1987). Our data fit this pattern with the highest silver levels from Islais Creek (0.59 (0.40) µg/g), followed by Oakland estuary (0.43 (0.14) µg/g), Hunters Point (0.42 (0.29) µg/g), and Redwood City (0.34 (0.07) µg/g).
Because municipal waste discharges are common sources of silver (Phillips 1987), the high sediment concentrations of silver at west Santa Monica Bay site may have been due in part to its location approximately 2.5 km north of the Seven-Mile Hyperion Sewer outfall. Even, one year after the Hyperion sludge discharge to the nearshore environment was terminated (November 1987), the silver levels in the immediate vicinity remained unchanged in the range of roughly 20 to 80 µg/g (Konrad 1989). At sites off the Palos Verdes Peninsula sewage discharge site, located between our east Santa Monica Bay (0.36 (0.07) µg/g) and San Pedro Outer Harbor (0.64 (0.19) µg/g) sites, Hershelman et al. (1981) reported surficial levels of silver ranging from 1.4 to 27 µg/g. Word and Mearns (1978) reported baseline levels for silver in coastal southern California marine sediments ranging from 0.06 to 1.7 µg/g (average: 0.2 µg/g), which are also similar to the values in our report.
Our mean (sd) for National City (1.4 (0.7) µg/g) was similar to those (2.0 and 4.0 µg/g) reported by Ladd et al. (1984), as well as those for south San Diego Bay (NBSP mean (sd): 1.2 (0.7) µg/g vs. 0.5-2.5 µg/g) and north San Diego Bay (NBSP mean (sd): 0.95 (0.46) vs. 1.0 µg/g).
Caution is needed in interpreting absolute levels of silver in sediments because no certified reference sediment materials are currently available for establishing accuracy of analysis. We rely on NOAA interlaboratory comparison studies to gauge our accuracy, which have shown our Pacific coast analyses to be comparable to silver analyses by other NBSP and Mussel Watch investigators.
Analytical results for single-sediment samples (not shown on Figure 6) collected from the San Luis Obispo (0.6 µg/g), Estero Bay (0.5 µg/g), outside Mission Bay (0.4 µg/g), Dana Point Inside (0.11 µg/g), and the Farallon Islands (<0.01 µg/g) were grouped around the reference site.
Tin--The concentration of total tin in Pacific coast sediments was significantly higher than the reference site (Dana Point UCI: 3.2 µg/g) at only two sites: Oakland estuary (10 (6.1) µg/g) and south San Diego Bay (6.3 (3.2) µg/g) (Figure 6). It is apparent that tin was elevated in some urban sites, possibly due to organotin release from antifouling paint on boats. It is possible that tin in Dana Point sediments was elevated beyond background levels and choosing the next reference site, Bodega Bay, would have caused many more sites (eight) to be significantly higher than the UCI of that reference site. This was not done because there was no clear pattern of elevated concentrations at Dana Point.
Analytical results for single-sediment samples (not shown on Figure 6) collected from Dana Point Inside (3.1 µg/g), Farallon Islands (2.2 µg/g), outside Mission Bay (2.2 µg/g), San Luis Obispo (0.8 µg/g), and Estero Bay (<0.7 µg/g) were all below the maximum reference site value.
The activities of humans have radically increased the mobilization of tin from the earth's crust by a factor of more than 10 (Phillips 1980); however, the impact of tin on aquatic ecosystems has received little attention because of the low toxicity exhibited by the inorganic form (Phillips 1987). Due to difficulties in analyzing tin in marine samples and to the low toxicity of inorganic tin to aquatic systems, few large-scale monitoring efforts on the tin content in sediments are available for the Pacific coast. Concern about tin in the aquatic environment was altered dramatically during the 1970s with the widespread use of organotins as an antifouling agent for vessels. Organotins have been used for stabilizing polyvinylchloride polymers and as bactericides, fungicides, and insecticides (Champ 1986). As an antifouling agent in hull bottom paints, the organotins proved highly effective, but unfortunately they also exerted toxic effects in the parts-per-trillion range on nontarget aquatic organisms. Before the U.S. Congress imposed restrictions on their use in 1988, organotins could be found at toxic levels in water bodies with limited circulation (e.g., marinas, channels, harbors, urban bays).
The present study has generated one of the largest collections of total tin data for surficial sediments for this region. As the interest in organometals, such as tributyltin in antifoulants, increases, more information will undoubtedly become available.
Zinc--The total zinc concentrations in Pacific coast sediments were significantly higher than the maximum reference site (Nisqually Reach UCI: 130 µg/g) at sites representing the four major urban areas of Oakland estuary (310 µg/g), south (282 µg/g) and north San Diego Bay (230 µg/g), Long Beach (189 µg/g) and San Pedro Outer Harbor (178 µg/g), and Elliott Bay (187 µg/g) off Seattle (Figure 6).
In the section on baseline metals and excess concentrations in urban sediments, we discuss the possibility that zinc at Nisqually Reach was elevated beyond normal background concentrations, thus this site was excluded in the determination of baseline zinc concentrations. The reference site with the next lowest concentration (Dana Point) was used instead; hence, 23 additional sites have been determined to be significantly higher than this reference site's upper comparison interval (64 µg/g). Because the background concentration of zinc is strongly affected by grain size, determination of contaminant levels is difficult. We address grain size effects in the section on chemical correlations.
Zinc can be considered an essential element because it is a requirement of several metalloproteins, particularly metalloenzymes (Thompson 1990). Phillips (1987) estimates a mobilization rate of zinc due to human activities of 10-15 times that of the natural rate, much like copper and silver, although the impact is considered to be significantly less.
High levels of zinc in Elliott Bay sediments are reflected in dissolved zinc distribution in this area. Paulson and Feely (1985) found low water concentrations of zinc in the Duwamish River upstream of most point sources and increasing surface water concentrations in Elliott Bay as they approached the mouth of the Duwamish Waterway downstream of the point sources. These findings strongly suggests that anthropogenic inputs injected by industrial and other sources along the Duwamish River and Duwamish Waterway on Harbor Island were responsible for an elevation in dissolved levels and ultimately in high sediment concentrations in underlying sediments. The mean value for our Elliott Bay NBSP site sediments of 187 (45) µg/g was in the mid-range (86-319 µg/g) for values reported by Malins et al. (1980) from this area.
The mean (sd) NBSP site concentration (92 (8) µg/g) from Commencement Bay was higher than that reported by Schults et al. (1987) for a nearby site off Blair Waterway (50 µg/g), but less than that found elsewhere in local waterways (to 1190 µg/g). Crecelius et al. (1985) reported zinc values of 87-94 µg/g from nearby sites and a preindustrial concentration of about 70 µg/g. While zinc inputs can occur from the pulp and paper industry (Dexter et al. 1985), Grieve and Fletcher (1977) showed that major river systems can contribute significant zinc to estuarine and coastal sediments.
Our mean (sd) zinc value (90 (15) µg/g) was higher that than reported by Fuhrer and Rinella (1983) for the Columbia River estuary (22 µg/g) as was our Youngs Bay site (NBSP mean (sd): 132 (80) vs. 35 µg/g). Our Coos Bay value (64 (37) µg/g) was similar to that reported by Fuhrer and Rinella (1983).
Phillips (1987) concluded that the average San Francisco Bay zinc sediment concentration was about 100 µg/g with relatively little variation from basin to basin, except in the immediate vicinity of sources which appeared largely to be rivers and sewage treatment plants. Risebrough et al. (1978) found levels ranging from 150 to 174 µg/g near Treasure Island and Hunters Point, in agreement with our zinc values for nearby sites (Oakland NBSP mean: 172 (9) µg/g; and Hunters Point 136 (25) µg/g); however, Luoma and Phillips (1988) more recently reported 405 µg/g in Oakland Outer Harbor near Treasure Island and 202 µg/g at Hunters Point. It is not known whether these differences reflect local variability in sampling or temporal increases. Risebrough et al. (1978) also reported concentrations of 100-124 µg/g near our Southampton Shoal site (104 (21) µg/g) and 100 µg/g in southwest San Pablo Bay (116 (24) µg/g).
The east Santa Monica Bay site contained low zinc sediment concentrations (35 (9) µg/g) whereas the west Santa Monica site near the Hyperion sewage sludge outfall had considerably more zinc (102 (56) µg/g). Brown et al. (1986) reported levels of 54 µg/g for a nearshore site and 70 µg/g for a site near the Hyperion outfall. In their 71-station, 60 m depth contour survey between Point Conception and the U.S.-Mexico border, Word and Mearns (1978) found a background range of 37-47 (mean: 44) µg/g zinc in surficial sediments. Our San Pedro Canyon (118 (11) µg/g) and Seal Beach (125 (30) µg/g) sites showed elevated concentrations above the background levels of Word and Mearns (1978). Higher yet, were the sites at nearby Long Beach (189 (23) µg/g) and San Pedro Outer Harbor (179 (35) µg/g) sites. Hershelman et al. (1981) reported that their southeasterly most transect off Palos Verdes had zinc sediment levels of 118-186 µg/g; this transect was between the Palos Verdes sewer outfall diffusers and our San Pedro Canyon site.
Our mean value for south San Diego Bay (281 (75) µg/g) was in the range (43-765 µg/g) reported by Ladd et al. (1984), although we found higher levels (NBSP mean (sd): 230 (40) µg/g vs. 89.5-120 µg/g) at our north San Diego Bay site and lower levels (NBSP mean (sd): 206 (109) vs. 490-4300 µg/g) at the National City site.
The values for zinc at the Oliktok Point and Endicott Field sites (81 (24) and 72 (12) µg/g, respectively) were in the range reported by Sweeney and Naidu (1989) for nearby sites (28-160 and 40-105 µg/g, respectively). Likewise, our mean (sd) concentration found at Port Valdez (150 (8) µg/g) was similar to that (128 µg/g) from inshore sediments off nearby Island Flats (Feder et al. 1990). Feely et al. (1981) found 133 µg/g zinc in Copper River particulate matter and 210-292 µg/g in surface nearshore suspended matter in the northeastern Gulf of Alaska which suggests that our central Alaskan values could represent geologically-derived zinc. In Hecate Strait, south of Boca de Quadra (104 (8) µg/g), Harding and Goyette (1989) reported zinc concentrations of 33.1 (7.6) µg/g.
Analytical results for single-sediment samples (not shown on Figure 6) collected from the Dana Point Inside (118 µg/g), Farallon Islands (48 µg/g), outside Mission Bay (47 µg/g), San Luis Obispo (44 µg/g), and Estero Bay (38 µg/g) were all less than the reference site.
Floating-bar plot presentations of the major sediment (crustal) components aluminum (Al) and iron (Fe) in sediments are given in Figure 7. The maximum comparison interval value for the reference site is not included on the figures because these elements are considered as strictly geologically derived.
Aluminum--The mean aluminum concentrations group within the ranges for estuarine sediments (3.5-8.5%) and all comparison intervals overlap within the maximum (7.9%) and minimum (4.3%) reference site values, as do the exploratory site single determinations.
Sediment concentrations at our Puget Sound sites varied from 6.5 to 7.5%, which is in line with values cited by Crecelius et al. (1985) for Commencement Bay (5.1-6.6%) sediments. In California, Chapman et al. (1986) reported southwestern San Pablo Bay sediment concentrations of 7.5-7.9% (NBSP mean (sd): 6.4 (1.6) %), 7.2-9.9% for Oakland Outer Harbor (NBSP: 6.8 (0.15) %), and 7.5-8.0% in Islais Creek Channel (NBSP mean: 5.2 (0.3) %).
The mean (sd) aluminum concentrations ranged from high values along the margins of the Gulf of Alaska (8.5 (0.9) % at Port Moller, Skagway (8.3 (0.7) %, and Port Valdez (8.1 (0.3) %) to low values in the Chukchi (3.6 (0.4) %) and Beaufort Seas (Endicott: 3.5 (0.3) %). The aluminum concentrations at the three lowest Alaskan sites (Endicott Field, Chukchi Sea, and Oliktok Point) were significantly lower than at the three highest Alaskan sites. Robertson and Abel (1990) found 3.5% aluminum off Prudhoe Bay (NBSP Endicott Field mean 3.5%), 4.2 % at Kotzebue (NBSP Chukchi Sea mean 3.6%), 6.1% off Oliktok Point (NBSP mean 4.4 (1.1) %), and 7.7 % in Kamishak Bay sediments near our NBSP site (7.5 (0.4) %). Feder et al. (1990) reported even higher (16.9%) surficial sediment levels at their experimental mudflat site east of Port Valdez (NBSP mean: 8.2 (0.3) %). In Hecate Strait, south of Boca de Quadra (NBSP mean: 7.0 (1.0) %), Harding and Goyette (1989) reported a mean (sd) aluminum concentration of 7.2 (2.3%).
Iron--The mean (sd) iron concentrations fell between 6.4 (0.9) % at Port Moller and 0.8 (0.56) % at Monterey Bay (Figure 7) and is probably related to grain size (see section on correlations). The comparison intervals for all sites overlap with the maximum (5.4%) and minimum (1.5%) intervals found for all reference site values (Nisqually Reach, Bodega Bay, and Dana Point) as do the exploratory single-site determinations (1.8-3.3%). No obvious geographic or anthropogenic patterns are apparent from the data (Figure 7).
Crecelius et al. (1985) reported iron levels in Commencement Bay sediments of 3.8-4.2% compared with our NBSP value of 3.7 (0.1) %. Likewise, in southwestern San Pablo Bay, Chapman et al. (1986) found 4.6-5.1% vs. our NBSP site 4.4 (0.5) %. They also report at Oakland Outer Harbor values were 4.3-6.2% compared with 5.1 (0.2) %, and in Islais Creek Channel the values were 4.8-5.0% compared with NBSP mean of 4.9 (0.3) %. The NBSP sites in Santa Monica Bay were near the lower end of the range of values for the Pacific coast sites (2.1 and 1.5%) but were not as low as Jan and Hershelman (1980) reported for their Hyperion outfall site (1.0%).
In the Beaufort Sea, Sweeney and Naidu (1989) report Oliktok Point and Prudhoe Bay area concentrations range from 0.8-4.8% and 0.8-2.9%, respectively, and Robertson and Abel (1990) report 3.7% and 3.0%, respectively. This is compared with the corresponding NBSP site values of 2.5 (0.7) % and 2.0 (0.2) %. Robertson and Abel (1990) reported a surficial iron sediment concentration of 3.5 % near the Kamishak Bay NBSP site (3.5 (0.08) %) and 2.5% off Kotzebue near our NBSP Chukchi Sea site (NBSP mean: 2.5 (0.4) %). At Port Valdez, the NBSP mean value of 5.1 (0.3) % was less than what Feder et al. (1990) reported (7.4%).
Floating-bar plot presentations of the major sediment (crustal) components manganese (Mn) and iron (Fe) in sediments are given in Figure 8. The maximum comparison interval value for the reference site is not included on the figure for silicon because this element is considered as strictly geologically derived.
Manganese--Manganese is a crustal element and, although measured in hundreds of parts per million, is important in understanding the biogeochemistry of estuarine and coastal sediments (Figure 8). The mean surface sediment concentrations for the Pacific coast varied by about an order of magnitude (110 to 1250 µg/g) and is probably related to grain size. The reference site values (Nisqually Reach: comparison interval of 1117 µg/g) for manganese overlapped with all the urban sites and the single determination sites fell within the midrange of the remaining Pacific coast samples. The single analysis sites are outside Mission Bay (666 µg/g), Dana Point Inside (326 µg/g), Estero Bay (310 µg/g), Farallon Islands (263 µg/g), and San Luis Obispo (192 µg/g).
Crecelius et al. (1985) reported manganese levels in Commencement Bay surface sediments in the range of 470-570 µg/g (NBSP mean (sd): 540 (86) µg/g) with little change in concentration over a vertical profile which covered the last 100 years. Schults et al. (1987) reported a lower concentration (111 µg/g) at their nearby Blair Waterway entrance station as well as all of their other Commencement Bay and Tacoma waterways sites (maximum: 368 µg/g) which is probably due to variable grain size.
Our manganese values in the San Francisco Bay area closely approximated those reported by Chapman et al. (1986) for southwestern San Pablo Bay at 640-790 µg/g (NBSP mean (sd): 730 (129) µg/g), for Oakland Outer Harbor at 510-760 µg/g (NBSP: 530 (55) µg/g), and for Islais Creek Channel at 380-450 µg/g (NBSP: 410 (65) µg/g). Jan and Hershelman (1980) reported a mean manganese concentration of 89 µg/g off the Hyperion outfall in Santa Monica Bay compared with 410 µg/g at our nearby, but more northern site (west Santa Monica Bay).
Sweeney and Naidu (1989) reported ranges of manganese of 115-555 µg/g and Robertson and Abel (1990) found 491 µg/g for the Oliktok Point area (NBSP: 390 µg/g); likewise, the former reported 205-420 µg/g and the latter 369 µg/g for the Prudhoe Bay area (NBSP: 330 (109) µg/g). Robertson and Abel (1990) reported 698 µg/g in surface sediments in Kamishak Bay near our NBSP site (766 (19) µg/g) and 405 µg/g off Kotzebue compared with our Chukchi Sea mean of 460 (97) µg/g. Our Port Valdez site's mean (sd) concentration of 1030 (73) µg/g corresponded with Feder et al. (1990) value of 1140 µg/g in a mudflat to the north of the Lowe River mouth.
Silicon--The mean (sd) silicon concentrations fell between 22.9 (0.7) % at Dutch Harbor and 42.9 (13.5) % at Humboldt Bay (Figure 8) and is probably related to grain size (see section on correlations). The mean West Coast silicon values group within the ranges for estuarine sediments (23-33%), with the values for Humboldt Bay (42.9%), Monterey Bay (38.9%), and Chukchi Sea (35.8%) being higher. The comparison intervals overlap within the maximum (36%) and minimum (25%) reference site values, as do the exploratory site's single determinations (26-33%).
Chapman et al. (1986) reported silicon levels in southwestern San Pablo Bay of 26.9-31.2% compared with our value of 29.9%, Oakland Outer Harbor of 27.4-33.6% compared with 28.6%, and Islais Creek Channel of 23.3-24.4% compared with our 25.8%. Feder et al. (1990) reported a value of near 60% for his Port Valdez site compared with our value of only 26.9%.
General Discussion of Sediment Chemistry
Of the 16 elements routinely analyzed in Pacific coast sediments, copper, lead, mercury, nickel, tin, and zinc appear to be the most likely pollution-associated elements because they were found in the highest concentrations in one or more urban sites. When compared to one of the three West Coast reference sites (Nisqually Reach, Bodega Bay, or Dana Point), many urban sites had significantly higher concentrations of these elements. In the section on baseline metals and excess concentrations in urban sediments we will explore this subject in greater depth.
Although the concentrations of a variety of elements were significantly higher in the sediments from certain urban sites (e.g., south San Diego Bay, west Santa Monica Bay, Oakland estuary, and Elliott Bay) compared to the reference sites, the high degree of intrasite variability in chemical concentrations at many of the sites frequently made it difficult to identify statistically significant intersite differences with the comparison intervals. For example, the mean cadmium concentration in sediment from Oceanside (Figure 3) was up to 50 times less than at 32 other Pacific coast sites, yet these differences were not statistically significant.
A number of factors may have influenced this intrasite variability, including 1) heterogeneous spatial distribution of chemicals in sediment within a small geographical area due to variable sediment characteristics, and 2) temporal changes in the accumulation of contaminants in sediment, such as episodic releases. Clearly, Factor 1 is well documented and may have had a strong influence on the observed variability, and Factor 2 can be evaluated using trend analyses, even though it is difficult to distinguish spatial variability from temporal variability. Additional years of sampling and sample analyses should provide sufficiently large sample sizes to reduce the intrasite variability, permit more realistic trend analyses, and thus allow more accurate estimates of intersite differences.
In order to assess the general health of the coastal and estuarine environments, fish were also analyzed for selected elements to determine if elements deposited in sediments from human activities would be transferred to biota. Livers and, in a limited number of cases, stomach contents of bottom-dwelling or bottom-feeding fish were analyzed for the first four cycles of Pacific coast sampling. As mentioned earlier, we made few comparisons of liver concentrations to published values because the data are sparse for the Pacific coast, and such information is less useful without information on exposure concentrations for comparison.
Ideally, in order to avoid species-specific differences, individuals of the same fish species should be compared when evaluating the exposure of fish to contaminants at different sites. However, because the same fish species do not inhabit all of the Pacific coast sampling sites, a total of eight target species were sampled for comparative purposes. Each target species (except black croaker) was sampled at a minimum of one urban (human-impacted) and one nonurban site. The target species and the geographical areas from which they were captured are given in Table 2a.
Interspecies differences in the uptake and fate of toxic chemicals by fish have been demonstrated by Varanasi et al. (1986, 1987, 1989b). For example, although both English sole and starry flounder (Platichthys stellatus) metabolized benzo[a]pyrene at similar rates, the two species showed differences in the detoxification of certain metabolites. Due to possible species-specific differences in uptake and metabolism of toxic elements, comparisons of tissue chemistry between species has not been strongly emphasized, however some interspecies comparisons are presented because we felt they highlighted potentially important patterns. In general, we will make intraspecies comparisons to indicate the degree to which fish were being exposed to contaminants at different sites.
The data on liver concentrations for each species and site (mean of three largest female specimens) are displayed graphically in Figures 9 through 43. In the section on correlations in liver, stomach contents, and sediment, we present pooled data from all individuals of a species over all geographic locations collected and compare liver concentrations to other compartments (stomach contents and sediment) in order to search for patterns of contamination. In all but a few cases, the concentrations of toxic elements in the livers of fish from urban sites were similar to, and occasionally lower than, those in livers from nonurban sites. There were few significant differences between populations at these different sites, but trends are evident and we believe that as comparison-sample sizes increase, significant differences will become more apparent. In the section on the relationship between chemical parameters (tissue and sediment) we propose a hypothesis which may explain the counterintuitive nature of high concentrations in reference site fish.
Fourhorn Sculpin, Spotted Sand Bass, Spotted Turbot, and Black Croaker
Floating-bar plot presentations for elements in liver tissue from fourhorn sculpin (Myoxocephalus quadricornis) from the Beaufort Sea and spotted sand bass (Paralabrax maculatofasciatus), spotted turbot (Pleuronichthys ritteri), and black croaker from southern California are combined in Figures 9-13 because of the limited amount of data available for each species. The plots of antimony (Sb), arsenic (As), and cadmium (Cd) are shown in Figure 9; chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 10; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 11; nickel (Ni), selenium (Se), and silver (Ag) in Figure 12; and tin (Sn) and zinc (Zn) in Figure 13. In comparing the Alaskan sites, the overlap of comparison intervals for the Endicott Field site (human impacted) with the Oliktok Point site (nonurban) was complete for all 14 elements in the fourhorn sculpin, except for zinc and arsenic which were significantly higher at the former site. Arsenic in another species from Alaska (flathead sole (Hippoglossoides elassodon)) was also high and is presented next. Black croaker, spotted sand bass, and spotted turbot were all taken in the San Diego area and are shown not for comparison but for informational purposes only. It is noteworthy that concentrations of mercury and cadmium were very high in spotted sand bass, and black croaker contained elevated concentrations of several elements, including tin.
Floating-bar plot presentations of antimony (Sb), arsenic (As), and cadmium (Cd) in liver tissue from flathead sole from Alaskan sites are given in Figure 14. The plots of chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 15; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 16; nickel (Ni), selenium (Se), and silver (Ag) in Figure 17; and tin (Sn) and zinc (Zn) in Figure 18. Lutak Inlet was used as the reference site for this Alaskan species. The concentration of arsenic in the flathead sole livers from Boca de Quadra (85 (26) µg/g) was significantly higher than the reference site (UCI: 34 µg/g), as well as for fish from Dutch Harbor and Skagway. Lead in liver of fish from Skagway (4.2 (5.0) µg/g) was significantly higher than from the reference site (Lutak Inlet upper comparison interval (UCI): 0.44 µg/g) and all other Alaskan sites except Dutch Harbor. The concentration of mercury in fish caught at the Dutch Harbor (0.42 (0.06) µg/g) and Kamishak Bay (0.27 (0.18) µg/g) sites was significantly higher than the reference site (UCI: 0.10 µg/g). Likewise, chromium (1.0 µg/g) and manganese (10.4 µg/g) concentrations in flathead sole from Dutch Harbor were significantly above the reference site UCI (0.32 µg/g and 4.7 µg/g, respectively). Manganese displayed a strong pattern with high levels in fish from Dutch Harbor to over an order of magnitude lower concentration in fish from Skagway. Other elements (arsenic, iron, and selenium) were also low in fish from Skagway, which may be related to the high levels of lead observed.
Floating-bar plot presentations of antimony (Sb), arsenic (As), and cadmium (Cd) in liver tissue from English sole are given in Figure 19. The plots of chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 20; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 21; nickel (Ni), selenium (Se), and silver (Ag) in Figure 22; and tin (Sn) and zinc (Zn) in Figure 23. Only tin in livers from English sole caught from Elliott Bay (1.8 (2.7) µg/g) was significantly higher than the UCI for the reference site Nisqually Reach (0.72 µg/g). For English sole, there were two reference sites to choose from, Nisqually Reach or Bodega Bay. When all sites were considered, liver tissue from the reference site Nisqually Reach, often contained the first or second highest mean concentration of the selected element (copper, silver, nickel, chromium, arsenic, cadmium, iron, manganese, and antimony), but the concentration was generally not significantly different from other sites. For example, the nickel concentration in liver from Nisqually Reach and Commencement Bay was significantly higher than concentrations in livers from Elliott Bay and Bodega Bay. Livers from English sole in Commencement Bay contained more iron and manganese than livers from English sole in Monterey Bay and more iron than in English sole livers from Bodega Bay. As mentioned earlier, Nisqually Reach may have contained elevated concentrations of some elements in the sediment. In comparing the different sites for English sole, only copper, nickel, zinc, and iron would be affected if the alternate reference site (Bodega Bay) was chosen for the benchmark.
Floating-bar plot presentations of antimony (Sb), arsenic (As), and cadmium (Cd) in liver tissue from starry flounder are given in Figure 24. The plots of chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 25; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 26; nickel (Ni), selenium (Se), and silver (Ag) in Figure 27; and tin (Sn) and zinc (Zn) in Figure 28. Only three means were significantly higher than their reference site UCI; chromium in liver of fish from Coos Bay (0.73 (0.66) µg/g) and the Columbia River estuary (0.63 (0.44) µg/g) and tin in fish from Coos Bay (2.8 (4.2) µg/g) . The UCI for chromium and tin at Bodega Bay was 0.35 and 0.63 µg/g, respectively. Silver in fish liver from the Bodega Bay reference site was significantly higher than silver in fish from the San Francisco Bay sites of San Pablo Bay, Southampton Shoal, and Castro Creek and from the Columbia River estuary and nearby Youngs Bay sites. Arsenic in liver of starry flounder from San Pablo Bay was significantly lower than from fish caught in the Columbia River estuary and zinc in liver of fish from Bodega Bay was significantly higher than in fish from San Pablo Bay.
Floating-bar plot presentations of antimony (Sb), arsenic (As), and cadmium (Cd) in liver tissue from white croaker are given in Figure 29. The plots of chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 30; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 31; nickel (Ni), selenium (Se), and silver (Ag) in Figure 32; and tin (Sn) and zinc (Zn) in Figure 33. Several elements (lead, mercury, selenium, and cadmium) show strong trends; however, none were significantly different. For this species, there was a strong trend in mercury levels in liver; however, none were significant. The levels of cadmium, iron, silver, selenium, and zinc in one or both of the reference sites were the first or second highest mean value determined for white croaker livers. Interestingly, cadmium in liver of fish from San Pedro Outer Harbor and Long Beach was significantly lower than in fish from the reference site which was unexpected. Even though the range between minimum and maximum mean concentrations for some elements (lead, antimony, and tin) was 10 to 15 times, the large variability within sites precluded statistically significant differences.
Floating-bar plot presentations of antimony (Sb), arsenic (As), and cadmium (Cd) in liver tissue from hornyhead turbot (Pleuronichthys verticalis) are given in Figure 34. The plots of chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 35; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 36; nickel (Ni), selenium (Se), and silver (Ag) in Figure 37; and tin (Sn) and zinc (Zn) in Figure 38. No site produced a significantly higher concentration of any element in hornyhead turbot liver when compared to the reference site (Dana Point) concentration (horizontal line) although a few were close. However, when comparing all sites to each other, copper in fish caught in San Pedro Canyon was significantly higher than that for fish from west Santa Monica Bay and outer San Diego Bay. This observation might be explained by sewage effluents from the Palos Verdes sewer outfall (Brown et al. 1986).
The mean (sd) concentration of silver in west Santa Monica Bay (1.9 (1.0) µg/g) hornyhead turbot liver was significantly higher in individuals from outer San Diego Bay (0.5 (0.3) µg/g). This pattern was reversed for zinc with outer San Diego Bay having a higher concentration. East Santa Monica Bay (11 (7) µg/g) and west Santa Monica Bay (9 (2) µg/g) fish liver contained significantly higher selenium concentrations than individuals from San Pedro Canyon (1.5 (0.4) µg/g) and outer San Diego Bay (2.0 (0.9) µg/g). Also, fish liver from San Pedro Canyon (7.2 (3.4) µg/g) and outside San Diego Bay (6.1 (1.5) µg/g) contained significantly higher concentrations of manganese than fish liver from west Santa Monica Bay (1.8 (2.1) µg/g) and east Santa Monica Bay (2.8 (0.8) µg/g) .
Floating-bar plot presentations of antimony (Sb), arsenic (As), and cadmium (Cd) in liver tissue from barred sand bass (Paralabrax nebulifer) are given in Figure 39. The plots of chromium (Cr), copper (Cu), iron (Fe) are shown in Figure 40; lead (Pb), manganese (Mn), and mercury (Hg) in Figure 41.; nickel (Ni), selenium (Se), and silver (Ag) in Figure 42; and tin (Sn) and zinc (Zn) in Figure 43. Mean (sd) lead 1.3 (0.06) µg/g) and silver (0.12 (0.02) µg/g) in Mission Bay fish were significantly higher than their respective reference site concentration (Dana Point UCI: 0.53 and 0.07 µg/g, respectively). Lead was also significantly higher in Mission Bay fish than in fish from south San Diego Bay (0.5 (0.3) µg/g) and National City (0.4 (0.1) µg/g). Zinc in Dana Point barred sand bass livers (lower comparison interval (LCI): 82 µg/g) was significantly higher than in fish from south San Diego Bay (72 (17) µg/g) and National City (54 (23) µg/g). These values were unexpected because Mission Bay is not an industrialized area; however, this bay can receive substantial amounts of surface water runoff which may contain contaminants (Cole et al. 1984), including lead from automobiles.
In several species there was a trend for the concentration of elements in stomach contents to be higher than the liver concentration; however, none were significantly different. In hornyhead turbot and barred sand bass, several elements (silver, mercury, chromium, lead, zinc) displayed a trend of higher means at the more urban sites (e.g., south San Diego) vs. a reference site (e.g., Dana Point). The same is true for silver, copper, lead, zinc in white croaker. In the section on correlations in liver, stomach contents, and sediment, we pooled stomach contents chemistry for all individuals of a species for comparisons between species and for comparisons to sediment and liver within a species.