Potential Use of the Benthic Foraminifers Bulimina denudata and Eggerelloides advenus in Marine Sediment Toxicity Testing

Abstract

The benthic foraminifers Bulimina denudata and Eggerelloides advenus are commonly abundant in offshore regions in the Pacific Ocean, especially in waste-discharge sites. The relationship between their abundance and standard macrofaunal sediment toxicity tests (amphipod survival and sea urchin fertilization) as well as sediment chemistry analyte measurements were determined for sediments collected in 1997 in Santa Monica Bay, California, USA, an area impacted by historical sewage input from the Hyperion Outfall primarily since the late 1950s. Very few surface samples proved to be contaminated based on either toxicity or chemistry tests and the abundance of B. denudata did not correlate with any of these. The abundance of E. advenus also did not correlate with toxicity, but positively correlated with total solids and negatively correlated with arsenic, beryllium, chromium, lead, mercury, nickel, zinc, iron, and TOC. In contrast, several downcore samples proved to be contaminated as indicated by both toxicity and chemistry data. The abundance of B. denudata positively correlated with amphipod survival and negatively correlated with arsenic, cadmium, unionized ammonia, and TOC; E. advenus negatively correlated with sea urchin fertilization success as well as beryllium, cadmium, and total PCBs. As B. denudata and E. advenus are tolerant of polluted sediments and their relative abundances appear to track those of macrofaunal toxicity tests, their use as cost- and time-effective marine sediment toxicity tests may have validity and should be further investigated.

1. Introduction

Typically, sediment toxicity, sediment chemistry measurements, and benthic macrofaunal community structure have been used to assess the toxicity of sediments in polluted regions [1,2,3,4,5]. Sediment toxicity tests, in particular, measure the direct effect of the combined toxic effect of the mixture of contaminants that are present in the interstitial water or surface sediments on benthic macroorganisms [6,7]. Microfauna, specifically foraminifera, have also been used for more than a half-century investigating ocean pollution in bays, harbors, and coastal margins [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]. Each method provides valuable information on the impact of anthropogenic pollution on the local biology and ecosystem.

Many benthic macrofaunal workers suggest their methods are valid because they analyze complex, multi-celled organisms that are sedentary and have a short lifespan so they can respond quickly to environmental change [31,32,33,34,35]. Microfaunal proponents suggest marine protozoa are useful because they, too, are sedentary and have a short lifespan [8], are primary consumers so they are lower on the food chain [36,37], and are ubiquitous in marine and estuarine habitats [38,39,40]. Foraminifera also have similar faunal distributions in the sediment to those of shallow marine invertebrates [15,41] so they can act as proxies for larger organisms. Some would even claim that they are among the most important animal groups on earth because they are significant in biogeochemical cycles on a global scale [42,43,44]. Yet rarely, if ever, are macro- and microfauna used in conjunction with one another to assess contamination in the marine environment and, ultimately, if a potential human health issue exists [15]. This study attempts to bridge that gap by integrating standardized macrofaunal methods (amphiopod survival and sea urchin fertilization success) used to assess sediment quality in coastal areas that are integral components of local municipal wastewater treatment agencies (e.g., Southern California Bight Regional Monitoring Programs, referred to as Bights ’94, ’98, ’03, ’08, ’13, and ‘18 [45]) and national programs (EPA, NOAA, and USGS) in the United States, as well as internationally [46,47] see summary in [48], with foraminiferal species’ abundances. The focus of this study is on surface and downcore sediments in one southern California embayment known as Santa Monica Bay.

Southern California has a long history of environmental research and monitoring. Some of the earliest studies investigating the relationship between foraminifera and pollution were initiated here [9,10,11,12,13,24,25,29,30], and several locations have been reexamined to assess whether remediation efforts had a positive impact on the biota [49,50,51]. Similarly, several multi-agency studies (i.e., Bights 1994 to 2018) have attempted to assess the health of the harbors and offshore regions based on repeated sediment toxicity, sediment chemistry, macrofaunal, and fish studies.

Previous work has shown that six species dominate the foraminiferal assemblage of Santa Monica Bay: Bulimina denudata Cushman and Parker, Eggerelloides advenus (Cushman), Portatrochammina pacifica (Cushman), Buliminella elegantissima (d’Orbigny), Nonionella stella Cushman and Moyer, and Nonionella basispinata (Cushman and Moyer) [12,30,50]. A time-series from 1955 to 1998 illustrated dramatic changes in the abundance of some of these species that was attributed to municipal wastewater discharge and later changes in water treatment protocols [49,50]. A comparison of sediment chemistry and foraminiferal distribution data acquired in 1998 showed that of the dominant six species, B. denudata and E. advenus occurred in high abundance most frequently where contaminants were most concentrated [49]. The present study expands on these initial investigations [49,50] to better document the temporal association between two of the dominant benthic foraminiferal species (B. denudata and E. advenus) with commonly used macrofaunal sediment toxicity measures and sediment chemistry analytes in Santa Monica Bay, using a reinterpreted chronology from that utilized in 2003.

Bulimina denudata is a Pacific Ocean species [52] known as a waste-indicating foraminiferal species tolerant of polluted sediments except at extreme levels [9,13,53,54] and low oxygen conditions [55,56,57]. The species is also an indicator of high food availability in natural settings [56,58] and shows an opportunistic response to anthropogenic enrichment of organic matter, trace metals, and trace organic compounds [57]. Eggerelloides advenus also commonly inhabits waste-discharge regions [29,59,60,61,62], is a pioneer colonizer of remediated areas possibly because of its high mobility [63] and has a cosmopolitan distribution as well [52]. As B. denudata and E. advenus are widely distributed, tolerant of contaminants, and prefer living on the shelf and upper slope [49,52] where pollutants are often dumped, further investigation of the association between their abundance and other toxicity measures commonly used in environmental monitoring studies is warranted.

2. Setting

Santa Monica Bay is located on the southwestern coast of the United States near the city of Los Angeles (Figure 1A,B). The bay extends from Point Dume to Palos Verdes Point and is bordered by the Santa Monica Mountains and the Palos Verdes Hills (Figure 1C). The continental shelf of Santa Monica Bay generally ranges from 5–8 km wide, except near the center where the shelf extends ~18 km. The shelf is incised by three canyons: Dume Canyon in the north, Santa Monica Canyon in the center, and Redondo Canyon to the south. To the north, below the shelf break at ~100 m depth, is an abbreviated and steep slope referred to as the “Ramp”; to the south, the relief is less abrupt and is known as “Short Bank.” Both lead to the Santa Monica Basin to the west.

One of the largest sewage outfall systems in southern California is located in Santa Monica Bay (Figure 1C); its history has been summarized by Dojiri [64]. In 1894, the City of Los Angeles began discharging raw sewage from the Hyperion treatment plant directly into the surf zone of Santa Monica Bay. Complaints of locally fouled beaches resulted in the construction of a screening facility and a submarine outfall one mile in length in 1925. However, due to construction problems and increased effluent flow, the extent of fouled beaches expanded, so that by 1943, coastlines as much as 10 miles distant from the outfall were affected. A new treatment plant and 1-Mile outfall were constructed in 1951 (Figure 1C). Chlorinated secondary effluent was discharged through the outfall at a depth of 14 m, while the sludge was processed and made into fertilizer. Later, this fertilizer production was not considered cost-effective and the sludge was also discharged into the Bay.

In the late 1950s two new outfalls were constructed: the 7-Mile and 5-Mile outfalls (Figure 1C). From 1957–1987, the 7-Mile outfall discharged digested sludge into the Bay at a depth of 98 m directly into the head of Santa Monica Canyon. In 1960, the 5-Mile outfall became operational, discharging unchlorinated liquid effluent at a depth of 58 m from a Y-shaped diffuser. As population increased in the city of Los Angeles, a corresponding decrease in the quality of the effluent occurred. Flow gradually exceeded the ability of the plant to provide secondary treatment and by 1985, the flow reached a maximum of 415 million gallons per day. The level of discharged solid effluent peaked as well, as only about 35% of the solids were being removed at this time. Conditions improved in 1987, however, with the initiation of both a Hyperion treatment plant improvement project and a localized water conservation effort due to drought. The combination of initiating full secondary waste treatment and decreased flow to the plant dropped the total average suspended solids discharged and improved the removal efficiency of solids to >91% in 1995–1996. Today, the 5-Mile outfall is still in use and is the main effluent outfall for the Hyperion Treatment Plant. Only in emergencies is the 1-Mile outfall used to discharge chlorinated secondary effluent.

3. Materials and Methods

3.1. Field Sampling

In June 1997, a standard NEL box corer on board the R/V Sproul was used to obtain sediment at 21 sites selected in a stratified random design [5] on the shelf and slope of Santa Monica Bay for geological, geochemical, and biological studies (Figure 2;

Table 1. Metadata (core number, latitude, longitude, and water depth) for the 21 sampling sites used in this study from USGS Cruise S2-97-SC.

Core	Latitude (°N)	Longitude (°W)	Water Depth (m)
4	34.01350	118.53405	16
6	34.02117	118.65417	2
10	33.92133	118.56017	139
16	33.94183	118.58983	180
18	33.97933	118.57517	60
20	33.96883	118.53217	47
22	33.98017	118.55683	50
24	33.97117	118.65150	182
28	33.83850	118.55133	108
30	33.86333	118.51367	65
33	33.84683	118.46083	78
34	33.88667	118.47800	53
36	33.83333	118.51133	125
42	33.93333	118.50667	47
44	33.92483	118.50050	46
48	33.83850	118.42700	134
49	33.95033	118.48233	28
50	33.92783	118.53500	60
51	33.94283	118.57483	144
52	33.91383	118.57233	181
55	33.90900	118.52517	60

). The box core had a surface area of 20 by 30 cm and a maximum depth of penetration of 60 cm. Subsamples of the upper 0–1 cm and 0–2 cm intervals were taken and preserved for microfaunal and textural analysis, respectively. Four subcores, each consisting of a plastic tube with an 8.4-cm inside diameter, were also taken from the box cores for geochemistry, toxicity, geochronology, and stratigraphy. Each tube contained a piston that was firmly attached to the sub-sampling rig that maintained the sediment surface so that virtually no core shortening occurred and the proper sub-bottom depth reference in each subcore was preserved. The tubes were driven into the intact box core at a constant speed with an electric-powered actuator. A polycarbonate liner was used for the geochemistry and toxicity subcores; the former were then frozen at −30 °C and the latter were refrigerated. A polybutyrate liner was used for both the stratigraphy and geochronology subcores. Two-centimeter-long sections of the geochronology subcores were extruded on board ship, bagged, and sent to Skidaway Institute of Oceanography for Pb-210, Cs-137, and Th-234 dating to determine sediment accretion rates and rates of biological mixing [65,66].

3.2. Chronology

Eighty sediment samples from Santa Monica Bay were prepared for radiochemical analysis at the Skidaway Institute of Oceanography [65] following standard techniques described in Alexander et al. [66]. Sediment samples for Pb-210 dating were ground into a powder, let to equilibrate for 20 days in sealed jars, and then counted for approximately 24 h. Total Pb-210 activities were directly determined by measuring the 45.5-KeV gamma peak using ORTEC (Oak Ridge, TN, USA) LO-AX germanium detectors. Supported levels of Pb-210 from Ra-226 were determined by measuring the gamma activity of Pb-214 (295 and 352 KeV) and of Bi-214 (609 KeV), the short-lived daughter products of Ra-226. Lead-210 has a half-life of 22.3 years, thereby providing a steady-state chronology [67] for the past ca. 100 years [65,68]. Accretion rates for bulk sediment were calculated from the least- squares regression of the downcore excess PB-210 activity data for each core using the mass flux rate, which accounts for changes in porosity that can affect the vertical accretion rates [65].

Caesium-137 is a transient tracer [67] resulting from atmospheric nuclear testing and has a half-life of 30.0 years [65,68]. Chronologic markers using this isotope include: (1) the onset of the Cs-137 record in 1954 due to initial bomb fallout; (2) the first fallout peak in 1959; and (3) another peak in 1963 after which input declined when atmospheric testing was banned [67,69]. A later peak was recorded in some locations due to the Chernobyl accident in 1986, but it is largely absent from terrestrial deposits in North America [70]. In sediments off southern California, the peak in 1963 has now been recalibrated to 1971 [71] due to the region’s dry climate, delaying the transport of Cs-137-absorbed terrestrial deposits to the ocean. The Cs-137 record is often used to corroborate the sediment accretion rates determined by the Pb-210 profiles [67], although clarity of the peaks has become an issue in some cases over the last few decades [71] with the two peaks merging into one, becoming elongated, appearing as multiple peaks, or not being clearly defined, all most likely the result of bioturbation [72]. Large increases in sand-sized particles are also known to introduce potential error in Cs-137 profiles [71]. In this study, Cs-137 activities were determined by measurement of the 661.6-KeV gamma peak.

The naturally occurring isotope Th-234 has a short half-life of only 24 days and is instrumental in determining biological mixing rates and depths because it adheres to sinking particles in the water column and is quickly delivered to the seafloor where it is mixed into the sediment. Samples to be used for determining Th-234 activities in this study were analyzed soon after collection by drying the sediment, grounding it into powder, and counting it for 24 h in order to quantify its gamma peak at 63 KeV [65,68]. The samples were recounted four months later to quantify the supported Th-234 levels, from which the excess levels could be determined.

Several factors, such as grain size, changes in porosity, accretion rates, and, most importantly, rates and depth of mixing due to bioturbation, introduce error into the dating process so that Pb-210 accretion rates are only estimates of the true accretion rates [65,67]. In Santa Monica Bay, measured apparent accretion rates were found to be as much as twice the actual accretion rates at some locations because of deep mixing [65]. As a result, the error for individual accretion rates was estimated at approximately 10% based on the standard error of all linear regressions calculated for Pb-210, resulting in an error for the calculated ages of about 8 years [65]. Estimates of uncertainties from just south of Santa Monica Bay on the shelves off Palos Verdes Point and San Pedro (Figure 1B) report errors of sedimentary increments of 5–10 years due just to slight differences in the Pb-210 slope mean estimates, thereby shifting the age by 10–40 years [72], or as great as centennial to millennial scales for ~5 cm intervals based on molluscan assemblages [73]. Due to these inherent dating errors, only approximate calendar ages based on the average sediment accretion rates are reported in the present study.

Additional factors taken into consideration in determining the chronology of the cores were peaks in Cs-137, the abundance of dichlorodiphenytrichlorethane (DDT) and polychlorinated biphenyls (PCBs), and the granulometry of the core sediments. Due to the fact that (1) the production of DDT was banned in 1970 [74], (2) input of the organic compound to the Los Angeles County Sanitation Districts sewer system ceased in 1971 [75], (3) DDT has a half-life of days [76,77] to years [78,79], and (4) it is broken down quickly as a result of dehydrochlorination by microorganisms [80], a dramatic decline was seen in DDT concentrations in Santa Monica Bay after 1979 [81]. Following a similar trend, PCBs were first manufactured in 1929 and were banned in the United States in 1979 [82], so their abundance dropped precipitously in Santa Monica Bay after 1979 as well [81]. Interestingly though, the concentration of metals has not followed this pattern, showing little temporal change since the mid-1960s, except near the 7-Mile outfall [81]. Finally, a shift from a shell-gravel substrate to one of predominantly mud on the shelves of Santa Monica Bay (investigated on Short Bank) and Palos Verdes due to siltation resulting from grazing and agriculture without concern of soil management [83] provides yet another means to date the core sediments. An epifaunal assemblage of brachiopods and scallops associated with the shell-gravel substrate persisted for thousands of years, then declined in ca. 1820–1825 AD due to the pervasive soil runoff, with the shells of those organisms ceasing to be produced by ca. 1910 [83]. Therefore, the absence of any shelly remains in the three shelf cores of this study suggests the sediment that was recovered dates to after 1910. Using these parameters, three chronological horizons of interest were identified, representing the following time periods: (1) pre-pollution (prior to ca. 1957); (2) peak pollution (ca. 1957–1987); and (3) post-pollution (after the initiation of treated discharge in 1987).

3.3. Grain Size

Grain size analysis was performed by the U.S. Geological Survey on 21 surface and 46 downcore sediment samples using 2 cm sampling intervals with a Horiba (Kyoto, Japan) Model LA-900 Laser Scattering Particle Size Distribution Analyzer [84]. Sediment samples were circulated in the sample reservoir and the particle diameters determined by detection of scattered laser light. Seventy-four different size classes between 0.88 and 1.024 µm were measured. The mean was reported for each sample and the results were converted to phi units between −1 and 9 (very coarse sand to very fine silt to clay). The percent sand, mud, and clay were also determined. Only the % mud content (i.e., the combined % silt and % clay fractions) is reported in this study. As the sampling intervals often differed slightly for the grain size and foraminiferal analyses, the mud content was estimated for the foraminiferal sample intervals using the nearest grain size determination because those values were generally consistent downcore within each core.

3.4. Foraminifera

Fifty to 100 cm³ of core top sediment (0–2 cm; “recent/present day” = 1997) was sampled at each station for foraminiferal analysis. Once the Pb-210 dating information was available [65], 50 1-cm sub-surface intervals were also analyzed for foraminifers from seven cores (Figure 2). Sediment samples were wet-sieved through nested 0.63 mm, 0.150 mm and 1.0 mm screens to segregate the size fractions. Sediment remaining on the screens was transferred to filter paper and air-dried. Foraminifera were extracted from the >0.63 mm fraction. Each sample was split with the aid of a microsplitter into an aliquot containing at least 300 benthic foraminifers, and all specimens were picked and identified. Sandy samples containing few foraminifers were subjected to sodium polytungstate flotation at a specific gravity of 2.4 g/cc in order to concentrate the foraminifers before splitting and picking. Relative foraminiferal species abundances were computed using a sum of total benthic foraminifera and converted to frequency data. The slides and residues of this study are on file at the Pacific Coastal and Marine Science Center Micropaleontology Laboratory, U.S. Geological Survey, Menlo Park, CA, USA.

3.5. Macrofaunal Toxicity Testing

The macrofaunal toxicity tests used in this study included amphipod survival and sea urchin fertilization success. Unfortunately, a broader study of bivalve community structure was not also undertaken, which would have allowed for a better comparison between the foraminiferal associations and a wider range of macrofaunal toxicity tests. Both amphipod survival and sea urchin fertilization tests used standard analytical methods [85,86] respectively and were conducted by Southern California Coastal Water Research Project (SCCWRP) personnel [5]. Juvenile amphipods of the species Grandidierella japonica Stephensen were collected from nearby Upper Newport Bay (Figure 1B,D) less than one week prior to testing. Each core was sliced into a 6-cm section (0–6 cm), and 4-cm sections thereafter. Twenty amphipods were added to a homogenized sediment-water mixture from each section and were removed by screening after ten days. Sediment was considered toxic if less than 80% of the controls survived, a standard used in other studies [87,88].

The sea urchin fertilization test used sperm and eggs of the purple sea urchin Stronglyocentrotus purpuratus (Stimpson) collected from the intertidal zone in northern Santa Monica Bay (Figure 1B,C) and water from the sediment water-interface of the amphipod survival test on the first day of the amphipod exposure. Sperm were first subjected to 20 min of exposure to this water. Eggs were then added and left for 20 min, allowing fertilization to proceed. The eggs were removed, preserved, and later examined with the aid of a microscope to determine if fertilization occurred. Sediments were considered toxic to sea urchins if the mean fertilization was significantly lower (p < 0.05; less than 80%) than the controls. Amphipod survival and sea urchin fertilization results presented as greater than 100% indicate that the animals did better in the sediment than in the controls.

3.6. Chemical Analysis

Cores were stored frozen at −20 °C until the metals, organic compounds, and total organic carbon (TOC) (together, referred to as Potential Toxic Elements in Martínez-Colón et al. [17,18,19] were analyzed by staff at SCCWRP [79]. The cores were then allowed to thaw, and sub-sampled with a piece of beveled glass into 1 or 2 cm sections at the desired depths based on the Pb-210-dated horizons. Trace metals analyzed included arsenic, beryllium, cadmium, chromium, copper, lead, mercury, nickel, silver, zinc, and iron; other sediment constituents included unionized ammonia (toxicity is thought to be primarily attributable to this form, NH₃, compared to ionized ammonia, NH₄) and percent total solids; and organics included total DDT, total PCBs, and total polycyclic aromatic hydrocarbons (total PAHs). The analyses were conducted as described in Bay et al. [81].

Each sediment sample to be measured for metals was digested using Method 3055 of the United States Environmental Protection Agency SW-846 protocols (third edition), and eventually analyzed using Method 6010 of the U.S. EPA SW-846 protocols. Some metals expected to have low concentrations were analyzed with graphite furnace atomic absorption spectroscopy (U.S. EPA Method 7010), and mercury was analyzed with a cold vapor method (U.S. EPA Method 7471).

Samples to be measured for organics were extracted using a microwave-assisted solvent extractor (MSP 1000 Microwave Sample Preparation System; CEM Corporation, Matthews, NC, USA). Total DDT components and PCB congeners were measured using an HP 5890 II GC equipped with a 63Ni electron capture detector (ECD) and a 60 m 0.25 μm I.D. (0.25 mm film thickness) DB-5 silica fused capillary column (J&W Scientific, Folsom, CA, USA). Total PAHs were measured using an HP 5890 II GC equipped with a 5970 mass spectrometer (MS) and a 60 m 0.25 mm I.D. (0.25 μm film thickness) DB-5 silica fused capillary column (J&W Scientific, Folsom, CA, USA).

Sediment TOC was obtained from the difference between the measured concentrations of total carbon and total inorganic carbon, the latter of which were analyzed on a carbon dioxide coulometer (System Model 5010; UIC, Inc., Fremont, CA, USA).

3.7. Data Analysis

A Spearman rank-order correlation using Aabel (v.3), a statistical software package created by Gigawiz Ltd. Co. (Tulsa, OK, USA), was used to analyze the foraminiferal abundances of B. denudata and E. advenus, as well as the toxicity and chemistry data, to determine any associations that might exist between them. This nonparametric correlation was applied to the biotic (percentage abundances of benthic foraminifera, sea urchin fertilization success, and amphipod survival) and abiotic (environmental) data, the latter comprised of sediment concentrations of trace metals and organics as well as TOC, total solids (% TS), mud content (%), and unionized ammonia. When the biotic data, % TS and % mud were correlated among themselves, the variables were not transformed because they were all expressed in similar units (percentage abundance) and none of these were rare so there was no need to down-weight the importance of the highly abundant constituents [89]. In contrast, the abiotic data were measured on different scales, warranting a log transformation followed by normalization [89,90,91]. Both biotic and abiotic correlations were interpreted at the p < 0.05 confidence level. Species abundance, toxicity, and chemistry data were also compared using effects range-low (ERL) and effects range-median (ERM) values; the ERL and ERM concentrations are the concentrations above which adverse biological effects are possible (10%) and probable (50%) for organisms living in the sediment, respectively [92].

Sediment concentrations in this study are occasionally presented as less than a threshold instead of an actual number (Appendix A). For example, in the case of silver, the concentration is reported in 39 samples as <2.6 µg/g, making it impossible to determine exceedances above the ERL of 1.0 µg/g (

Table 2. Sediment quality guidelines (ERL and ERM). ERL = Effects Range-Low concentration and ERM = Effects Range-Median concentration. ERL and ERM values from Long et al. [76].

Sediment Chemistry Analyte		Sediment Quality Guidelines
		ERL Value (µg/g Dry Weight)	ERM Value (µg/g Dry Weight)
Trace Metals	Arsenic	8.2	70
	Cadmium	1.2	9.6
	Chromium	81	370
	Copper	34	270
	Lead	46.7	218
	Mercury	0.15	0.71
	Nickel	20.9	51.6
	Silver	1	3.7
	Zinc	150	410
Trace Organic Compounds	Total PAHs	4022	44,792
	Total DDT	1.58	46.1
	Total PCBs	22.7	180

). As a result, the abundance of silver was not plotted in respect to ERL or ERM concentrations in the cores, nor included in the Spearman correlation. For three other sediment chemistry analytes (<17 µg/g of copper in 29 samples, <0.42 µg/g of cadmium in four samples, and <0.04 µg/g of mercury in one sample), the concentrations did not affect determinations of any ERL or ERM exceedances, so these were plotted downcore but were eliminated from the Spearman correlation.

4. Results

Of the seven cores collected in this study, average sediment accretion rates ranged from 0.31 to 0.52 cm/year on the shelf and 0.19 to 0.49 cm/year on the slope (Figure 3; [65]). The locations with the highest accretion rates were Core 6 (0.52 cm/year) off Malibu Point and Core 10 (0.49 cm/year) near the 7-Mile outfall (Figure 2). Based on the Pb-210 profiles and, where available, penetration depths of excess Th-234, the SML ranged in depth from <1.5 cm to 23 cm, with the deepest mixed layers occurring in cores 6 and 34 from the shelf. Calendar ages based on the Pb-210 accretion rates suggest that most of the cores date back to the early 1900s, and the lack of shell-gravel sediment at the bottom of the three shelf cores also indicates they are younger than ca. 1910 [83]. However, two cores (36 and 52) obtained on the slope may have captured the latter portion of the 19th Century, although the Cs-137 record of Core 36, as well as the persistence of PCBs downcore in Core 52, suggests they may be younger. The chronologies are supported by the depths of the SML, as the effective resolution of the sampled intervals (the SML/accretion rate) compares well with the calendar years assigned to the upper portion of each core. Most cores also displayed two or more peaks in Cs-137.

Bulimina denudata averaged 16% of the foraminiferal assemblage of the 21 surface samples analyzed (Figure 4A). Eggerelloides advenus was locally more abundant and the species’ abundance only slightly higher on average (18%; Figure 4B). The amphipod survival test identified only three of 21 surface samples as being toxic (Cores 6, 36, and 52; Figure 4C; Appendix A), with survival rates of 68, 38 and 74%, respectively, whereas the sea urchin fertilization test indicated that the surface of only Core 10 was toxic (28% fertilization success; Figure 4D; Appendix A). The ERM values were exceeded for the trace metals cadmium, copper, mercury, and silver, and trace organics (total DDT and total PAHs) in Core 10; for silver in Cores 16, 51 and 52; and for DDT in Cores 28, 30, 33, 36, and 48 (Figure 5 and Figure 6) [83]. The ERL values were exceeded in two cores for arsenic, seven for cadmium, six for chromium, four for copper, one for lead, eleven for mercury, seven for nickel, one for zinc, 15 for total DDT, and one for total PCBs.

The abundance of B. denudata in the 48 downcore samples was higher in most cases than in the surface samples (18%) whereas E. advenus was less abundant (averaging 10%) except for near-recent samples in Cores 6 (4–5 cm, 66%) and 20 (2.5–3.5 cm; 53%) (Figure 7A–G; Appendix A). The amphipod survival test suggests far more downcore samples were toxic compared to the surface material (21 versus 3), yet the sea urchin fertilization test only identified the five samples from Core 10 as toxic. The ERM values were exceeded for nearly all metals in Core 10 (cadmium, chromium, copper, lead, mercury, nickel, silver, and zinc), as well as total DDT and total PCBs. The ERM value for total DDT was also exceeded in Cores 20, 34, 36, and 52. The ERL values were exceeded in a minimum of three cores for cadmium, chromium, mercury, nickel, total DDT, and total PCBs, two for copper, and one for arsenic.

In all samples (surface and downcore), a weak but significant positive correlation was found between the abundance of B. denudata and both amphipod survival (r_s = 0.263;

Table 3. Spearman rank correlations for species abundances of Bulimina denudata and Eggerelloides advenus with macrofaunal toxicity results (sea urchin fertilization and amphipod survival). Values square root transformed before correlation. Significant correlations (p < 0.05) are in square brackets and bold.

			Macrofaunal Toxicity Test
Samples	Number of Samples	Benthic Foraminiferal Species	Sea Urchin Fertilization		Amphipod Survival
			rs	ps	rs	ps
Surface & Downcore	67	Bulimina denudata	−0.015	>0.5	0.263	[0.032]
		Eggerelloides advenus	−0.252	[0.040]	0.022	>0.5
Surface & Downcore (except core 20)	59	Bulimina denudata	−0.028	>0.5	0.330	[0.011]
		Eggerelloides advenus	−0.292	[0.025]	−0.028	>0.5
Surface	21	Bulimina denudata	0.012	>0.5	0.129	>0.5
		Eggerelloides advenus	−0.222	0.334	0.012	>0.5
Downcore	46	Bulimina denudata	−0.037	>0.5	0.275	0.064
		Eggerelloides advenus	−0.340	[0.021]	−0.010	>0.5
Downcore (except core 20)	39	Bulimina denudata	−0.041	>0.5	0.350	[0.029]
		Eggerelloides advenus	−0.437	[0.005]	−0.083	>0.5

) and total DDT (r_s = 0.294;

Table 4. Spearman rank correlations for species abundances of Bulimina denudata and Eggerelloides advenus with sediment constituents, including sediment chemistry analytes, TOC, total solids (% TS), and mud content (% mud). Values log transformed and normalized before correlation except for % mud and % TS that needed no transformation. Significant correlations (p < 0.05) are in square brackets and bold.

	Surface & Downcore (49 Samples)				Surface (21 Samples)
	Bulimina denudata		Eggerelloides advenus		Bulimina denudata		Eggerelloides advenus
Sediment Constituent	rs	ps	rs	ps	rs	ps	rs	ps
As	−0.296	[0.039]	−0.285	[0.048]	−0.120	>0.5	−0.438	[0.047]
Be	−0.068	>0.5	−0.558	[<0.001]	0.154	>0.5	−0.592	[0.005]
Cd	−0.326	[0.022]	−0.217	0.134	0.127	>0.5	−0.334	0.139
Cr	0.126	0.387	−0.436	[0.002]	0.362	0.107	−0.832	[<0.001]
Pb	0.115	0.431	−0.070	>0.5	0.159	0.490	−0.501	[0.021]
Hg	0.016	>0.5	−0.274	0.057	0.209	0.362	−0.481	[0.027]
Ni	0.036	>0.5	−0.345	[0.015]	0.344	0.127	−0.479	[0.028]
Zn	0.006	>0.5	−0.442	[0.001]	0.237	0.302	−0.704	[<0.001]
Fe	0.025	>0.5	−0.426	[0.002]	0.230	0.316	−0.631	[0.002]
Total DDT	0.294	[0.041]	−0.183	0.209	0.217	0.346	−0.257	0.260
Total PCBs	−0.029	>0.5	−0.364	[0.010]	−0.063	>0.5	−0.221	0.337
Unionized Ammonia	−0.317	[0.026]	0.357	[0.012]	−0.146	>0.5	0.266	0.243
TOC	−0.289	[0.044]	−0.234	0.106	0.075	>0.5	−0.521	[0.015]
% TS	0.023	>0.5	0.181	0.213	−0.282	0.216	0.611	[0.003]
% Mud	0.116	0.354	−0.113	0.354	0.370	0.108	−0.143	>0.5
	Downcore (28 Samples)				Downcore (Except Core 20) (24 Samples)
	Bulimina denudata		Eggerelloides advenus		Bulimina denudata		Eggerelloides advenus
Sediment Constituent	rs	ps	rs	ps	rs	ps	rs	ps
As	−0.388	[0.041]	−0.258	0.185	−0.452	[0.026]	−0.203	0.341
Be	−0.242	0.215	−0.501	[0.007]	−0.197	0.356	−0.477	[0.018]
Cd	−0.560	[0.002]	−0.375	[0.049]	−0.521	[0.009]	−0.307	0.145
Cr	−0.0004	>0.5	−0.308	0.111	0.008	>0.5	−0.271	0.201
Pb	0.132	>0.5	0.131	>0.5	0.204	0.338	0.184	0.389
Hg	−0.019	>0.5	−0.242	0.215	−0.001	>0.5	−0.212	0.321
Ni	−0.252	0.195	−0.167	0.396	−0.215	0.312	0.002	>0.5
Zn	−0.130	>0.5	−0.352	0.066	−0.137	>0.5	−0.306	0.146
Fe	−0.042	>0.5	−0.319	0.098	0.088	>0.5	−0.257	0.225
Total DDT	0.270	0.164	−0.050	>0.5	0.304	0.180	−0.315	0.164
Total PCBs	−0.093	>0.5	−0.375	[0.049]	−0.212	0.356	−0.552	[0.010]
Unionized Ammonia	−0.407	[0.032]	0.181	0.356	−0.443	[0.039]	0.131	>0.5
TOC	−0.437	[0.020]	−0.217	0.268	−0.488	[0.018]	−0.337	0.116
% TS	0.173	0.379	0.098	>0.5	0.072	>0.5	0.014	>0.5
% Mud	0.022	>0.5	−0.077	>0.5	0.019	>0.5	−0.101	.0.5

). A weak but significant negative correlation was evident between the abundance of E. advenus and sea urchin fertilization success (r_s = −0.252; Table 3) and the abundance of B. denudata and arsenic, cadmium, unionized ammonia, and TOC (r_s = −0.289 to −0.326, Table 4). Nine sediment constituents significantly correlated with the abundance of E. advenus, eight were weakly to moderately negatively correlated (arsenic, beryllium, chromium, copper, nickel, zinc, iron, TOC, and total DDT; r_s = −0.285 to −0.558) and unionized ammonia was moderately positively correlated (r_s = 0.357). When only the surface samples were analyzed, no significant correlations were found between any sediment constituent and the abundance of B. denudata, or between the abundance of either species versus amphipod survival and sea urchin fertilization. In contrast, a moderate to strong negative correlation was evident for eight trace metals (arsenic, beryllium, chromium, lead, mercury, nickel, zinc, and iron) and TOC with the abundance of E. advenus (r_s = −0.438 to −0.832) as well as a strong positive correlation with percent total solids (r_s = 0.611). In the downcore samples, a moderate negative correlation (r_s = −0.388 to −0.560) was found between the abundance of B. denudata and arsenic, cadmium, unionized ammonia, and TOC, whereas only beryllium, cadmium, total PCBs, and sea urchin fertilization success correlated moderately negatively with the abundance of E. advenus (r_s = −0.340 to −0.501).

Grain size, reported here as % mud content, is lowest on average on the outer shelf at site 50 (19%) and on the mid–shelf north of Short Bank at site 34 (32%), whereas the highest values occur west of Short Bank at site 36 (95%), at the bottom of core 24 (82% at 32–33 cm), in surface sediment at site 16 (81%), at site 10 located near the terminus of the 7-Mile outfall from a depth of 25 cm to the core top (63–76%), and at site 6 near Malibu Point (66%) (Figure 2). No correlation was found between mud content and the abundance of either B. denudata or E. advenus in the surface or downcore samples when analyzed separately, or when all of the surface and downcore samples were analyzed together (Table 4).

5. Discussion

5.1. General Trends of Toxicity in Santa Monica Bay

The sediment chemistry measurements [81] and macrofaunal (amphipod survival and sea urchin fertilization) toxicity tests [5] indicate that many surface and downcore sediment samples in Santa Monica Bay were toxic, including those dated prior to 1957–1960 when the 7-Mile and 5-Mile outfalls became operational. The most likely explanation for the presence of toxic sediments downcore is due to the downward migration of contaminants by bioturbation [5]. The spatial and temporal pattern of toxicity suggests that the historical inputs of the 7-Mile sludge outfall was the most likely source of the pollution, and that remediation has improved sediment quality and decreased the size of the impacted area, so that only a few sites in the bay were toxic when the samples were collected in 1997 [5].

In Santa Monica Bay, the amphipod survival tests identified toxic samples more often than the sea urchin fertilization tests, whereas the toxicity effect of the latter was more pronounced [5]. According to Greenstein et al. [5], amphipod survival rates negatively correlated (r_s = −0.242) only with the concentration of silver in the sediment, whereas poor success with sea urchin fertilization was associated with high concentrations of zinc, silver, copper, cadmium, and total PCBs and PAHs. The high levels of DDT found throughout the bay [81] did not correlate significantly with either amphipod survival or sea urchin fertilization, possibly due in part to the fact that the amphipod used in this study (Grandidierella japonica) is tolerant of the pesticide [93], and neither did grain size [5]. These tests suggest that although both amphipods and sea urchins are sensitive to pollutants, they respond to sediment constituents in a different manner.

5.2. Temporal Trends of Toxicity in Santa Monica Bay

Seven times more downcore samples proved to be toxic according to the amphipod survival test than at the surface (Appendix A). This is due to the fact that the major source of contamination historically to Santa Monica Bay was the contaminant inputs from wastewater discharge from the Hyperion Outfall system that declined drastically during the last few decades as a result of improved treatment processes and better source control [81].

The temporal pattern of toxicity was most evident in Core 10 which was collected near the terminus of the 7-Mile outfall (Figure 2) and is the longest and most contaminated of this study. The Pb-210 profile indicates the core contains sediment dating back to the early 1900s and the SML extends down to 6.5 cm (Figure 3). However, the Cs-137 record suggests deposition starting in the middle 1900s. Unfortunately, no trace organic compounds were measured prior to 24–25 cm, but at that depth DDT and PCBs were already at ERM levels (Appendix A) and at the highest concentrations recorded in any of the cores. Therefore, the Pb-210 chronology appears to be indicating calendar ages that are too old downcore. Based on the Cs-137 peaks and the DDT and PCB values, the pre-pollution period is thought to range from the bottom of the core to ca. 34 cm, the peak pollution period from ca. 34–5 cm, and the post-pollution from ca. 5 cm to the core top.

Although both macrofaunal toxicity tests suggest the sediment was toxic during the pre-pollution period (Figure 7A), at 38–39 cm sediment chemistry values of the trace metals were low (only cadmium and mercury exceeded ERL values; Appendix A), mud content was 40%, TOC was moderately low at 0.907%, and the foraminiferal density (i.e., FD = the number of foraminifera/gram dry sediment) increased to 2511 from 1621 further downcore at 44–45 cm (Figure 7A). Together, these indicate that the concentrations of the trace metals were not, in fact, toxic to the benthic foraminifera. However, since the trace organic compounds were not analyzed at this depth in the core, they cannot be ruled out as being responsible for the macrofaunal toxicity results.

From 1960 to 1987, the 7-Mile outfall discharged untreated effluent and roughly 7x10⁹ L of sludge annually, and even larger amounts from 1982–1987 due to reduced levels of treatment because of decreased efficiency of plant operations [94]. As a result, the % total solids decreased by nearly one half (Figure 6) and the sediment accretion rate at the core site which averaged 0.49 cm/year below the SML (Figure 3) is reported to have been as high as 2.28 cm/year adjacent to the 7-Mile outfall [81]. Foraminifers were extremely rare during this peak pollution period, particularly at 6–7 cm and 15.5–16.5 cm when the site nearly became a dead zone. At 24–25 cm, all of the trace organic compounds and nearly all of the trace metals exceeded ERM levels, both amphipod survival or sea urchin fertilization toxicity tests identified the sediments as contaminated, mud content rose to 63%, TOC increased dramatically to 7.3%, and the FD dropped significantly to 379 (Figure 5, Figure 6 and Figure 7).

Most chemical constituents continued to exceed ERM levels at the latter stages of the peak pollution period. These exceedances are not surprising with the continued outfall discharges and the mud content reaching a peak of 76% at this time, as fine-grained sediments are known to adsorb heavy metals and organic compounds [95,96,97]. At 12–13 cm, total organic carbon increased to the highest concentration of any sample in this study (8.4%) and the FD continued to decline (to 141). The amphipod survival and B. denudata abundance curves were similar with depressed values, as were the sea urchin fertilization success and abundance of E. advenus (Figure 7A).

Site 10 was the only surface site (0–2 cm) considered toxic based on the sea urchin fertilization test, as well as being the most contaminated of the surface sites based on sediment chemistry, exceeding ERM values for cadmium, copper, mercury, silver, total DDT, and total PAHs, and ERL levels for arsenic, chromium, lead, nickel, zinc, and total PCBs (Figure 5 and Figure 6). However, the FD increased by nearly 1000 (1126; Figure 7A) as the mud content decreased to 69% and the TOC value, although still very high, decreased to 5%. The high FD possibly reflects the ability of foraminifera to use organic waste as a food source [10,29,57]. The abundance of B. denudata (28%) also increased by a factor of four over its abundance at 12–13 cm, exceeding its average of 16% for all of the surface samples (Appendix A), and again, possibly demonstrating the species’ propensity for settings with high food availability and low oxygen [56,57,58]. In contrast, the abundance of E. advenus did not change dramatically from that at 12–13 cm (3% to 7%), compared to an average of 18% for all of the surface sites, and its abundance negatively correlated with most trace metals, TOC, and % TS in the surface samples (Table 4).

The second most contaminated site was 6 (Figure 2), located far north of the 7-Mile or 5-Mile outfall sites. The core was collected on the shelf near Malibu Point, where sediment is introduced by Malibu Creek [91], one of the three sources (the others being Ballona Creek and the Hyperion outfalls) of fine-grained particulate material to the margin of Santa Monica Bay [98,99]. As a result, the mud content of the core averaged 66% and the average sediment accretion rate was the highest of any cores of this study (0.52 cm/year) [65]. The Pb-210 profile suggests the core contains a record of deposition beginning in the middle 1900s, much of which was highly bioturbated, with the SML reaching a depth of 23 cm (Figure 3). This is supported by the Cs-137 record which indicates deep mixing because the values never reached zero downcore. Despite this, Cs-137 peaks in the 1950s and ca. 1971 were observed, similar to those seen in Core 10. The deepest sample analyzed for organic compounds was at 24–25 cm and DDT was recovered at the ERL level, suggesting toxic sediment was already being deposited. Combined, the pre-pollution period appears to be from 29–23 cm, the peak pollution period from 23 to 5 cm., and the post-pollution period from 5–0 cm.

According to the amphipod survival test, sediment from Core 6 was toxic for most of the length of the core, including the core top (Figure 7B), even though no ERM values were exceeded (Figure 5 and Figure 6; Appendix A). However, ERL values were surpassed at many sub-surface levels and in the surface sediment for cadmium, mercury, nickel and total DDT; no PCBs were recovered. Throughout the core, the abundance of TOC was moderately low (0.858–1.065; Appendix A) and the FD remained low (288–528, average 375). The exception was the sample from 18–19 cm in which amphipod survival was no longer toxic (94%), the FD increased by a factor of ~10 (to 3358), and the abundance of B. denudata was one of the highest in the core (17%). These factors possibly suggest a short-term improvement in sediment quality, although some sediment chemistry data are missing (i.e., TOC and total organic compounds) which hinders an explanation. The curves for amphipod survival and the abundance of B. denudata do not track each other as well in this core as in some of the others, particularly from ca. 15 cm and above (Figure 7B). This is also the only core in which the abundance of E. advenus comprised a dominating >65% of the assemblage at two intervals: from 14.5–15.5 cm during the peak pollution period and at 4–5 cm after remediation began. In both instances, the abundance of B. denudata dropped to <5%. The poor tracking of B. denudata abundance with amphipod survival and dominance by E. advenus over B. denudata are not consistent with the other cores in this study and likely reflect the significant sediment mixing occurring throughout most of the core. However, the spikes in abundance of E. advenus may also reflect the species’ ability to quickly colonize formerly impacted areas [50,63]. Interestingly, the sea urchin fertilization success showed little change and indicated no toxicity throughout the core.

Core 52, obtained on the slope near Core 10 in the vicinity of the 7-Mile outfall terminus (Figure 2) contains a sediment record going back to the late 19th Century according to the Pb-210 profile (Figure 3), with an average sediment accretion rate of 0.25 cm/year [65] and an SML at the shallowest of any core (<1.5 cm). Peaks in the Cs-137 record occur in the upper half of the core, thereby indicating the latter half of the 1900s, although they are not as well defined as in other cores. Unfortunately, trace organics were not measured at every horizon in this core, yet PCBs are present at 24–25 cm, suggesting the sediment is at least from >1929 at that depth. The dramatic increase in DDT and PCBs to ERM and ERL levels, respectively, from 10–6 cm suggests this interval represents the peak pollution period. The core is characterized by sediment exceeding ERM values for silver and total DDT, as well as surpassing ERL values for cadmium, copper, chromium, nickel, and total PCBs downcore (Figure 5 and Figure 6; Appendix A). The only change from this pattern in the core top sediment was a slight reduction in the amount of total DDT, where it exceeded only the ERL level (Figure 6A). The mud content steadily increased from 44% at the bottom of the core (36–37 cm) to 53% at the top, as did the TOC (0.9% to 2.14%), reflecting increased input from the outfall, although the average sediment accretion rate for the core did not change significantly after the outfalls became active.

The amphipod survival test showed evidence of toxicity for most of the lower half of Core 52 and at the top of the core as well (Figure 7C). Similarities were seen between the amphipod survival and B. denudata curves, with both reaching peak abundances near the top of the peak pollution period at 6–7 cm (108% and 16%, respectively) and then declining toward the top of the core during the post pollution period (74% and 11%). Only the bottom of the core showed opposite trends, a pattern which cannot be explained at this time. The sea urchin fertilization success and the abundance of E. advenus exhibited reverse patterns, as the former exceeded the success of the controls, but the latter made up only a small proportion of the foraminiferal assemblage throughout the core (0–2%). Foraminiferal density fluctuated between extremely high (6878–10,775) and lower (537–981) values in the core, yet displayed a general trend of decreasing upcore, eventually with a FD = 1903 at the core top. This pattern did not coincide with an obvious change in macrofaunal or sediment chemistry toxicity and also cannot be explained at this time.

Core 36 was collected in the southern portion of Santa Monica Bay on Short Bank (Figure 2), far from the outfalls and at a depth of 125 m. As with Core 52, it contains late 19th Century deposits based on the Pb-210 record, although the Cs-137 record suggests the early 1900s instead. This site had the highest mud content (95%) and lowest average sediment accretion rate (0.19 cm/year; Figure 3) [65] of any in this study. The lower boundary of the SML, based on the Pb-210 and Th-234 records, was fairly shallow at 3.5 cm. However, multiple peaks are seen in the Cs-137 profile and values never reached zero downcore, suggesting both surficial and deep mixing in this core. An offset is also evident in the Pb-210 and Cs-137 records at ca. 20 cm, which marks a change in deposition. With missing chemical constituent data for the lower portion of the core, the estimate of the pollution boundaries are difficult; it is suggested that the onset of the peak pollution period coincides with the first (and already extreme) occurrence of DDT and PCBs at 12–13 cm, just below the 1959 Cs-137 peak, and continues up to 2 cm.

The ERM values were exceeded for total DDT for all sediment chemistry samples from the 13 cm to the core top, whereas ERL values were exceeded for cadmium, chromium, copper, nickel and total PCBs at many levels (Figure 5 and Figure 6; Appendix A). The DDT may have been transported to the core site by prevailing currents that flow toward the northwest along the shelf and upper slope [100] from the Palos Verdes margin where a large DDT deposit is located [101].

Generally, the amphipod survival and B. denudata curves are similar for Core 36 (Figure 7D), except for the addition of two foraminiferal samples at 6.5–7 cm and 4–5 cm where amphipod survival data were not available. The amphipod toxicity data identified the core top and lowermost (27–29 cm) sediment as toxic, and B. denudata abundances also declined there. Abundance of E. advenus was low throughout most of the core (<0.5–2%), only increasing to a maximum of 10% at 6.5–7 cm and 8% in the core top when B. denudata declined. Additionally, as with Core 52, E. advenus showed an opposite trend from the sea urchin fertilization success, which surpassed the controls for the entire core. This site had, by far, the highest FD (8047–63,127) of any core of this study, despite the fact that TOC was low from 8–9 cm on (0.062–0.017%) (Figure 7D; Appendix A). The extremely high FDs are attributed to a slow sediment accretion rate at the site. Although not quite as high, a FD = 26,258 was reported in the southwestern margin of Biscayne Bay, Florida, a region thought to be influenced by reduced salinity and elevated organic-carbon concentrations [102].

Core 20, obtained on the mid-shelf (Figure 2), features a Pb-210 profile suggesting a sediment record back to the early 1900s with an average sediment accretion rate of 0.31 cm/year (Figure 3) [65], although the Cs-137 profile indicates only sediments from the middle 1900s were deposited (Figure 3). The Pb-210 and Th-234 records indicate the SML is from 0–5.5 cm. As DDT and PCBs occur at high concentrations near the bottom of the core (20–22 cm), most of the core is considered to represent the peak pollution period. The boundary with the post-pollution periods is taken to be at ca. 6 cm based on foraminiferal and macrofaunal toxicity data.

Site 20 was most affected by total DDT, exceeding ERM values once downcore and at all other levels for ERL values (Figure 5 and Figure 6; Appendix A). Otherwise, ERL values were exceeded only rarely for cadmium and mercury. As the average sediment accretion rate for this was not as high as those found at several other sites, especially those near the outfall, the low FD values (81–572, averaging 313; Figure 7E) from 25–12 cm indicate stressed environmental conditions existed in the peak pollution period.

A saw-toothed pattern of both amphipod survival and success of sea urchin fertilization is evident in Core 20 (Figure 7E), with amphipod survival also indicating sediment toxicity at 20–22 cm and 12–13 cm, and sea urchin fertilization indicating only non-toxic sediment. This unique pattern is also seen in the abundance of E. advenus, with peaks in abundance alternating with B. denudata. This is the only other core besides Core 6 in which the abundance of B. denudata does not generally follow that of amphipod survival; the saw-tooth pattern is present but is offset slightly downcore. However, because (1) every effort was made to disturb the sediment as little as possible during box coring and subsequent sub-coring, (2) the Pb-210 and Th-234 records indicate the SML did not extend through most of the length of the core like Core 6 but was restricted to the upper 5.5 cm, (3) no unique array of contaminants were found at this site, and (4) the abundance of B. denudata and amphipod survival has tracked fairly well in all other cores, another factor may be responsible for the offset pattern. That is, it may be due to the presence of unseen gently dipping beds in the box core that resulted in the sediment of the two subcores (for foraminiferal analysis and macrofaunal toxicity) actually being non-contemporaneous when sampled at the same depth downcore.

Core 24 was obtained on the northern portion of the Ramp (Figure 2) and had an average sediment accretion rate of 0.43 cm/year (Figure 3) [65]. Based on the Pb-210 and Th-234 values, the SML extended to 4.5 cm. The Cs-137 profile is not well defined in the core and few samples were measured for sediment constituents, making the depositional history difficult to interpret. Based on the presence of DDT at ERL levels at 12–13 cm, the pre-pollution period is thought to range from 33 to 12.5 cm, the peak pollution period from 12.5–5 cm, and the post-pollution period from 5–0 cm.

The only sediment chemistry constituents to exceed ERL values in Core 24 were nickel and total DDT (Figure 5 and Figure 6; Appendix A). Except for an anomalously low value from 7.5–8.5 cm, the curves of B. denudata abundance and amphipod survival were similar, whereas the abundance of E. advenus was opposite that of B. denudata (Figure 7F). The amphipod survival test suggested sediment quality was poor prior to the 1950s, although the FD did not indicate this with values in the 1400s. Bulimina denudata abundances were also low at this time, reached a maximum during the peak pollution period (at 6–7 cm), and then declined after remediation was undertaken in 1987. The FD value was the highest (3422) for this core at 2.5–3.5 cm, indicating a return to non-impacted conditions after remediation.

Core 34 was obtained on the shelf near Short Bank (Figure 2). This core was similar to the other shelf cores by having an average sediment accretion rate of 0.32 cm/year like Core 20 [65] and having a substantial SML like Core 6, but here it only reached a depth of 17 cm downcore instead of 23 cm (Figure 3). The Pb-210 profile suggests the record goes back to the early 1900s, but the Cs-137 record indicates only the middle 1950s. The latter is supported by the occurrence of PCBs at an ERL level at the bottom of the core (28–29 cm), making the age > 1929. Maximum concentrations of DDT and PCBs occur from the bottom of the core to 16–17 cm, suggestive of the peak pollution period with no pre-pollution period represented. This coincides with the second Cs-137 peak, interpreted as 1971. No PCBs were recovered from 8–9 cm to the core top and DDT values also decreased, marking the post-pollution period (9–0 cm).

Only one sediment sample (22–23 cm; Appendix A) in Core 34 was contaminated according to the amphipod survival test. Except for total DDT, ERL values were exceeded only for mercury and total PCBs at two levels (Figure 5 and Figure 6). The Total DDT concentration exceeded ERL values in the upper 5 cm of the core and ERM levels for the remainder of the core. The abundance curves of B. denudata and E. advenus are opposite one another, with B. denudata maxima at 8–9 cm and 4–5 cm (Figure 7G). Like that seen in Cores 6 and 20 also from the shelf, this pattern of alternating dominance by B. denudata and E. advenus, particularly upcore from 17 cm, coincides with the position of the SML and may reflect bioturbation. The amphipod survival and B. denudata abundance curves are generally similar, except for the latter’s low value at 12–13 cm, whereas sea urchin fertilization shows little change throughout the length of the core. The FD is generally ≤1000 in the core except for particularly low values at the bottom of the core at 28–29 cm (303), at the beginning of the post-pollution period (422–527), and especially in the 1997 core top (194).

5.3. Bulimina denudata and Eggerelloides advenus

Bulimina denudata comprised an average of 16% of the foraminiferal assemblage of the surface samples, residing most commonly on the outer shelf and upper slope (Figure 4A; Appendix A) [50]. No significant correlation was found in these samples between the abundance of B. denudata and high concentrations of any trace metals, trace organic compounds, other constituents measured, or macrofaunal toxicity (amphipod survival or sea urchin fertilization) (Table 3 and Table 4). These results are in contrast to the amphipod survival test that identified three surface sites as toxic, and the sea urchin fertilization test having identified an additional site (Appendix A). The species increased in abundance to an average of 19% of the assemblage downcore and was negatively correlated with arsenic, cadmium, unionized ammonia, and TOC. Comparison of amphipod survival and the abundance of B. denudata illustrates that they are generally in good agreement except for the unique saw-toothed pattern in Core 20 that may be due to analyzing non-contemporaneous sediments or in the regions impacted by sediment mixing in cores 6 and 34 (Figure 7B,E,F). When Core 20 was eliminated from the data set, the abundance of B. denudata downcore showed a moderately positive correlation with amphipod survival (r_s = 0.350; Table 3) and arsenic, cadmium, unionized ammonia, and TOC (r_s = 0.443–0.521; Table 4). As the species is presently abundant on the outer shelf and upper slope and was so prior to the addition of the major sewage outfalls [50], no point source of contamination appears to be responsible for its present spatial distribution in Santa Monica Bay (Figure 4A).

Eggerelloides advenus comprised an average of 18% of the assemblage of the surface samples and was most abundant on the shelf (Figure 4B; Appendix A) [50]. The species’ abundance strongly correlated (r_s = 0.611) with total solids as well as negatively with TOC, and all of the trace metals analyzed except cadmium (r_s = −0.438 to −0.832; Table 4). This species’ tolerance of cadmium is similar to that found in Chaleur Bay, New Brunswick, Canada where E. advenus was the single most abundant species in cadmium-rich tailings [63]. The abundance of E. advenus also showed no significant correlation with either amphipod survival or sea urchin fertilization success in the surface samples (Table 3). In the downcore samples, beryllium, cadmium, total PCBs (r_s = −0.375 to −0.501), and sea urchin fertilization success (r_s = −0.340) negatively correlated with the abundance of E. advenus. However, when Core 20 with its presumably offset sediments was removed from the analyses, only beryllium and total PCBs (r_s = −0.477 to −0.552) as well as sea urchin fertilization displayed a moderate correlation (r_s = −0.437), the latter even more significantly (dropping from p_s = 0.021 to p_s = 0.005). Although the species is presently abundant on the shelf of Santa Monica Bay, E. advenus occurred there only rarely prior to the addition of the major sewage outfalls [50]. It is assumed the species’ spatial distribution and abundance reflects its role as a pioneer colonizer of impacted areas, most likely due to the effluent that was released by the local outfalls.

5.4. Macrofaunal Toxicity Measures and Foraminiferal Abundance

Amphipods are the most common organism used to measure sediment toxicity [103,104,105,106] and the 10-day survival protocol is currently the benchmark test [107,108] because amphipods are among the most sensitive infaunal organisms to pollutants [109]. Grandidierella japonica was thought to be a valuable test species in sediment bioassays [5] because it: (1) is widespread in coastal areas [110,111,112]; (2) can tolerate wide ranging temperatures, salinity, and sediment types; (3) is sensitive to numerous contaminants [112,113,114,115]; and (4) has a relatively high tolerance to ammonia [114,115]. However, it is no longer considered for standardized testing because of concerns the species may adapt physiologically to contaminated sediments [93,116,117].

Other amphipod species are now recommended for routine use although the choice is regionally dependent due to local availability issues, lack of a reliable commercial supply, lack of physiological adaptation, or because the test species being used needs to be sensitive to the local pollutants of concern [118]. According to the standards set by the American Society for Testing and Materials [107], the U.S. Environmental Protection Agency [86], and the California Environmental Protection Agency [108], suggested test species in the U.S. are Ampelisca abdita Mills, Eohaustorius estuarius Bosworth, Rhephoxynius abronius (Barnard), and Letocheirus plumulosus Shoemaker. Most of these have been used in toxicity studies in the Southern California Bight (A. abdita [87], E. estuarius [7,119], and R. abronius [120]). However, E. estuarius has become the species of choice for southern California toxicity studies [6,7,119] because it is widely used outside the region and, therefore, allows for better data comparability.

On the East and Gulf coasts of the U.S., use of L. plumulosus is widespread in monitoring [121,122]. In Europe, several different taxa are used for sediment toxicity testing protocols as recommended by ICES, OSPAR, HELCOM, and SETAC-Europe, including the amphipods Corophium orientale Schellenberg, C. arenarium Crawford, C. multisetosum Stock, C. volutator (Pallas), Monocorophium insidiosum (Crawford), Monoporeia affinis (Linström), and Pontoporeia femorata Krøyer (see review [123]). Thus, it would be valuable to expand our investigation of the correlation between the abundance of B. denudata and amphipod survival to determine if it also applies to other amphipod species used in testing in the U.S. and elsewhere internationally, as it does to G. japonica.

Although the sea urchin fertilization test is a common marine bioassay used worldwide for environmental monitoring and regulatory testing of effluents and sediment pore waters [124], molluscan fertilization has been used in the southern California monitoring program since 2008 [6,7,125]. In this test, the normal development of embryos of the mussel Mytilus galloprovincialis Lamarck is measured using a sediment-seawater interface test following the methodology of the U.S. Environmental Protection Agency [126] and Anderson et al. [127], much in the same way sea urchin fertilization was measured. Thus, as with the different species of amphipods, one could investigate whether the abundance of B. denudata and E. advenus correlates with molluscan fertilization tests as well.

Finally, B. denudata is morphologically very similar to Bulimina marginata d’Orbigny which lives on the shelf and slope [128] in many other oceans exclusive of the Pacific [52]. As B. marginata has also been reported as a dominant species at contaminated sites [57,129], it may be true that both of these species respond in similar ways to contaminants. If, as with B. denudata, the abundance of B. marginata positively correlates with amphipod survival, then the possibility exists that the use of these species as sediment toxicity measures may be applied to pollution studies worldwide.

5.5. Use of Foraminifera as Toxicity Measures

The standard macrofaunal toxicity tests of amphipod survival, sea urchin fertilization, and molluscan embryo development success presently used are costly and burdensome laboratory analyses that must be completed shortly after the acquisition of field samples if the results are to be considered valid [130]. Recommendations for sediment holding times before the toxicity tests are initiated range from less than two weeks [131], to less than four weeks [6,7], to less than eight weeks [98,132,133]. One advantage of using foraminifera is that such a time constraint is not an issue. Although using living benthic foraminifera has been shown to be valuable in assessing the ecological status of marine ecosystems [15], samples can be stained for later analysis. In addition, foraminiferal shells are preserved in the sediment which provides a long-term record that may help define reference conditions or gauge the biotic response to polluted conditions and remediation, all of which are not possible with the standard macrofauna toxicity tests presently in use.

When comparing macrofaunal toxicity and foraminiferal abundances (such as those of B. denudata and E. advenus) as proxies of pollution, a few issues need to be considered. The standard macrofaunal toxicity tests use live organisms that interact with toxic compounds that are present in the sediment in a relatively uncomplicated manner (i.e., short-term direct exposure in a laboratory). In contrast, benthic foraminifera reside directly in the sediment for their lifetime. As a result, they may be a more sensitive measure of toxicity because they may be exposed to a higher contaminant load than the organisms being tested in the laboratory. This is especially true if the sediment is fine-grained and contains absorbed heavy metals and organic compounds that are bioavailable [17,18,19]. However, other factors that affect any complex marine ecosystem, such as variations in local temperature, oxygen content, amount of organic matter, prey availability, predation pressures, or changes in regional climate (e.g., El Niño/La Niña-Southern Oscillation or Pacific Decadal Oscillation) exert some control on foraminiferal abundances and distribution as well. For example, agglutinated taxa such as Eggerelloides have been shown to decline in abundance to near extinction in dysoxic environments whereas Bulimina spp. are tolerant of such conditions [134], so long-term records of variations in species abundance may reflect more than just the impact of anthropogenically induced toxicity. Additionally, foraminiferal shells are subject to taphonomic processes such as dissolution and bioturbation after death, resulting in time-averaging that may smooth the faunal trends [72,73] or in some cases, greatly bias the assemblage [135].

Despite the fact that the factors listed above may mask or enhance the impact of trace metals and organic compounds on the organisms, foraminifera are still extremely sensitive to their environment, are ubiquitous and easily sampled, and the preservation of their shells provide a historic record of environmental change due to anthropogenic impact. The findings of this study suggest that the abundance of B. denudata and E. advenus as sediment toxicity measures may have merit. Further investigations are recommended to see if this trend applies to other sites worldwide, other species of macrofauna that are locally available, and potentially other foraminiferal species.

6. Conclusions

The benthic foraminifera B. denudata and E. advenus are valuable taxa in pollution studies because they are tolerant of numerous trace metal and organic contaminants, and as such, are good indicator species. This study suggests that the surface and downcore abundance of B. denudata and E. advenus, as well as amphipod survival and sea urchin fertilization success, can provide a modern assessment, and historical record of change, of sediment quality due to anthropogenic inputs. The abundance of B. denudata was positively correlated with the survival success of amphipods and the abundance of E. advenus was negatively correlated with sea urchin fertilization. As B. denudata (with the related species B. marginata) and E. advenus are present nearly worldwide and often dominate assemblages, the temporal variation of their abundances may become a widespread cost- and time-effective tool for defining and monitoring areas affected by waste discharge. The outcome of this proof-of-concept study suggests that further investigation of the relationship between standard macrofaunal sediment toxicity tests (amphipod survival, sea urchin fertilization, and the development of normal mussel embryos) and the abundance of benthic foraminifera (B. denudata, E. advenus, and others yet to be identified) is warranted.