Multimedia Screening of Contaminants of Emerging Concern (Cecs) in Coastal Urban Watersheds in Southern California (Usa)

SCCWRP #0902 Environmental Toxicology and Chemistry, Vol. 9999, No. 9999, pp. 1–9, 2016 # 2016 SETAC Printed in the USA

Environmental Chemistry

MULTIMEDIA SCREENING OF CONTAMINANTS OF EMERGING CONCERN (CECS) IN COASTAL URBAN WATERSHEDS IN SOUTHERN CALIFORNIA (USA)

KEITH A. MARUYA,*y NATHAN G. DODDER,y ASHMITA SENGUPTA,y DEBORAH J. SMITH,z J. MICHAEL LYONS,z ANN T. HEIL,x and JÖRG E. DREWESk ySouthern California Coastal Water Research Project Authority, Costa Mesa, California, USA zCalifornia Regional Water Quality Control Board, Los Angeles Region, Los Angeles, California, USA xLos Angeles County Sanitation Districts, Whittier, California, USA kUrban Water Systems Engineering, Technical University of Munich, Munich, Germany

(Submitted 11 September 2015; Returned for Revision 28 October 2015; Accepted 23 December 2015)

Abstract: To examine the occurrence and fate of contaminants of emerging concern (CECs) and inform future monitoring of CECs in coastal urban waterways, water, sediment, and fish tissue samples were collected and analyzed for a broad suite of pharmaceuticals and personal care products (PPCPs), commercial and/or household chemicals, current use pesticides, and hormones in an effluent-dominated river and multiple embayments in southern California (USA). In the Santa Clara River, which receives treated wastewater from several facilities, aqueous phase CECs were detectable at stations nearest discharges from municipal wastewater treatment plants but were attenuated downstream. Sucralose and the chlorinated phosphate flame retardants tris(1-chloro-2-propyl) phosphate (TCPP), tris(1,3-dichloro-2-propyl) phosphate (TDCPP), and tris(2-chloroethyl) phosphate (TCEP) were most abundant in water, with maximum concentrations of 35 mg/L, 3.3 mg/L, 1.4 mg/L, and 0.81 mg/L, respectively. Triclocarban, an antimicrobial agent in use for decades, was more prevalent in water than triclosan or nonylphenol. Maximum concentrations of bifenthrin, permethrin, polybrominated diphenyl ethers (PBDEs), and degradates of fipronil exceeded CEC-specific monitoring trigger levels recently established for freshwater and estuarine sediments by factors of 10 to 1000, respectively. Maximum fish tissue concentrations of PBDEs varied widely (370 ng/g and 7.0 ng/g for the Santa Clara River and coastal embayments, respectively), with most species exhibiting concentrations at the lower end of this range. These results suggest that continued monitoring of pyrethroids, PBDEs, and degradates of fipronil in sediment is warranted in these systems. In contrast, aqueous pharmaceutical concentrations in the Santa Clara River were not close to exceeding current monitoring trigger levels, suggesting a lower priority for targeted monitoring in this medium. Environ Toxicol Chem 2016;9999:1–9. # 2016 SETAC

Keywords: Contaminants of emerging concern Pharmaceuticals Pyrethroids Polybrominated diphenyl ethers Water quality

INTRODUCTION Monitoring of CECs is essential in highly populated coastal Contaminants of emerging concern (CECs) are a diverse urban regions where the input and loading of CECs via group of largely unmonitored and unregulated chemicals [1] controlled and uncontrolled discharges and/or pathways are that have been shown to occur in municipal wastewater maximized [16]. In an inaugural study in southern California discharges, stormwater runoff, and ultimately in ambient (USA), home to more than 20 million people, CECs were fl receiving waters [2–4]. Some of the more potent CECs (e.g., analyzed in water collected from 2 ef uent-dominated rivers fl pesticides and hormones) exhibit toxic or endocrine-disrupting during low- ow conditions [17]. Maximum in-stream concen- potential to aquatic life at relatively low concentrations [5,6], trations of pyrethroid pesticides exceeded monitoring trigger fi whereas others may persist [7,8], accumulate in tissues [9,10], levels established by a panel of scienti c experts who employed and ultimately biomagnify in higher trophic level species, a risk-based framework driven by published maximum including birds, marine mammals, and humans [11–13]. For measured concentrations and aquatic toxicity thresholds (e.g., many CECs, insufficient information is available to characterize predicted no-effect concentrations, no-observed-effect concen- their occurrence, fate, and transport in receiving water systems, trations, and lowest-observed-effect concentrations) to identify and also to determine scientifically credible thresholds above CECs for monitoring in these receiving waters [17]. In contrast, which ecologically relevant effects would be expected [14]. maximum concentrations of most pharmaceuticals and personal Although it is impossible to characterize the combined risk for care products (PPCPs) commonly detected in treated wastewa- all known CECs using the traditional chemical-by-chemical ter (e.g., acetaminophen, N,N,diethyl-meta-toluamide [DEET] fi paradigm, prioritizing and monitoring the occurrence of known, and gem brozil) were less than 10% of monitoring trigger high-priority CECs is necessary to inform managers and levels established by the aforementioned expert panel. Little scientists of the potential for exposure to CECs and ultimately attenuation of CECs was observed downstream of wastewater to protect the integrity of aquatic and human health [15]. treatment facilities discharging to surface waters (water reclamation plants), partly because of the short hydraulic residence times and low biological activity in these relatively This article includes online-only Supplemental Data. short, highly modified systems [17]. However, the above study * Address correspondence to [email protected] Published online 8 January 2016 in Wiley Online Library focused on CECs in the water column and was limited in scope (wileyonlinelibrary.com). to waterways with minimal or intermittent natural bottom DOI: 10.1002/etc.3348 sediments [17].

1 2 Environ Toxicol Chem 9999, 2016 K.A. Maruya et al.

In the present study, we sought to address: 1) the occurrence homogenized using a solvent-rinsed stainless steel blender and and fate of aqueous phase CECs discharged from water composited into precleaned glass jars. Within 48 h of collection, reclamation plants into an effluent-dominated river with a samples were stored at –20 8C, and later shipped to the Colorado natural bottom structure during low-flow conditions; 2) the School of Mines Advanced Water Technology Center (Golden, occurrence of CECs in river and coastal embayment sediments; CO, USA) on ice and with chain of custody forms for CEC and 3) the occurrence of CECs in fish tissue of various species analysis. collected in river and coastal embayments. Water, sediment, and fish samples were collected at multiple stations in an effluent- Coastal embayments dominated river and in multiple coastal embayments along the Twenty-two surficial sediments were collected in southern southern California coast and analyzed for a broad suite of California embayments stretching from Mugu Lagoon (near CECs. Measured concentrations were subsequently compared Port Hueneme) to San Diego Bay between July and with current monitoring trigger levels established to protect the October 2013 (Figure 1B and Supplemental Data, Table S1). health of aquatic life in these receiving water systems. These embayments represented a broad spectrum of adjacent land use and human impact, ranging from restricted access areas MATERIALS AND METHODS (e.g., the upper Newport Bay Ecological Reserve), moderately developed coastal lagoons (e.g., Agua Hedionda, Batiquitos, Santa Clara River watershed Los Penasquitos, San Dieguito, and San Elijo), densely The Santa Clara River watershed encompasses an area of populated marinas (e.g., Marina del Rey), and urban flood 4140 km2 and extends 134 km from the headwaters in the San control waterways (e.g., Ballona Creek, Dominguez Channel, Gabriel Mountains north of Los Angeles (CA, USA) before and the Los Angeles River estuary). Triplicate grab samples entering the Pacific Ocean near the town of Ventura were collected at each station using a petite Ponar grab or (Figure 1A). The Santa Clara River has been subject to flow precleaned shovel, and composited into a single sample using diversions in the middle to lower reaches for groundwater trace organics clean techniques. Specimens of 5 fish species recharge for water storage and crop irrigation, but, unlike other targeted by sport fishermen (barred sandbass, cabezon, kelp heavily modified coastal rivers in this region [17], has bass, rockfish, and white croaker) were collected in the Los maintained a relatively natural riparian corridor and a riverbed Angeles–Long Beach Harbor complex in August 2013 and that is largely natural substrate. Major tributaries of the Santa again in May 2014 (Figure 1B and Supplemental Data, Clara River include the Castaic, Piru, and Sespe Creeks. The Table S2). river is home for several threatened and endangered species with sections protected from human access and recreational use to Sample processing and analytical chemistry preserve habitat. In the upper watershed, the Saugus and Target compounds were isolated from unfiltered 1-L water Valencia water reclamation plants discharge a combined samples using solid-phase extraction and subsequent elution 75 000 m3/d of tertiary treated effluent into the Santa Clara into methanol. Sediment and fish tissue samples were freeze- River. During dry weather, low-flow conditions, more than 90% dried, and aliquots were extracted using accelerated solvent of in-stream flow in the upper watershed is supplied by water extraction. Smaller fish (e.g., silversides, cichlids) were reclamation plant discharge in most sections of the river. deheaded and composited in triplicate using a solvent-rinsed Surface flow is intermittent with dry gaps between water stainless steel scalpel. Larger fish (carp, marine fish) were fileted reclamation plants. On re-emerging downstream of the Piru (skin-off) and homogenized individually as described above in Creek confluence, the river flows west before terminating at the Santa Clara River watershed. The CECs targeted for analysis coast, where the Ventura water reclamation plant discharges included those recommended for investigative monitoring in approximately 35 000 m3/d tertiary effluent into holding ponds California receiving waters by a scientific advisory panel [14]. prior to release into the Santa Clara River estuary. Target analytes for water included PPCPs (e.g., caffeine, Water was collected from 10 stations in the Santa Clara River carbamezepine, DEET, diazepam, diclofenac, dilantin, fluoxe- watershed—6 stations along the freshwater main stem (SCR-1, tine, gemfibrozil, ibuprofen, iopromide, meprobamate, sucra- SCR-2, SCR-3, SCR-4, SCR-5, and SCR-6), 3 stations in the lose, and sulfamethoxazole), natural and synthetic hormones estuary (SCE-1, SCE-2, and SCE-3), and a reference station on (e.g., 17b-estradiol, estrone, ethinylestradiol, progesterone), Piru Creek (SCR-REF)—during 2 low-flow events (27 July herbicides (e.g., atrazine, diuron, simazine), and commercial and 15 October 2013; Figure 1A). Bed sediment was collected and industrial chemicals such as bisphenol A (BPA), during a single event (4 September 2013). One-liter unfiltered 4-nonylphenol, and the chlorinated phosphate flame retardants water samples were collected using a stainless steel bucket (CPFRs) tris(2-chloroethyl) phosphate (TCEP), tris(1-chloro-2- and stored in amber bottles with preservative. Sampling was propyl) phosphate (TCPP), and tris(1,3-dichloro-2-propyl) performed during daylight hours, with stations in the upper phosphate (TDCPP; Supplemental Data, Table S3). For Santa Clara River watershed sampled during the morning and sediment and fish tissue samples, current-use pesticides stations in the lower watershed sampled in the afternoon. (8 pyrethroids along with fipronil and its degradates) and a Surficial sediment samples (500 g) were collected using a suite of 14 di- to octabromimated diphenyl ethers (PBDEs) stainless steel hand shovel and stored in glass jars. All were analyzed (Supplemental Data, Table S4). Sediment equipment and storage vessels were precleaned. Convict (also samples from the Santa Clara River watershed and 11 of the called zebra) cichlids were collected at SCR-1, and common coastal embayment sites were also analyzed for BPA, carp (Cyrpinus carpio) and silversides (Atherinopsis spp.) were diclofenac, triclosan, triclocarban, 4-nonylphenol, and CPFRs. collected at SCE-3 on 4 and 5 September 2013. Fish samples Sample extracts were analyzed using isotope-dilution liquid were wrapped in solvent-rinsed aluminum foil. All samples chromatrography–tandem mass spectrometry (LC–MS/MS) were kept in the dark on ice in the field, and transferred to with atmosphere pressure chemical ionization (hormones) and storage within 48 h of collection. Water was stored at 5 8C electrospray ionization negative and positive (PPCPs, CPFRs) and sediment and fish at –20 8C until analysis. Tissue was multiple reaction monitoring modes using an ABSCIEX 3200 CECs in sediment and fish tissue Environ Toxicol Chem 9999, 2016 3

Figure 1. (A) Sampling stations for the Santa Clara River (SCR) watershed. River water and sediment samples were collected during low-ﬂow conditions (July through October 2013). (B) Sampling locations in southern California estuaries. Sediment and ﬁsh tissue samples were collected June through October 2013. Inset shows the Los Angeles/Long Beach Harbor complex. Shading indicates urban areas with a population density > 6500/km2. WWTP ¼ wastewater treatment plant; SCE ¼ Santa Clara River estuary; WRP ¼ water reclamation plant.

QTRAP LC–MS/MS [18,19]. Pyrethroids, fipronil, and its blanks containing preservatives were processed and analyzed in degradates and PBDE congeners were analyzed by gas parallel with each batch of samples. No target analyte was chromatrography–negative chemical ionization (GC-NCI)/MS detected in procedural and/or field blanks for aqueous samples. with an Agilent 7890 GC/5975 quadrupole mass selective In sediment blanks, nonylphenol and the 3 CFPRs were detector [20]. detectable, resulting in adjustment of reporting limits to 3 times Water samples were preserved with ascorbic acid and the highest reported blank concentration. Mean recoveries of the sodium azide prior to extraction within 7 d of collection to surrogates polychlorinated biphenyl (PCB)-208 and brominated minimize target analyte loss. Procedural (laboratory) and field diphenyl ether (BDE)-172 in sediment and fish tissue extracts 4 Environ Toxicol Chem 9999, 2016 K.A. Maruya et al. were 88 3.5% and 85 2.9%, respectively, and 80 18% and reach, where during dry weather and in the absence of 84 15%, respectively. Recovery of TCEP-d12 was variable dewatering activity, surface flow from the Saugus water and less than 10% for 7 of 23 sediment extracts, resulting in reclamation plant completely submerges into the sandy riverbed higher uncertainty and likely underestimated concentrations for downstream of SCR-1, only to reappear just upstream of SCR-2. the CPFRs. Analysis of fish tissue triplicates resulted in analyte- The much reduced CEC concentrations, coupled with higher specific relative standard deviations ranging from 1.5% to 41%. sulfate levels, strongly suggest the surface flow at SCR-2 to be For a minimum signal-to-noise ratio of 10, reporting limits for largely influenced by groundwater discharges. Although surface water samples ranged from 1.0 ng/L (multiple analytes) to 250 flow at SCR-2 was much reduced compared with SCR-1 (4-nonylphenol) ng/L (Supplemental Data, Table S3); reporting (4000 m3/d vs 20 000 m3/d), it could not be ascertained limits for sediment and fish tissue ranged from 0.06 ng/g whether CECs in the Saugus water reclamation plant discharge (triclocarban) to 60 ng/g (4-nonylphenol; Supplemental Data, were simply diluted by groundwater before re-emerging, or Table S4). whether a more complex interaction with the subsurface environment had occurred. Monitoring trigger quotients The sharp increase in CEC concentrations as well as flow (at Measured concentrations of fipronil and its 3 degradates, 55 000 m3/d) at SCR-3, attributable to the discharge from the 8 pyrethroids, and 8 PBDE congeners were reported individu- Valencia water reclamation plant, was followed by a steady ally and as their respective sums in units of ng/g dry (sediment) decline to nondetectable levels at stations SCR-5 and SCR-6, or wet weight (fish tissue). Maximum concentrations of CECs 15 km and 43 km downstream of the Valencia water reclamation were compared with their corresponding monitoring trigger plant, respectively. Sulfate concentration also increased levels, derived from toxicity thresholds for ecological receptors, downstream of the Valencia water reclamation plant, albeit to compute monitoring trigger quotients at a slower pace than was observed between SCR-1 and SCR-2 (Figures 2 and 3). Although historical data (from 1999–2000) ¼ ð Þ MTQ Cmax MTL 1 indicated a net loss of surface flow between SCR-3 and SCR-5 [21], loss of surface flow alone does not explain the where MTQ is the monitoring trigger quotient, Cmax is the decreasing concentration profiles. Interestingly, CECs consid- maximum measured concentration, and MTL is the monitoring ered to be resistant to transformation and thus serve as trigger level. Values of monitoring trigger quotients that conservative tracers in riverine systems (e.g., carbamazepine, exceeded and/or approached unity were used to screen CECs sucralose) [8,17] showed decreases in aqueous concentration of for future investigative monitoring [14]. approximately 30% between SCR-3 and SCR-5 (Supplemental Data, Tables S5 and S6). Assuming this loss was because RESULTS AND DISCUSSION of dilution by groundwater, it can be surmised that CECs exhibiting a greater decrease (e.g., 60–90% for DEET, Santa Clara River watershed gemfibrozil, iopromide, meprobamate, the CPFRs, and triclo- Several CECs (carbamazepine, DEET, dilantin, diuron, carban) were subject to additional attenuation mechanisms gemfibrozil, meprobamate, sucralose, sulfamethoxazole, TCEP, (e.g., sorption to riverbed sediments, biotransformation, and TCPP and TDCPP, and triclocarban) were widely detectable in photodegradation). water samples from the Santa Clara River watershed (Supple- In sediments, BPA, diclofenac, and 4-nonylphenol were not mental Data, Tables S5 and S6). In contrast, the target hormones detected. Triclosan and triclocarban had the highest detectable (17a-ethinyl estradiol, 17ß-estradiol, estrone, and testosterone), concentrations of 110 ng/g and 183 ng/g dry weight, respec- atrazine, BPA, and 4-nonylphenol were not detected in water. tively (Supplemental Data, Table S7). The CPFRs were detected No target analytes were detectable in the samples from the at concentrations less than 60 ng/g dry weight. However, the low reference station on Piru Creek (SCR-REF), whose source is a recovery of TCEP-d12 in several samples resulted in higher recreational reservoir with a largely undeveloped watershed. measurement uncertainty for CPFRs; thus, caution should be Aqueous concentrations were highest for stations SCR-1 and used when interpreting data for these CECs. Fipronil degradates SCR-3, just downstream from the Saugus and Valencia water were detectable in most samples, but concentrations were reclamation plants, respectively, and in the estuary near the <1 ng/g dry weight at all locations except SCE-1 (Supplemental discharge from the Ventura water reclamation plant (SCE-1, Data, Table S8). Pyrethroids were infrequently detected, with SCE-2, SCE-3; Figure 1A). The highest water concentrations bifenthrin being most commonly found, at concentrations up were measured for sucralose (35 000 ng/L); followed by TCPP to 5 ng/g dry weight (Supplemental Data, Table S8). The (3300 ng/L), TDCPP (1400 ng/L), TCEP (660 ng/L), and maximum concentration when all pesticide analytes were meprobamate (620 ng/L). summed was 16 ng/g dry weight, which was observed at station Concentration profiles of target CECs in river water showed SCE-1. The second highest concentration was at SCR-4, which a rapid decline after discharge from the Saugus water is well below the Saugus and Valencia water reclamation plants reclamation plant in the uppermost reach of the Santa Clara and well upstream of the Ventura water reclamation plant, River. For example, aqueous concentrations of the CPFRs indicating a significant source of pyrethroids other than the (Figure 2) triclosan and triclocarban (Figure 3) decreased water reclamation plants. The PBDEs were detected at 4 of several fold between SCR-1 and SCR-2, a distance of 5 km. the sites (SCR-1, SCR-3, SCE-1, and SCE-2), at summed Water quality monitoring data courtesy of the Los Angeles concentrations of less than 7 ng/g dry weight (Supplemental County Sanitation Districts showed a sharp increase in sulfate at Data, Table S9). The PBDEs detected were limited to BDE-47, SCR-2 (Figures 2 and 3), indicating a large contribution of BDE-49, BDE-99, BDE-100, BDE-153, and BDE-154, with relatively CEC-free surface flow at this location, which may be concentrations dominated by BDE-47 and BDE-99. because of groundwater discharge from the shallow aquifer In contrast to the profiles observed in water, CEC as well as construction dewatering projects or other local concentrations in river sediment were highest at stations further sources [21]. This is consistent with the ephemeral nature of this downstream from the water reclamation plants (e.g., at SCR-4; CECs in sediment and fish tissue Environ Toxicol Chem 9999, 2016 5

Figure 2. Longitudinal concentration proﬁles of chlorinated phosphate ﬂame retardants for the Santa Clara River. (A) Water column. (B) Sediment. Aqueous sulfate concentration as a marker for groundwater interaction is plotted on the secondary axis. The sulfate measurement for station SCR-5 was for a single date in September 2013. TCPP ¼ tris(1-chloro-2-propyl) phosphate; TCEP ¼ tris(2-chloroethyl) phosphate; TDCPP ¼ tris(1,3-dichloro-2-propyl) phosphate; WRP ¼ water reclamation plant; WWTP ¼ wastewater treatment plant.

Figures 2 and 3). These profiles are evidence that sorption may analytes. These sediment profiles showed that sorption may have contributed to the attenuation of aqueous concentrations contribute to the loss of CECs from the water column, and that between SCR-3 and SCR-5. Including the estuary, CEC sediments enriched in organic carbon accumulated higher concentrations were highest at SCR-4, SCE-1, and SCE-2 concentrations of the more hydrophobic CECs. and lowest for SCR-5 and SCR-6. As is well established in the In fish tissue, the summed concentrations of pesticides were literature, total organic carbon (TOC) content (fOC) appeared to 14 ng/g wet weight (Supplemental Data, Table S10). Pesticide be a factor, with SCR-4 (2.6%) and SCE-1 (2.1%) accumulating concentrations were dominated by fipronil sulfone; in contrast, more CECs (in most cases) than sediments with lower fOC pyrethroids were largely nondetectable. Tissue concentrations (Supplemental Data, Table S8). Sediment concentrations for of fipronil and its degradates varied by 2 orders of magnitude CPFRs were relatively low compared with their water between the silversides and the carp (Table 1). Tissue concentrations (Figure 2), suggesting a relatively low potential concentrations of PBDEs wereP highly variable, with the cichlids for accumulating in sediment. On the other hand, sediment collected at SCR-1 exhibiting PBDE as high as 370 ng/g wet concentrations of triclosan and triclocarban reflected a higher weight (Supplemental Data, Table S11). In all samples, the potential for partitioning from water into bed sediments congener profile was dominated by PBDE-47, PBDE-100, (Figure 3). Interestingly, the occurrence of triclocarban in PBDE-28, and PBDE-99. These tissue concentrations, relative SCE (estuarine) sediments was lower than in Santa Clara River to the low concentrations in sediment, confirm the potential of (freshwater) sediments (Supplemental Data, Table S7), suggest- PBDEs to bioaccumulate in fish tissues. The high lipid content ing that salinity and/or pH may influence its fate. The trend for (18–27%; Supplemental Data, Table S10), proximity to the triclosan was the opposite of that for triclocarban, with higher Saugus water reclamation plant, and possibly the low sediment concentrations in the SCE compared with the Santa Clara River fOC may have influenced the elevated PBDE concentrations in (Figure 3 and Supplemental Data, Table S7). The fate of the cichlid tissues. Although these fish are a nonnative, tropical analytes that were barely or not detectable in both water and aquarium species, they can fortuitously be used to indicate the sediments could not be assessed, although it should be noted that degree of bioaccumulation at SCR-1, particularly because their reporting limits for 4-nonylphenol, for example, were 1 order of mobility is constrained by shallow flow and the nearby magnitude to 2 orders of magnitude higher than for most other downstream dry gap. However, because of their high relative 6 Environ Toxicol Chem 9999, 2016 K.A. Maruya et al.

Figure 3. Longitudinal concentration profiles of triclosan (TCS) and triclocarban (TCB) for the Santa Clara River. (A) Water column. (B) Sediment. Aqueous sulfate concentration as a marker for groundwater interaction is plotted on the secondary axis. The sulfate measurement for station SCR-5 was for a single date in September 2013. WRP ¼ water reclamation plant; WWTP ¼ wastewater treatment plant. lipid concentration relative to native fish, bioaccumulation on a and/or water. Additional studies on the bioaccumulation wet tissue basis for the cichids may not be applicable to other potential of this degradate are warranted. fish. Lipid normalization removed much, but not all, of the variability in comparing PBDE concentrations in tissues (data Coastal embayments not shown). The difference in fipronil (and degradates) Sediment. Bisphenol A, triclocarban, TCEP, TCPP, and concentration for silversides and carp may be partially TDCPP were detected in most of the 11 samples analyzed for explained by an order of magnitude difference in lipids, these constituents, with 4-nonylphenol detectable in approxi- particularly because the carp were analyzed as skin off filets. mately half of these samples, and triclosan and diclofenac Pyrethroids, which were largely below nondetect levels, did not largely nondetectable (Supplemental Data, Table S12). Con- accumulate in fish tissue. The predominance of fipronil sulfone centrations for 6 of these 8 analytes were highest in the sample in fish tissue was not expected (Supplemental Data, Table S10), from the Los Angeles River. In contrast to the Santa Clara River, particularly in comparison with its concentrations in sediment pyrethroids were frequently detected in embayment sediments,

Table 1. Fish tissue concentrations (ng/g wet wt) of ﬁpronil and its 3 degradates (SFIP), pyrethroids (SPYR), and polybrominated diphenyl ethers (SBDE)

Species Station Tissue type Lipid (%) SFIP SPYR SBDE

Alaskan halibuta Muscle 2.4 <0.01 <0.04 0.57 Convict cichlidb SCR-1 Whole body 22 3.7 5.6 0.92 0.24 0.17 300 62 Silversideb SCE-1 Whole body 12 0.82 13 0.94 <0.04 6.6 0.57 Carpc SCE-1 Muscle 2.1 0.62 0.22 0.10 <0.04 6.0 4.0 Sandbassb Outer LAH Muscle 2.7 1.1 0.0055 0.0038 <0.04 0.76 0.27 White croakerc Inner LAH Muscle 5.4 1.1 0.10 0.058 <0.04 2.9 2.2 aReference material. bn ¼ 3. cn ¼ 5. cSCR – Santa Clara River; SCE ¼ Santa Clara Estuary; LAH ¼ Los Angeles Harbor. CECs in sediment and fish tissue Environ Toxicol Chem 9999, 2016 7 with bifenthrin and permethrin typically found at the highest contents for the collected specimens, ranging between 1.5% and concentrations. Elevated concentrations of pyrethroids were 7.6%, were considerably lower than that measured in convict found in selected Los Angeles regional watersheds, for cichlids tissues taken from the Santa Clara River. Moreover, example, Los Angeles River, Ballona Creek, Marina del Rey, species such as white croaker and kelp bass can roam over a and Dominguez Channel (Supplemental Data, Table S13), wide range within as well as outside the Los Angeles–Long which drain highly urbanized and/or industrialized landscapes. Beach Harbor [22], thus reducing their exposure to relatively Except for Upper Newport Bay, pyrethroid concentrations were small or patchy areas of elevated sediment contamination. relatively low in comparison in coastal lagoons and river mouths in northern San Diego County, which lies southeast of Los Implications for environmental quality and future monitoring Angeles. Fipronil was rarely detected, and its degradates were Maximum aqueous concentrations of TCPP, TDCPP, and generally detectable but at relatively low concentrations TCEP for the Santa Clara River watershed were similar to those (< 5 ng/g dry wt). Total PBDEs, also analyzed at all sites, reported previously in 2 different effluent-dominated waterways ranged up to 45 ng/g dry weight (again for the Los Angeles in southern California [17], indicating similar discharge loading River sample), with BDE-47, BDE-99, BDE-153, and BDE-154 (i.e., loading from the water reclamation plants) for all 3 as the most abundant congeners (Supplemental Data, effluent-dominated watersheds in the present study region. In Table S14). Similar to the other CEC groups, the highest contrast, the maximum Santa Clara River concentration for PBDE concentrations were observed in embayments and river sucralose was 10-fold higher than was found in other regional mouths within the Los Angeles region. As was observed for watersheds, suggesting that this compound had a correspond- Santa Clara River sediments, embayment samples enriched in ingly higher loading into the upper Santa Clara River water TOC contained the highest levels of CECs (e.g., Los Angeles reclamation plants relative to the other watersheds [17]. River, fOC ¼ 7.4%; Dominguez Channel, fOC ¼ 3.3%; Supple- Maximum sediment concentrations of triclocarban in the mental Data, Table S13). However, 1 exception was the present study were several fold higher than was reported in sediment from Ballona Creek (fOC ¼ 0.5%), which contained Puget Sound (WA, USA; maximum of 17 ng/g dry wt) [23], but CECs at higher concentrations relative to most other stations. consistent with surface sediment concentrations reported in the Target CEC concentrations were in general higher in coastal vicinity of the marine outfall for a large municipal wastewater embayment sediments compared with those from the Santa treatment plant discharging to the coastal ocean near Los Clara River watershed. Angeles [24]. For triclosan, coastal sediment concentrations Fish tissue. Measured concentrations of pesticides and were largely below detection limits, similar to the occurrence PBDEs for fish tissues collected in and around Los Angeles– pattern of this CEC reported for Narragansett Bay (RI, Long Beach Harbor (Supplemental Data, Tables S15–S18) were USA) [25]. uniformly less than 10 ng/g wet weight (each analyte) and were The maximum monitoring trigger quotients for aqueous lower than maximum tissue concentrations measured in the phase CECs targeted in the present study were uniformly less Santa Clara River (Table 1). Fipronil desulfinyl and fipronil than unity (Table 2). However, the data for 17ß-estradiol and sulfone were the most commonly detected pesticides (parent estrone were inconclusive, because reporting limits were not fipronil and fipronil sulfide were infrequently detected); low enough to estimate monitoring trigger quotients < 1. The moreover, the sum of pesticide concentrations was uniformly monitoring trigger quotients for the remaining target CECs in low (0.21 ng/g wet wt). No pyrethroids were detected. Total the Santa Clara River were consistently lower than was PBDE concentrations were highest among the target analytes at calculated for 2 other effluent-dominated coastal rivers in the 7.0 ng/g and 5.3 ng/g wet weight for white croaker collected region [17]. The CEC concentrations downstream of water near Pier J inside Los Angeles–Long Beach Harbor, and kelp reclamation plants discharging to the upper Santa Clara River bass collected outside the harbor breakwater, respectively. Lipid watershed decreased more rapidly than those measured in other

Table 2. Maximum aqueous concentrations (ng/L) for target contaminants of emerging concern (CECs) in the Santa Clara River compared with other efﬂuent-dominated waterwaysa

Las Vegas Los Angeles San Gabriel Santa Clara MTQf Compound Washb Riverc Riverc Riverd MTLe Santa Clara Riverd

Bisphenol A <1.0 691 657 <25 60 <1 Diclofenac 18 77.0 124 20.1 100 0.20 17b-estradiol n/a <1.25 <1.25 <52 <2.5 Estrone n/a <2.5 <2.5 <56 <1 Ibuprofen n/a 38.3 40.5 <25 100 <0.25 Galaxolide n/a 2619 2753 n/a 700 n/a Permethrin n/a <0.17 1.72 n/a 1.0 n/a Fipronil n/a 28.7 22.2 n/a 3.3g n/a Carbamazepine 203 231 330 285 25 000 0.011 Gemfibrozil 170 179 324 118 78 000 0.0015 Triclosan 12 10.7 26.0 38.4 250 0.15 aThese concentrations were used to compute monitoring trigger quotients (MTQs) for the Santa Clara River. bFrom Vanderford and Snyder [19]. cFrom Sengupta et al. [17]. dResults of present study. eMonitoring trigger level [14]. fMaximum measured concentration/MTL. gFrom Weston and Lydy [26]. MTL ¼ monitoring trigger level; MTQ ¼ monitoring trigger quotient; n/a ¼ data not available. 8 Environ Toxicol Chem 9999, 2016 K.A. Maruya et al. effluent-dominated rivers in the region, namely, the Los convict chichlids collected just downstream of the Saugus water Angeles and San Gabriel Rivers [17]. This was likely the result reclamation plant (near SCR-1) was >1 (Table 3). Their very of the ephemeral nature of the Santa Clara River below the high lipids relative to other fish likely played a role in their Saugus water reclamation plant, combined with the attenuation ability to bioaccumulate PBDEs to a much higher degree than by biotransformation and sorption below the Valencia water the other species targeted in the present study. The monitoring reclamation plant, and potential interactions with the subsurface trigger quotients for PBDE-47 and PBDE-99 in species other environment. The interaction of effluent-dominated surface than the cichlids were <1, including those for fish tissue flow with shallow groundwater in the upper Santa Clara samples collected from Los Angeles–Long Beach Harbor River watershed would also suggest the need for more (Table 3). To better understand the potential impact of elevated investigation regarding the fate and transport of CECs in the PBDEs in the convict cichlids, additional exploratory investi- shallow aquifer. gation into the occurrence and/or consumption of these fish by In contrast, monitoring trigger quotients for sediment- wildlife in the upper watershed should be pursued. associated bifenthrin, permethrin, and PBDEs (represented by Until very recently, aquatic toxicity thresholds for CPFRs congeners 47 and 99) were >>1, with the exception of have been reported as typically in the parts per million or higher permethrin for the Santa Clara River sediments (Table 3). range, and thus occurrence measured in our previous screening Moreover, sediment concentrations in coastal embayments for study [17] did not approach such thresholds. However, newly these analytes far exceeded those in Santa Clara River published information suggests that TDCPP in particular may sediments, indicating a higher priority for continued monitoring exert toxicity (multiple endpoints including larval fish survival, in coastal habitats, particularly those draining highly urbanized gene expression, and egg production) in the parts per billion watersheds (e.g., the Los Angeles River). The maximum range [29]. Using this information, the gap between occurrence, concentration of fipronil and its degradates in Santa Clara River from Sengupta et al. [17] and the present study, and these newly sediments also exceeded the freshwater monitoring trigger published toxicity values is reduced from several thousand to a level [26]. Assuming equal potencies of the degradates relative factor of 10 or less. Moreover, little is known about time trends to parent fipronil and thus summing their concentrations is in CPFR usage, removal by wastewater treatment processes, and justified for screening level assessment based on the relatively ultimately discharge rates into and occurrence in receiving narrow range of sediment median lethal concentrations (LC50s) waters [14]. Thus, it would be prudent to monitor selected available (0.12–0.16 mg/g org C) [27]. Although maximum CPFRs in aqueous samples to determine if exposure concen- concentrations of fipronil were higher in coastal embayment trations are stable into the future. sediments, monitoring trigger quotients did not exceed unity, In summary, future monitoring of pyrethroids and PBDEs in because published toxicity thresholds for estuarine/marine coastal embayment sediments would be the highest priority in species were up to 50 times higher than for freshwater species. this region. Pyrethroids and fipronil degradates in Santa Clara Although other investigators have identified PBDEs as a River sediments would be the next highest priority, followed by monitoring priority for sediments and fish tissue [28], compar- PBDEs in fish tissue. Monitoring of CPFRs such as TDCPP in ing PBDEs in fish tissue with the monitoring trigger level aqueous samples is also recommended. Based on the findings established for California receiving waters [14] is not of the present study, continued measurement of other CECs straightforward. The published monitoring trigger level for (e.g., PPCPs, hormones, and synthetic phenolics like BPA and PBDEs is based on bioaccumulation in bird eggs, and thus 4-nonylphenol) in surface water initially recommended for cannot be easily or directly translated into an appropriate investigative monitoring would be of lower priority; however, monitoring trigger level for fish tissue. Assuming that their loading and fate in the subsurface environment of the biomagnification of PBDE-47 and PBDE-99 occurs, and that upper Santa Clara River watershed could become a topic of bird egg residues can be expected to be higher than fish tissue, consideration. Because CEC research is an active and evolving comparing fish tissue PBDEs with the published threshold of field, these recommendations may change in light of future 28.9 ng/g would constitute a conservative approach. In this advancements in risk assessment that may lead to revised scenario, the monitoring trigger quotient for PBDE-47 for monitoring trigger quotients.

Table 3. Maximum sediment and ﬁsh tissue concentrations (ng/g dry and wet wt, respectively) and monitoring trigger quotients (MTQs) for selected contaminants of emerging concern (CECs) in the Santa Clara River (SCR) and southern California coastal embayments

Other SCR Other embayment Monitoring Compound Matrix watersheda embaymenta MTLb SCR MTQc MTQc priority

Bifenthrin Sediment (embayment) 4.8 54 0.052 92 1000 High PBDE-47 Sediment (embayment) 0.70 11 0.030 23 370 High PBDE- 99 Sediment (embayment) 0.63 19 0.030 21 630 High PBDE-47 Tissue 11/230d 3.8 28.9 0.38/8.0d 0.13 Medium PBDE- 99 Tissue 1.0/26d 0.23 28.9 0.035/0.90d 0.0080 Low Permethrin Sediment (embayment) <1.4 105 0.073 <2 1400 High Fipronile Sediment (freshwater) 1.2 n/a 0.090 13 n/a High Fipronile Sediment (embayment) 6.8 3.7 6.5 1.0 0.57 Medium aResults of present study. bMonitoring trigger level for coastal embayments [14]. cMonitoring trigger quotient ¼ maximum measured concentration/MTL. dIncluding convict cichlids collected at SCR-1. eConcentrations represent the sum of the parent compound and its 3 measured degradates (desulﬁnyl, sulfone and sulﬁde). PBDE ¼ polybrominated diphenyl ether; MTL ¼ monitoring trigger level; MTQ ¼ monitoring trigger quotient; n/a ¼ data not available. CECs in sediment and fish tissue Environ Toxicol Chem 9999, 2016 9

Supplemental Data—The Supplemental Data are available on the Wiley 14. Maruya KA, Schlenk D, Anderson PD, Drewes JE, Denslow ND, Online Library at DOI: 10.1002/etc.3348. Olivieri AW, Scott GI, Snyder SA. 2014. An adaptive, comprehensive monitoring strategy for chemicals of emerging concern (CECs) in Acknowledgment—The present study was supported by the California Water California’s aquatic ecosystems. Integr Environ Assess Manag Resources Control Board (Agreement 10-085-140). The authors thank W. 10:69–77. Lao, D. Tsukada, D. Diehl, D. Heil, R. Yoshida, M. Bao, M. Gasca, and the 15. Scott GI, Fulton MH, Weisberg SB, Maruya KA, Lauenstein G. field sampling crews from the Los Angeles County Sanitation Districts. 2012. Contaminants of concern in the marine environment: The need for new monitoring and assessment strategies. J Mar Biol Oceanogr Data availability—All data can be accessed in the primary article and 1:1–3. supplemental information. For more information, please contact the 16. Maruya KA, Dodder NG, Weisberg SB, Gregorio D, Bishop JS, corresponding author at [email protected] Klosterhaus S, Alvarez DA, Furlong ET, Bricker S, Kimbrough KL, Lauenstein GG. 2014. The Mussel Watch California pilot study on REFERENCES contaminants of emerging concern (CECs): Synthesis and next steps. Mar Pollut Bull 81:355–363. 1. Diamond JM, Latimer HA, Munkittrick KR, Thornton KW, Bartell 17. Sengupta A, Lyons JM, Smith DJ, Heil A, Drewes JE, Snyder SA, SM, Kidd KA. 2011. Prioritizing contaminants of emerging concern Maruya KA. 2014. The occurrence and fate of chemicals of emerging for ecological screening assessments. Environ Toxicol Chem concern (CECs) in coastal urban rivers receiving discharge of treated 30:2385–2394. municipal wastewater effluent. Environ Toxicol Chem 33:350–358. 2. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber 18. Vanderford BJ, Pearson RA, Rexin DJ, Snyder SA. 2003. Analysis of LB, Buxton HT. 2002. Pharmaceuticals, hormones, and other organic endocrine disruptors, pharmaceuticals, and personal care products in wastewater contaminants in U.S. streams, 1999–2000: A national water using liquid chromatography/tandem mass spectrometry. Anal reconnaissance. Environ Sci Technol 36:1202–1211. Chem 75:6265–6274. 3. Pal A, Gin KYH, Lin AYC, Reinhard M. 2010. Impacts of emerging 19. Vanderford BJ, Snyder SA. 2006. Analysis of pharmaceuticals in water organic contaminants on freshwater resources: Review of recent by isotope dilution liquid chromatography/tandem mass spectrometry. occurrences, sources, fate and effects. Sci Total Environ 408:6062– Environ Sci Technol 40:7312–7320. 6069. 20. Lao W, Tsukada D, Greenstein DJ, Bay SM, Maruya KA. 2010. 4. Scott PD, Bartkow M, Blockwell SJ, Coleman HM, Khan SJ, Lim R, Analysis, occurrence and toxic potential of pyrethroids and fipronil McDonald JA, Nice H, Nugegoda D, Pettigrove V, Tremblay LA, in sediments from an urban estuary. Environ Toxicol Chem 29:834– Warne MSJ, Leusch FDL. 2014. An assessment of endocrine activity in 851. Australian rivers using chemical and in vitro analyses. Environ Sci 21. Cox MH, Mendez GO, Kratzer CR, Reichard EG. 2003. Evaluation of Pollut Res 21:12951–12967. trace tests completed in 1999 and 2000 on the upper Santa Clara River, 5. Chandler GT, Cary TL, Volz DC, Walse SS, Ferry JL, Klosterhaus SL. Los Angeles and Ventura Counties, California. Water Resources 2004. Fipronil effects on copepod development, fertility, and Investigations Report 03-4277. US Geological Survey, Sacramento, reproduction: A rapid life-cycle assay in 96-well microplate format. CA. Environ Toxicol Chem 23:117–124. 22. Ahr B, Farris M, Lowe CG. 2015. Habitat selection and utilization of 6. Ankley GT, Miller DH, Jensen KM, Villeneuve DL, MartinovicD. white croaker (Genyonemus lineatus) in the Los Angeles and Long 2008. Relationship of plasma sex steroid concentrations in female Beach Harbors and the development of predictive habitat use models. fathead minnows to reproductive success and population status. Aquat Mar Environ Res 108:1–13. Toxicol 88:69–74. 23. Long ER, Dutch M, Weakland S, Chandramouli B, Benskin JP. 2013. 7. Fono LJ, Kolodziej EP, Sedlak DL. 2006. Attenuation of wastewater- Quantification of pharmaceuticals, personal care products, and derived contaminants in an effluent-dominated river. Environ Sci perfluoroalkyl substances in the marine sediments of Puget Sound, Technol 40:7257–7262. Washington, USA. Environ Toxicol Chem 32:1701–1710. 8. Guo YC, Krasner SW. 2009. Occurrence of primidone, carbamazepine, 24. Maruya KA, Dodder NG, Tang C, Lao W, Tsukada D. 2015. Which caffeine, and precursors for N-nitrosodimethylamine in drinking water coastal and marine contaminants are truly emerging? Environ Sci Pollut sources impacted by wastewater. J Am Water Resour As 45:58–67. Res 22:1644–1652. 9. Dodder NG, Maruya KA, Ferguson PL, Grace R, Klosterhaus S, La 25. Katz DR, Cantwell MG, Sullivan JC, Perron MM, Burgess RM, Ho KT, Guardia MJ, Lauenstein GG, Ramirez J. 2014. Occurrence of Charpentier MA. 2013. Factors regulating the accumulation and spatial contaminants of emerging concern in mussels (Mytilus spp.) along distribution of the emerging contaminant triclosan in the sediments of the California coast and the influence of land use, storm water an urbanized estuary: Greenwich Bay, Rhode Island, USA. Sci Total discharge, and treated wastewater effluent. Mar Pollut Bull Environ 443:123–133. 81:340–346. 26. Weston DP, Lydy MJ. 2014. Toxicity of the insecticide fipronil and its 10. Ramirez AJ, Brain RA, Usenko S, Mottaleb MA, O’Donnell JG, Stahl degradates to benthic macroinvertebrates of urban streams. Environ Sci LL, Wathen JB, Snyder BD, Pitt JL, Perez-Hurtado P, Dobbins LL, Technol 48:1290–1297. Brooks BW, Chambliss CK. 2009. Occurrence of pharmaceuticals and 27. Maul, JD, Brennan AA, Harwood AD, Lydy MJ. 2008. Effect of personal care products in fish: Results of a national pilot study in the sediment-associated pyrethroids, fipronil, and metabolites on Chiro- United States. Environ Toxicol Chem 28:2587–2597. nomus tentans growth rate, body mass, condition index, immobiliza- 11. Sedlak MD, Greig DJ. 2012. Perfluoroalkyl compounds (PFCs) in tion, and survival. Environ Toxicol Chem 27:2582–2590. wildlife from an urban estuary. J Environ Monit 14:146–154. 28. Luigi V, Giuseppe M, Claudio R. 2015. Emerging and priority 12. Meng XZ, Blasius ME, Gossett RW, Maruya KA. 2009. Polybromi- contaminants with endocrine active potentials in sediments and fish nated diphenyl ethers (PBDEs) in pinnipeds stranded along the southern from the River Po (Italy). Environ Sci Pollut Res 22:14050–14066. California coast. Environ Pollut 157:2731–2736. 29. Wang Q, Lam JCW, Wang X, Guo Y, Lam PKS, Zhou B. 2015. 13. Fischer D, Hooper K, Athanasiadou M, Athanassiadis I, Bergman Å. Developmental exposure to the organophosphorus flame retardant 2006. Children show highest levels of polybrominated diphenyl ethers tris(1,3-dichloro-2-propyl) phosphate: Estrogenic activity, endocrine in a California family of four: A case study. Environ Health Perspect disruption and reproductive effects on zebrafish. Aquat Toxicol 114:1581–1586. 160:163–171.