Persistent organic pollutants were assessed in Humboldt Penguins (Spheniscus humboldti) from the Punta San Juan Marine Protected Area, Peru, in the austral winter of 2009. Plasma samples from 29 penguins were evaluated for 31 polychlorinated biphenyl (PCB) congeners and 11 organochlorine pesticides (OCPs) by using gas chromatography coupled to an ion trap mass spectrometer and for 15 polybrominated diphenyl ether (PBDE) congeners by using gas chromatography coupled with high-resolution mass spectrometry. The detection rate for PCBs in the samples was 69%, with congeners 105, 118, 180, and 153 most commonly detected. The maximum ΣPCB concentration was 25 ng/g. The detection rate for DDT, DDD, and/or DDE was higher than for other OCP residues (90%; maximum concentration=10 ng/g). The detection rate for PBDEs was 86%, but most concentrations were low (maximum ΣPBDE concentration=3.81 ng/g). This crucial breeding population of S. humboldti was not exposed to contaminants at levels detrimental to health and reproductive success; however, the identified concentrations of legacy and recently emerged toxicants underscore the need for temporal monitoring and diligence to protect this endangered species in the face of regional human population and industrial growth. These results also provide key reference values for spatial comparisons throughout the range of this species.

The Humboldt Penguin (Spheniscus humboldti) is native to coastal habitats in northern Chile and southern Peru. The species experienced drastic declines in the 19th century due to human activities and loss of nesting substrate throughout its range related to commercial harvesting of guano for fertilizer. Today, the effects of commercial fisheries and El Niño–Southern Oscillation events on prey availability restrict population recovery, and bycatch, hunting, predation from introduced species, and human disturbance remain active population threats (de la Puente et al. 2013; Trathan et al. 2015). The species is recognized as vulnerable by the International Union for Conservation of Nature (BirdLife International 2017), threatened by the US Endangered Species Act (US Fish and Wildlife Service 2017), and on Appendix I of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (2017).

The Punta San Juan (PSJ) marine protected area protects the largest breeding population of S. humboldti in Peru (3,000–5,000 penguins), accounting for 25–50% of the entire Peruvian population and 10% of the global population on an annual basis (Paredes et al. 2003; de la Puente 2013; BirdLife International 2017). This site is crucial to the long-term survival of the species in Peru and is the focus of multiple conservation efforts. Population health and reproductive success at PSJ have been intensively assessed over the past two decades, but characterization of toxicant exposure is lacking. Regional human population growth and industrial development in the adjacent town of San Juan de Marcona (SJM) raise concerns of environmental contamination in the local marine ecosystem. The largest open-pit iron ore mine in Peru is located 20 km from PSJ, and an expanding copper mine is within 6 km. Mine waste, as well as SJM sewage and refuse, are deposited in multiple locations, some as close as 4 km away. In addition to penguins, PSJ protects important populations of multiple seabird species and is a key rookery for South American sea lions (Otaria byronia) and Peruvian fur seals (Arctocephalus australis).

There is presently a paucity of information related to marine contamination in the PSJ region, making it difficult to assess potential toxicant impacts on species and ecosystem health. Toxicant exposure data for Spheniscus penguins throughout their range are also sparse (Smith et al. 2008; Baldassin et al. 2016), but studies in other locations have established the use of penguins as bioindicators of environmental pollution. Research in Antarctic penguins has long demonstrated exposure and bioaccumulation of persistent organic pollutants, such as polychlorinated biphenyls (PCBs), DDT, and others (George and Frear 1966).

Our aim was to quantify the occurrence of PCBs, organochlorine pesticides (OCPs), and polybrominated biphenyl ethers (PBDEs) in the blood of S. humboldti at PSJ. The project was conducted concurrently with other initiatives assessing S. humboldti population health to build effective conservation strategies. With established data, future research of acute and chronic perturbations in animal population health at PSJ can be assessed against known levels of environmental toxicant exposure.

Sample collection and processing

During May 2009, we collected samples from penguins within the PSJ marine protected area, Ica, Peru (15°22′S, 75°12′W), by using collection methods authorized under Peruvian permit 131-2009-AG-DGFFS-DGEFFS (Fig. 1). Twenty-nine adult penguins were removed from nests in burrows or sea caves. Eggs and small chicks were first removed from nests by using a cupped pole to avoid trauma, and then they were provided with supplemental heat while the adult was sampled. Penguins were released at the site of capture.

Figure 1

Map depicting the Punta San Juan, Peru marine protected area where plasma samples were collected from 29 Humboldt penguins (Spheniscus humboldti) in May 2009 for analysis of persistent organic pollutant concentrations. The dashed line represents a wall physically isolating the protected headland and associated beaches labeled north (N) and south (S) from 0 to 9.

Figure 1

Map depicting the Punta San Juan, Peru marine protected area where plasma samples were collected from 29 Humboldt penguins (Spheniscus humboldti) in May 2009 for analysis of persistent organic pollutant concentrations. The dashed line represents a wall physically isolating the protected headland and associated beaches labeled north (N) and south (S) from 0 to 9.

Close modal

All penguins were examined by a veterinarian and determined to be in normal health based on physical examination, along with hematologic and plasma biochemistry parameters. We used a 20-gauge needle to collect up to 24 mL of blood from the jugular vein. Blood was immediately placed into lithium heparin tubes (Vacutainer, BD, Franklin Lakes, New Jersey, USA) and stored in a cooler on ice until processing. Plasma was separated by centrifugation within 6 h of collection. The plasma was placed into cryovials (NUNC, Thermo Fisher Scientific, Rochester, New York, USA) and promptly frozen to −20 C. Within 1–5 d, samples were placed in liquid nitrogen for export and subsequently maintained at –70 C until laboratory analysis.

We prepared three control blanks (cryovial tubes) that were included at various points of sample collection and processing to determine the entrance of any environmental contaminants. Deionized water was aliquoted into the first blank tube and into a bottle before travel to the field site. This bottle was subsequently carried to the field site and used to fill two additional blank tubes with water at the time of blood collection and sample centrifugation.

Sample preparation and analytes

Plasma (3 g) from each animal was weighed and treated with methanol and mixed for 10 min with a laboratory mixer. Then, the samples were extracted with hexane-dichloromethane-acetone (1:1:0.6) and mixed again for 30 min. These extracts were centrifuged for 5 min at 250 × G, and the organic phase was removed. The extraction procedure was repeated two more times, and then the organic phases were pooled. Next, the extracts were processed by a solid phase extraction (silica gel cleanup) procedure per US Environmental Protection Agency (USEPA) Method 3630C (USEPA 1996a). Two silica gel fractions were collected from each sample and were concentrated to a final volume of 1.0 mL. Thirty-two samples were prepared in four extraction batches. Analytes and method detection limits are presented in Table 1.

Table 1

Method detection limits for gas chromatography coupled to an ion trap mass spectrometer used to determine concentrations of 42 polychlorinated biphenyl congeners and organochlorine pesticides, and method detection limits for gas chromatography coupled with high-resolution mass spectrometry used to determine concentrations of 15 polybrominated diphenyl ether congeners in the plasma of 29 Humboldt Penguins (Spheniscus humboldti) from Punta San Juan, Peru, that were sampled during May 2009 to assess exposure of a key breeding population to persistent organic pollutants.a

Method detection limits for gas chromatography coupled to an ion trap mass spectrometer used to determine concentrations of 42 polychlorinated biphenyl congeners and organochlorine pesticides, and method detection limits for gas chromatography coupled with high-resolution mass spectrometry used to determine concentrations of 15 polybrominated diphenyl ether congeners in the plasma of 29 Humboldt Penguins (Spheniscus humboldti) from Punta San Juan, Peru, that were sampled during May 2009 to assess exposure of a key breeding population to persistent organic pollutants.a
Method detection limits for gas chromatography coupled to an ion trap mass spectrometer used to determine concentrations of 42 polychlorinated biphenyl congeners and organochlorine pesticides, and method detection limits for gas chromatography coupled with high-resolution mass spectrometry used to determine concentrations of 15 polybrominated diphenyl ether congeners in the plasma of 29 Humboldt Penguins (Spheniscus humboldti) from Punta San Juan, Peru, that were sampled during May 2009 to assess exposure of a key breeding population to persistent organic pollutants.a

Instrumentation and methodology

We performed analyses for PCB and OCP residues per USEPA Method 8270C (USEPA 1996b). The analyses were done using a gas chromatograph (model 3800, Varian, Palo Alto, California, USA) coupled to an ion trap mass spectrometer (Saturn 2000, Varian). Separation was achieved with a capillary column (Rtx-5ms, 30 m×0.25 mm i.d., 0.25-μm film, Restek 12623, Restek, Bellefonte, Pennsylvania, USA) with helium at a flow rate of 2.0 mL/min as the carrier gas. A splitless injection of 4 μL was performed at 280 C. The oven was held at 45 C for 4 min and then was ramped to 320 C at a rate of 10 C/min. Compounds eluted from the column were fragmented by electron impact, with the mass spectrometer operated in a selective ion storage mode, whereas only fragments of interest were stored and measured.

We performed analyses of PBDE by gas chromatography coupled with high-resolution mass spectrometry as per USEPA Method 1614 (USEPA 2003), modified by the use of only two surrogates, a penta-brominated PBDE and a deca-brominated PBDE, for analysis due to the costs of purchasing labeled compounds for each congener. Separation of the target analytes was achieved with a gas chromatography system (model 6890, Agilent, New York, New York, USA) equipped with a Stx®-500 column (catalog no. 11543, 15 m×0.25 mm×0.15 μm df, Restek Corporation, U.S., Bellefonte, Pennsylvania, USA). The oven operating conditions were as follows: 80 C, hold 2 min; 20 C/min until 240 C was reached; 5 C/min to 265 C; and 10 C/min to 320 C/min, hold 6 min. A 4-μL splitless injection was performed at 320 C. Helium at 0.7 mL/min was used as the carrier gas. Detection of the PBDE congeners was achieved with a high-resolution mass spectrometer (Autospec Ultima NT, Waters Corporation, Milford, Massachusetts, USA). The instrument was operated at a resolution at or greater than 10,000, and single ion recording methods were constructed to improve sensitivity. Perfluorokerosene was used for mass calibrations, and the appropriate perfluorokerosene mass fragments were used as lock masses during PBDE analysis.

Instruments were calibrated before each run with reference materials purchased from a certified vendor (Accustandard, New Haven, Connecticut, USA). Pentachloronitrobenzene was used as an internal standard during PCB and OCP gas chromatography mass spectrometry analysis. Decachlorinated PCB 209 was used as an internal standard during the PBDE analysis by gas chromatography coupled with high-resolution mass spectrometry.

Quality control

Our quality control parameters included all surrogates, reagent blanks, reagent blank spikes, and sample matrix spikes. In addition, we performed a silica gel spike with each clean-up batch to ensure proper fractionation of the target compounds. Quality control parameters at the instrument level included an analytical sample duplicate and analytical sample spike for each sample batch. Surrogates (rare or isotope-enriched compounds) were added to all samples and were measured along with target analytes to assess the efficiency of the extraction based on their recovery. Surrogates included three PCBs and two isotope-enriched PBDEs. Reagent blanks consisted of 3 g of deionized water. Reagent blank spikes were prepared by spiking the target compounds into 3 g of deionized water. Sample matrix spikes were performed in duplicate on a random sample from the set with all the target compounds spiked into it. A silica gel spike was performed by spiking the target compounds onto a silica gel column and eluting the column in exactly the same manner as the other samples within the set. Analytical duplicate samples and analytical duplicate spikes were prepared during the sample analysis and were prepared by selecting random samples and either preparing them in duplicate or spiking them with the target analytes.

Statistics

Data were checked for normality using the Shapiro-Wilk and Anderson-Darling tests. Homogeneity of variances among groups was tested using Levene's test. We performed analyses using Systat 13.0 (Systat, Inc., Chicago, Illinois, USA) and SAS 9.1.3 (SAS Institute, Inc., Cary, North Carolina, USA). Values below the detection limit were proxied as the detection limit, thereby estimating a maximum concentration (hence, risk) in that penguin.

As a conservation measure, eggs from S. humboldti were not taken for analysis. Total DDE egg concentrations were estimated using an avian model relating ΣDDE concentrations in adult female birds to egg concentrations. We developed this model using published plasma and egg concentrations for American Kestrels (Falco sparverius), Northern Goshawks (Accipiter gentilis), Cooper's Hawks (Accipiter cooperii), and Sharp-shinned Hawks (Accipiter striatus) in a DDT spray area in the Pacific Northwest (Henny and Meeker 1981); Bald Eagles (Haliaeetus leucocephalus; Strause et al. 2007); Black-footed Albatrosses (Phoebastria nigripes) and Laysan Albatrosses (Phoebastria immutabilis; Auman et al. 1997); and Glaucous Gulls (Larus hyperboreus; Verreault et al. 2005). Initially, point estimates of total DDE in S. humboldti eggs (EggHP) were calculated from the means: EggHP=(PlasmaHP/PlasmaAVIAN)×EggAVIAN. Because the Glaucous Gull values were expressed relative to the percent lipid concentration in the tissue, the ratio Lipid%PLASMA/Lipid%EGG was used as a multiplicative adjustment. DDT concentrations in the two albatross species had to be estimated in a different way because the ratio of the means, , was smaller than the mean of the ratios from the n individual albatrosses . Specifically, the concentration of DDT was estimated as DDE/R. Then, to obtain a distribution for the concentrations in S. humboldti eggs, the means and SDs of the input variables, and R, were expressed as distributions in Crystal Ball 11.2.0.00 (Oracle, Redwood Shore, California, USA). The lognormal distribution was used for the pesticide concentrations in plasma and eggs of the avian species except for the gull; the normal distribution was used for all Glaucous Gull parameters. For each avian species, 1,000 estimates of EggHP were obtained. The 8,000 estimates were pooled, standardized, and z values >3 were excluded as outliers. Summary statistics were obtained for the included values.

Homolog concentrations were computed by summing the concentrations of the PCB congeners having the same number of chlorine molecules. Homolog concentrations for each penguin were expressed as proportions of the total PCB concentration. The proportion of dichloro-PCBs was not used in the repeated measures analysis of variance due to colinearity (caused by Σhomologs=1 for each penguin).

Quality control

The overall average recovery for the four PCB surrogates in all 32 samples was 88%, with a relative SD of 25%. One sample produced a surrogate recovery of 225% for labeled PBDE congener 118 and 0% for labeled PBDE congener 209. This sample was most likely inadvertently spiked twice with labeled PBDE congener 118, instead of once for each surrogate. In addition, one blank sample produced a surrogate recovery of 0% for labeled PBDE congener 118. This sample also was most likely not spiked with this surrogate. When calculating surrogate recoveries for PBDEs, these aberrant recoveries were not included in the mean surrogate calculations. The overall average for the labeled PBDE surrogates was 88% (relative SD 21%).

Reagent blanks were free from PCB, OCP, and PBDE contaminations. Reagent blank spikes, silica gel quality check spikes, sample matrix spikes, and analytical sample spikes recovered well, with average recoveries ranging from 67% to 122%. Analytical duplicates also recovered well for detectable target analytes.

Organohalogens

Overall, the detection rate (i.e., the proportion of birds having at least one detectable congener) for PCBs in the plasma of individual penguins was 69% (Table 2). The incidence of the more prevalent individual PCBs was as follows: congeners 105 (23%), 118 (39%), 180 (48%), and 153 (61%). The PCB congeners 52, 70, 66, 95, 99, 84/101, 77, 128, 163/138, 149, 187, and 183 were detected in less than 20% of the penguins. The PCBs 18, 31, 28, 33, 49, 44, 74, 126, 194, and 200 were not detected. The PCB congener profile in the plasma of the penguins in this study was depleted in the lower (<5 chlorine) chlorinated congeners (Fig. 2).

Table 2

Summary statistics for total polychlorinated biphenyls (∑PCB), total DDT metabolites (∑DDT), and total non-DDT organochlorine pesticides (∑OCP) concentrations (ng/g) in plasma samples obtained from 29 adult male and female Humboldt penguins (Spheniscus humboldti) at Punta San Juan, Peru in May 2009 determined using gas chromatography coupled to an ion trap mass spectrometer.

Summary statistics for total polychlorinated biphenyls (∑PCB), total DDT metabolites (∑DDT), and total non-DDT organochlorine pesticides (∑OCP) concentrations (ng/g) in plasma samples obtained from 29 adult male and female Humboldt penguins (Spheniscus humboldti) at Punta San Juan, Peru in May 2009 determined using gas chromatography coupled to an ion trap mass spectrometer.
Summary statistics for total polychlorinated biphenyls (∑PCB), total DDT metabolites (∑DDT), and total non-DDT organochlorine pesticides (∑OCP) concentrations (ng/g) in plasma samples obtained from 29 adult male and female Humboldt penguins (Spheniscus humboldti) at Punta San Juan, Peru in May 2009 determined using gas chromatography coupled to an ion trap mass spectrometer.
Figure 2

Polychlorinated biphenyls congener profile based on the concentrations found in the plasma of 29 Humboldt penguins (Spheniscus humboldti) from Punta San Juan, Peru. Horizontal dashed line=0.3 ng/g, used as the detection limit. Plotted value is arithmetic mean and error bars represent SD. Open bars represent values with a mean below the detection limit. Vertical dashed lines mark homolog groups (i.e., the first line is CL3, then CL4, and subsequent numbers).

Figure 2

Polychlorinated biphenyls congener profile based on the concentrations found in the plasma of 29 Humboldt penguins (Spheniscus humboldti) from Punta San Juan, Peru. Horizontal dashed line=0.3 ng/g, used as the detection limit. Plotted value is arithmetic mean and error bars represent SD. Open bars represent values with a mean below the detection limit. Vertical dashed lines mark homolog groups (i.e., the first line is CL3, then CL4, and subsequent numbers).

Close modal

At 90%, the detection rate for DDT, DDD, and/or DDE in blood was higher than for other OCP residues (Table 2). The mean recovery of DDE was 2.61 ng/g (range, 0–9.8 ng/g). With nondetection results replaced by the detection limit (0.2 ng/g for p,p′-DDE; 0.3 ng/g for p,p′-DDD and p,p′-DDT), pairwise Spearman rank correlations between p,p'-DDE, p,'p-DDD, and p,p'-DDT were not significant (all P values were >0.32). DDE represented 81%, on average, of the total DDT (DDT+DDD+DDE) concentration. Using the described statistical methods, a mean DDE concentration of 0.57 μg/g wet weight was estimated in eggs.

Aldrin, lindane, oxychlordane, α-chlordane, and dieldrin were not detected in any penguin. Trans-nonachlor was detected in five penguins, with concentrations ranging from 0.43 to 1.1 ng/g. Heptachlor and γ-chlordane were detected in two penguins at concentrations of 0.65 and 0.21 ng/g, respectively.

Residues of PBDEs were detected in 25 of 29 (86%) penguins; however, most concentrations were at or below the quantifiable detection limit (Table 1). Six penguins had congener concentrations above detection limits that could be quantified, with a maximum ΣPBDE concentration of 3.81 ng/g. Congeners 47, 99, and 100 were the most prevalent.

Residues of PCBs, OCPs, and PBDEs represent common and toxicologically important environmental contaminants. All are linked to endocrine disruption, neoplasia, eggshell thinning, impaired reproductive success, and population declines (Bustnes et al. 2004; Miljeteig et al. 2012). Responses vary amongst taxa, but seabirds commonly are affected (Blus 1996). The lipid-soluble nature of these compounds makes them prone to bioaccumulation and biomagnification through food webs. Contamination with PCBs and PBDEs may come from insulating materials, flame retardants, and multiple other industrial sources. They are persistent in the environment, resulting in greater risk for bioaccumulation in higher trophic-level animals with long lifespans, such as piscivorous penguins (Furness and Campyhuysen 1997). These compounds tend to cause little acute mortality, but they can have severe effects on avian endocrine systems, reproduction, and long-term population health (O'Hara and Rice 1996).

Site- and species-specific health data are important for guiding long-term conservation efforts. The only study of persistent organic pollutants in S. humboldti was conducted between 1992 and 1994 and concluded analyzed plasma OCP and PCB concentrations were below detectable limits of 0.001–0.007 ppm (μg/g) and 0.05 ppm, respectively (Smith et al. 2008). These historical data most likely no longer accurately reflect current conditions due to increased regional development, growth in coastal human populations, changes in chemical use, and other factors that all affect the fate and transport of chemicals in the environment. The utility of Smith et al. (2008) as a reference point is also difficult, as individual detection limits and specific PCB congeners analyzed were not reported, and no PDBEs were analyzed. With few reports of contaminant concentrations in S. humboldti, our study's results are best compared with data from other penguin and avian species (Yang et al. 2009; Feng et al. 2010; Corsolini et al. 2011).

The 69% detection rate and maximum ΣPCB concentration of 25 ng/g in this study indicated widespread exposure to low levels. Congener profiles in S. humboldti are affected not only by the global distribution of PCBs and concentrations in food, but also through recycling due to penguin activity at the colony site. Similar to the findings of Roosens et al. (2007), the penguins in our study demonstrated a predominance of persistent congeners such as PCB 99, 118, 138, and 153. The fact that the penguins at PSJ tend to stay at the colony year-round, depositing guano and other by-products of habitation, allows for the localized concentration of PCBs in the soils and nesting areas and amplification of these congeners due to biotransformation and accumulation in the environment and local biota (Roosens et al. 2007).

Corsolini et al. (2007) reports the total concentrations of 43 PCB congeners in whole blood samples from three Pygoscelis penguins and documents differences in PCB concentrations among them as attributable, in part, to dietary differences. In a study of Adélie Penguins (Pygoscelis adeliae), the blood PCB profiles exhibit seasonal fluctuation that reflect the physiologic condition and energy demands of the birds due to diet availability and acquisition, excretion, and maternal transfer (van den Brink et al. 1998). These results all emphasize the importance of establishing species- and site-specific baseline levels for toxicants, as it is difficult to make comparisons and generalizations between PCB levels across multiple avian species and locations, each with its own unique environmental variables.

Our detection rates for DDT and its metabolites DDD and DDE (DDTs) were higher than those of other OCPs (90%; maximum concentration=10 ng/g). The presence of some low-level exposure to DDTs is not surprising, given the continued global cycling of this pesticide and its relatively recent use in Peru. Use of DDT for agricultural purposes in Peru continued into the early 1990s (Roberts et al. 1997), and studies continued to find DDT in agricultural drainage beyond that time (Palm 2007). Residues of DDT, and to a lesser extent other OCPs, are detectable in marine fish and invertebrates from the coast near the major Peruvian population center of Lima (Cabello-Torres and Sanchez-Rivas 2006). Although the human population in the Ica region of Peru is smaller, towns are concentrated along the river courses near the coast, with several larger population centers located further inland. Narrow river floodplains are intensively cultivated, often with irrigated crops, and rivers thereby represent a potential source of coastal contamination. For PSJ, the closest sources of significant agricultural runoff are the Rio Acari (65 km to the south) and Rio Grande (51 km to the north), suggesting that wildlife at PSJ is relatively naïve to agriculture runoff. However, more comprehensive spatial data are needed on S. humboldti ocean use, along with OCP testing of penguins at other coastal sites.

Geisz et al. (2008) finds that although ratios of p,p'-DDT to p,p'-DDE in Adélie Penguins have declined since 1964, ΣDDT concentrations have increased in Adélie Penguins from 1964 to 2006. They attribute the increase to new food-web contamination from the 1–4 kg ΣDDT being released annually into coastal waters along the western Antarctic ice sheet from glacier ablation and melting south polar ice. If this is happening, it is plausible that such compounds are also carried on the Antarctic Circumpolar Current and then northward on the Humboldt Current toward PSJ, representing an additional source of exposure. Plasma DDE concentrations in our study were low compared with concentrations that were two to three times higher in whole blood from three Antarctic penguin species from King George Island (Corsolini et al. 2007), but this comparison must be made with caution because it crosses species and ignores metabolic considerations.

The DDTs found in eggs of Antarctic penguins vary considerably among locations, years, and species (Schiavone et al. 2009). Our estimation of DDE levels in S. humboldti eggs provided a useful tool to assess potential toxicosis and detrimental effects on reproduction when collection of eggs for direct measurements is not feasible. Our estimated mean DDE concentration of 0.572 μg/g (wet weight) in S. humboldti eggs was considerably greater than that reported in a previous study of penguins (Schiavone et al. 2009), but it was well below that associated with reproductive impairment in Bald Eagles (>3.6 μg/g; Wiemeyer et al. 1993) and Brown Pelicans (Pelecanus occidentalis; 3.0 μg/g; Blus 1996). Our maximum estimated value (2.4 μg/g) did approach these levels, but there is substantial variation in sensitivity across taxa, and generalizations must be made with caution. Research and biologic monitoring at PSJ over the past 20 yr has not revealed any evidence of impaired reproduction in S. humboldti consistent with DDE exposure. Examination of eggs in 2009 during this study period did not reveal evidence of shell thinning or abnormalities. In addition, Subramanian et al. (1986) and Tanabe et al. (1986) found that compared with other seabirds, very little of the OCP burden in penguins is transferred to eggs; therefore, our estimated mean DDE concentration in eggs (derived from blood-egg relationships in nonpenguin species) may be overestimated for penguins. Data are needed to confirm our estimated egg concentrations. Establishing a species-specific correlation between blood and egg concentrations would be beneficial for long-term monitoring of this and other populations.

Previous studies demonstrate the presence and distribution of PBDEs in the marine environment (Corsolini et al. 2007; Yogui and Sericano 2009; Van den Steen et al. 2011), but data are lacking for piscivorous marine bird species endemic to Peru. Species and location-specific data are desirable, as many variables (e.g., migratory behavior, prey availability, metabolism, and long-range transport) affect organohalogen profiles. The concentrations of PBDEs are much lower in the eggs of Gentoo Penguins (Pygoscelis papua) and Chinstrap Penguins (Pygoscelis antarcticus) compared with that of the migratory, piscivorous South Polar Skua (Stercorarius maccormicki) inhabiting the same breeding range (Yogui and Sericano 2009). Similarly, mean PBDE concentrations in the Falkland Islands are lower in the krill-feeding Rockhopper Penguins (Eudyptes chrysocome) compared with the more piscivorous Imperial Shag (Phalacrocorax atriceps; Van den Steen et al. 2011). Even between similar species, prey preference complicates comparisons; for example, Corsolini et al. (2007) report blood PBDE concentrations that are 40% lower in Gentoo and Chinstrap Penguins consuming predominantly krill diets compared with levels in Adélie Penguins that consume krill and fish. Although our detection rates and concentrations were low compared with other piscivorous species (Yogui and Sericano 2009; Van den Steen et al. 2011), PBDEs 17, 28, 47, 99, and 100 predominated in one or more of the penguins, with measureable concentrations. Continued monitoring at PSJ may help demonstrate the long-range effects of volatilization, the transport and persistence of PBDEs within this population, and the use of wildlife as a bioindicator of human exposure.

The low concentrations of industrial chemicals (i.e., PCBs and PBDEs) in S. humboldti at PSJ is not surprising given the distance (>400 km) from the major coastal population, industrial, and shipping center in Lima; however, the increasing scale and changing structure of regional industry in SJM raises significant concern regarding future environmental contamination. As coastal development continues to increase and creates increased human presence in previously remote areas, studies like this are critical for the conservation of wildlife. Muñoz and Becker (1999) and Cifuentes et al. (2003) note higher concentrations of toxicants such as PCBs in areas of industrial activity as well as higher concentrations of DDT and metabolites near estuaries draining agricultural lands. The dramatic growth of the SJM region over the past 50 yr and construction of a transcontinental highway linking SJM to Brazil will only spur further regional industrial development. Legislation passed in 2005 also deemed the development of a mega-port (240,000-metric ton ship capacity) in the adjacent San Nicolás Bay to be in Peru's national interest. This and other coastal developments pose great risk of increased industrial waste and environmental contamination, underscoring the need for robust, critical evaluation of such projects before decisions are reached that could impact wildlife populations. Baseline contaminant data are a key component of such environmental impact assessments.

At present, S. humboldti at PSJ do not seem to be exposed to environmental contaminants at levels detrimental to individual health or reproductive success; however, the detection of measurable concentrations of legacy and recently emerged toxicants underscores the need for continued monitoring and diligent conservation policies. Our study provides important reference data for monitoring this key population over time and establishes comparison data for assessing other penguin rookeries. Cohesive conservation strategies for S. humboldti and the Peruvian Humboldt Current ecosystem rely on objective data and the ability to monitor temporal changes, particularly in the face of continued growth of regional industries and increasing human populations.

We thank Val Beasley, Marco Cardeña, Franco Garcia, Paulo Guerrero, Maria José Ganoza, Wendy Flores, Alonso Bussalleu, and Michael Macek for support of this project. In addition, we thank Nandakishore Rajagopalan, Gerald Bargren, and Christie Teausant for technical assistance and laboratory support with chemical analysis. We gratefully acknowledge support from SERNANP and facility access by AGRO RURAL. Funding for this project was provided by the Saint Louis Zoo WildCare Institute, the Saint Louis Zoo's Field Research for Conservation Fund, and the Chicago Zoological Society's Chicago Board of Trade Endangered Species Fund. The Department of Comparative Biosciences, University of Illinois at Urbana-Champaign, contributed services in-kind.

Auman
HJ,
Ludwig
JP,
Summer
CL,
Verbrugge
DA,
Froese
KL,
Colborn
T,
Giesy
JP.
1997
.
PCBS, DDE, DDT, and TCDD-EQ in two species of albatross on Sand Island, Midway Atoll, North Pacific Ocean
.
Environ Toxicol Chem
16
:
498
504
.
Baldassin
P,
Taniguchi
S,
Gallo
H,
Maranho
A,
Kolesnikovas
C,
Amorim
DB,
Mansilla
M,
Navarro
RM,
Tabeira
LC,
Bicego
MC,
et al.
2016
.
Persistent organic pollutants in juvenile Magellanic Penguins (Spheniscus magellanicus) in South America
.
Chemosphere
149
:
391
399
.
BirdLife International
.
2017
.
Spheniscus humboldti
.
The International Union for Conservation of Nature red list of threatened species 2017: e.T22697817A111228184
. .
Blus
LJ.
1996
.
DDT, DDD, and DDE in birds
.
In
:
Environmental contaminants in wildlife: Interpreting tissue concentrations
,
Beyer
WN,
Heinz
GH,
Redmon-Norwood
AW,
editors
.
Lewis Publishers, CRC Press
,
Boca Raton, Florida
,
pp
.
49
71
.
Bustnes
JO,
Hanssen
SA,
Folstad
I,
Erikstad
KE,
Hasselquist
D,
Skaare
JU.
2004
.
Immune function and organochlorine pollutants in arctic breeding glaucous gulls
.
Arch Environ Contamin Toxicol
47
:
530
541
.
Cabello-Torres
RJ,
Sánchez-Rivas
G.
2006
.
Residuos de contaminantes orgánicos persistentes (COPs) en áreas costeras del Mar Peruano: Callao y Pisco-Paracas
.
Publicaciones Iremar Boletin
33
:
75
110
.
Spanish
.
Cifuentes
JM,
Becker
PH,
Sommer
U,
Pacheco
P,
Schlatter
R.
2003
.
Seabird eggs as bioindicators of chemical contamination in Chile
.
Environ Pollut
126
:
123
137
.
Convention on International Trade in Endangered Species of Wild Fauna and Flora
.
2017
.
Appendices I, II, and III
. .
Corsolini
S,
Borghesi
N,
Ademollo
N,
Focardi
S.
2011
.
Chlorinated biphenyls and pesticides in migrating and resident seabirds from east and west Antarctica
.
Environ Int
37
:
1329
1335
.
Corsolini
S,
Borghesi
N,
Schiamone
A,
Focardi
S.
2007
.
Polybrominated diphenyl ethers, polychlorinated dibenzo-dioxins, -furans, and -biphenyls in three species of Antarctic penguins
.
Environ Sci Pollut Res
14
:
421
429
.
de la Puente
S,
Bussalleu
A,
Cardeña
M,
Valdés-Velasquez
A,
Majluf
P,
Simeone
A.
2013
.
Humboldt Penguin (Spheniscus humboldti)
.
In
:
Penguins: Natural history and conservation
,
García-Borboroglu
P,
Boersma
PD,
editors
.
University of Washington Press
,
Seattle, Washington
,
pp
.
265
284
.
Feng
ZJ,
Yu
PS,
Lu
B,
Cai
MH,
Wu
GH.
2010
.
Distribution of organochlorine residues-PCBs and OCPs in tissue, eggs and dropping-amended soil of penguins in Antarctic
.
Mar Environ Sci
29
:
309
313
,
355
.
Furness
RW,
Camphuysen
K.
1997
.
Seabirds as monitors of the marine environment
.
ICES J Mar Sci
54
:
726
737
.
Geisz
HN,
Dickhut
RM,
Cochran
MA,
Fraser
WR,
Ducklow
HW.
2008
.
Melting glaciers: A probable source of DDT to the Antarctic marine ecosystem
.
Environ Sci Technol
42
:
3958
3962
.
George
JL,
Frear
DEH.
1966
.
Pesticides in the Antarctic
.
J Appl Ecol
3 (Suppl)
:
155
167
.
Henny
CJ,
Meeker
DL.
1981
.
An evaluation of blood plasma for monitoring DDE in birds of prey
.
Environ Pollut Ser A
25
:
291
304
.
Miljeteig
C,
Gabrielsen
GW,
Strøm
H,
Gavrilo
MV,
Lie
E,
Jenssen
BM.
2012
.
Eggshell thinning and decreased concentrations of vitamin E are associated with contaminants in eggs of ivory gulls
.
Sci Total Environ
431
:
92
99
.
Muñoz
J,
Becker
PH.
1999
.
The Kelp Gull as bioindicator of environmental chemicals in the Magellan region: A comparison with other coastal sites in Chile
.
Sci Mar
63
:
495
502
.
O'Hara
TM,
Rice
CD.
1996
.
Polychlorinated biphenyls
.
In
:
Noninfectious diseases of wildlife
,
2nd Ed.,
Fairbrother
A,
Locke
LN,
Hoff
GL,
editors
.
Iowa State University Press
,
Ames, Iowa
,
pp
.
71
86
.
Palm
B.
2007
.
Pesticide use in rice cultivation in Tarapoto, Peru: Usage patterns and pesticide residues in water sources
.
Master's Thesis
,
Swedish University of Agricultural Sciences
,
Uppsala, Sweden
,
130
pp
.
Paredes
R,
Zavalaga
CB,
Battistini
G,
Majluf
P,
McGill
P.
2003
.
Status of the S. humboldti in Peru, 1999–2000
.
Waterbirds
26
:
129
138
.
Roberts
DR,
Laughlin
LL,
Hsheih
P,
Legters
LJ.
1997
.
DDT, global strategies, and a malaria control crisis in South America
.
Emerg Infect Dis
3
:
295
302
.
Roosens
L,
Van Den Brink
N,
Riddle
M,
Blust
R,
Neels
H,
Covaci
A.
2007
.
Penguin colonies as secondary sources of contamination with persistent organic pollutants
.
J Environ Monit
9
:
822
825
.
Schiavone
A,
Corsolini
S,
Borghesi
N,
Focardi
S.
2009
.
Contamination profiles of selected PCB congeners, chlorinated pesticides, PCDD/Fs in Antarctic fur seal pups and penguin eggs
.
Chemosphere
76
:
264
269
.
Smith
KM,
Karesh
WB,
Majluf
P,
Paredes
R,
Zavalaga
C,
Reul
AH,
Stetter
M,
Braselton
WE,
Puche
H,
Cook
RA.
2008
.
Health evaluation of free-ranging Humboldt penguins (Spheniscus humboldti) in Peru
.
Avian Dis
52
:
130
135
.
Strause
KD,
Zwiernik
MJ,
Im
SH,
Newsted
JL,
Kay
DP,
Bradley
PW,
Blankenship
AL,
Williams
LL,
Giesy
JP.
2007
.
Plasma to egg conversion factor for evaluating polychlorinated biphenyl and DDT exposures in great horned owls and bald eagles
.
Environ Toxicol Chem
26
:
1399
1409
.
Subramanian
AN,
Tanabe
S,
Hidaka
H,
Tatsukawa
R.
1986
.
Bioaccumulation of organochlorines (PCBs and p,p′-DDE) in Antarctic Adelie penguins Pygoscelis adeliae collected during a breeding season
.
Environ Pollut Ser A
40
:
173
189
.
Tanabe
S,
Subramanian
AN,
Hidaka
H,
Tatsukawa
R.
1986
.
Transfer rates and pattern of PCB isomers and congeners and p,p′-DDE from mother to egg in Adelie penguin (Pygoscelis adeliae)
.
Chemosphere
15
:
343
351
.
Trathan
PN,
García-Borboroglu
P,
Boersma
D,
Bost
CA,
Crawford
RJM,
Crossin
GT,
Cuthbert
RJ,
Dann
P,
Davis
LS,
de la Puente
S,
et al.
2015
.
Pollution, habitat loss, fishing, and climate change as critical threats to penguins
.
Conserv Biol
29
:
31
41
.
USEPA (US Environmental Protection Agency)
.
1996
a
.
Method 3630C: Silica gel cleanup
. .
USEPA
.
1996
b
.
Method 8270C: Semivolatile organic compounds by gas chromatography/mass spectrometry (GC/MS)
. .
USEPA
.
2003
.
Method 1614: Brominated diphenyl ethers in water, soil, sediment and tissue by HRGC/HRMS
. .
US Fish and Wildlife Service
.
2017
.
Environmental conservation online system
. .
Van den Brink
NW,
van Franeker
JA,
Ruiter-Dijkman
EM.
1998
.
Fluctuating concentrations of organochlorine pollutants during a breeding season in two Antarctic seabirds: Adelie penguin and southern fulmar
.
Environ Toxicol Chem
17
:
702
709
.
Van den Steen
E,
Poisbleau
M,
Demongin
L,
Covaci
A,
Dirtu
AC,
Pinxten
R,
van Noordwijk
HJ,
Quillfeldt
P,
Eens
M.
2011
.
Organohalogenated contaminants in eggs of rockhopper penguins (Eudyptes chrysocome) and imperial shags (Phalacrocorax atriceps) from the Falkland Islands
.
Sci Total Environ
409
:
2838
2844
.
Verreault
J,
Letcher
RJ,
Muir
DCG,
Chu
S,
Gebbink
WA,
Gabrielsen
GW.
2005
.
New organochlorine contaminants and metabolites in plasma and eggs of glaucous gulls (Larus hyperboreus) from the Norwegian Arctic
.
Environ Toxicol Chem
24
:
2486
2499
.
Wiemeyer
SN,
Bunck
CM,
Stafford
CJ.
1993
.
Environmental contaminants in bald eagle eggs—1980–84—and further interpretations of relationships to productivity and shell thickness
.
Arch Environ Contam Toxicol
24
:
213
227
.
Yang
D,
Lu
B,
Pan
JM.
2009
.
Distribution and ecological environmental significance of POPs in dropping-amended soil, organisms and eggs of Antarctic penguins
.
Mar Environ Sci
22
:
427
433
.
Yogui
GT,
Sericano
JL.
2009
.
Levels and pattern of polybrominated diphenyl ethers in eggs of Antarctic seabirds: Endemic versus migratory species
.
Environ Pollut
157
:
975
980
.