Article pubs.acs.org/est
Dietary versus Maternal Sources of Organochlorines in Top Predator Seabird Chicks: An Experimental Approach Sophie Bourgeon,†,#,* Eliza K. H. Leat,‡ Robert W. Furness,‡ Katrine Borgå,§ Sveinn Are Hanssen,† and Jan Ove Bustnes† †
Norwegian Institute for Nature Research (NINA), Fram Centre, 9296 Tromsø, Norway College of Medical, Veterinary and Life Sciences, Graham Kerr Building, University of Glasgow, Glasgow G12 8QQ, U.K. § Norwegian Institute for Water Research (NIVA), Gaustadalleén 21, 0349 Oslo, Norway ‡
S Supporting Information *
ABSTRACT: We examined the relative importance of dietary sources and maternal transfer on organochlorine concentrations (∑OCs) in Great skua chicks (Stercorarius skua) in Shetland by food supplementing parents with known wintering area. We predicted that experimental chicks (whose parents were supplemented) should have (i) higher growth rates and, (ii) lower ∑OCs due to growth dilution effect and/or due to being fed with less contaminated food compared to control chicks. We also predicted a significant influence of maternal wintering area on chicks’ ∑OCs. Plasma ∑OCs of adults, assessed prior to the manipulation, significantly differed between wintering areas of birds. Chicks were weighed every 5 days and plasma ∑OCs were assessed at 20 days old. Based on nitrogen and carbon stable isotope analysis, the supplementary food contributed on average 20% of the dietary protein of the chicks. Although experimental chicks experienced better developmental conditions, supplementary food did not alleviate their organochlorine burden. Nevertheless, chicks whose mothers wintered in Europe showed ∑OCs 50% higher than chicks whose mothers wintered in Africa. Moreover, based on the positive relationship between ∑OCs of chicks and females, the contaminant load of Great skua chicks in Shetland appears to be more influenced by maternal transfer than by trophic transfer.
■
INTRODUCTION Many environmental contaminants are transported over long distances and have become globally distributed.1,2 One of the major concerns about persistent organic pollutants (POPs) is that they have been reported to induce a variety of adverse biological effects in birds compromising their breeding performance and survival.3−6 It is therefore of high ecological concern to assess their concentrations in the environment. In that respect, birds represent useful tools to monitor the concentrations of POPs. More specifically, seabirds appear to be particularly suitable biomonitors as they feed at the top of the marine food web (high trophic position) through which pollutants biomagnify7−9 and because they nest in colonies where adults, eggs, and chicks are easily accessible.10 Moreover, many seabirds breed over large geographical scales enabling the spatial screening of pollutants. Intraspecies studies reported latitudinal gradients of pollutant exposure. For example, Great skuas (Stercorarius skua) breeding at different geographic locations showed organochlorine concentrations up to 7-fold higher in some colonies than in others.11 Nevertheless, since many seabirds breeding in the northern hemisphere migrate over long distances, some to more heavily industrialized areas,12,13 the interindividual variability reported in pollutant levels among adults at their breeding sites might not only reflect local pollution.14 Contrary to adults, chicks a priori appear as suitable models in which to monitor local © 2013 American Chemical Society
environmental pollution, since they are confined to the breeding grounds while growing, potentially restricting their pollutant profiles to one specific environment. Accordingly, studies that use eggs as matrices to record temporal trends in pollutants assume that eggs, providing that appropriate conditions are fulfilled, can reliably indicate pollution at the breeding site (i.e., local contamination).15,16 However, maternal transfer of contaminants to the eggs17,18 and the nestlings may be important; most of the POPs found in 15 day old great tit (Parus major) nestlings were of maternal origin.19 This calls for caution whenever using the latter matrices and questions the relevance of using chicks as sentinels for local environmental pollution, in particular in migrating seabirds. Given that the wintering area of Great skuas was recently shown to significantly affect both concentrations and patterns of organochlorines measured in plasma of adults after egg-laying at the breeding site,14 it is reasonable to assume that pollutants measured in eggs and chicks could also reflect the influence of the maternal wintering area. The concentrations of POPs in Great skua eggs from Shetland are among the highest in North Atlantic seabirds with Received: Revised: Accepted: Published: 5963
January 29, 2013 April 23, 2013 May 3, 2013 May 3, 2013 dx.doi.org/10.1021/es400442q | Environ. Sci. Technol. 2013, 47, 5963−5970
Environmental Science & Technology
Article
up to 7500 μg/kg wet weight sum PCB.20 The breeding ecology of the Great skua population of Shetland has been extensively documented over the last three decades and qualitative changes in the diet arising from a decrease in sandeel (Ammodytes marinus) availability were reported to constrain egg size,21 breeding success,22 and adult survival.21,23 In addition, stable isotope ratios in winter grown primary feathers and geolocation data,14,24 showed that adults wintered over continental shelf areas in one of two distinct areas, either off west Africa or off western Europe.14 In this context, the current study proposes an alternative and novel experimental approach to assess the influence of dietary and maternal sources of organochlorines in Great skua chicks from Shetland by manipulating the dietary sources and/or relieving the contaminant load of chicks by food supplementing the adults during chick rearing with a different type of food which was less contaminated than their natural diet. Thereafter, we measured plasma organochlorine (OC) concentrations in chicks and adults. Red blood cell (RBC) stable isotopes of nitrogen (δ15N) and carbon (δ13C) of chicks and adults were measured and were used to determine what proportion of the supplementary food was being fed to the chicks. We predicted that experimental chicks (chicks whose parents were supplemented during chick rearing) should have (i) higher body mass and growth rate, (ii) lower OC concentrations (as a result of the growth dilution effect and/or being fed with less contaminated food) compared to control chicks. We also predicted a significant influence of maternal wintering area on chicks’ OCs, regardless of treatment. This is the first field experimental approach carried out on seabird top predator chicks to investigate the influence of dietary and maternal sources on pollutant profiles.
adults overwintered in Europe (7 females, 5 males) and 63% in Africa (15 females, 5 males). 2.2. Supplementary Feeding Procedure. Around two days prior to hatching, 16 pairs were chosen to be given 100 g of beef meat cat food daily (experimental birds), whereas 15 pairs were not given any extra food but were visited at the same frequency to exert the same stress levels on all the birds (control birds). Cat food (Landlord, Reitangruppen, Germany) was distributed during day time from the moment when one egg was starting to crack, that is, around two days before hatching. In addition, for the first three days one raw hen’s egg was given at the same time as cat food to train the birds into taking the cat food. Beef cat food was chosen because it has a different stable isotope signature (terrestrial rather than marine) and is less contaminated in comparison to the natural diet of chicks. Contaminant analyses performed on the cat food batch used in the current experiment confirmed that OC compounds (the same as those measured in adults and chicks; see below) were all close to or under the detection limit of 10 μg/kg. One hundred grams of cat food represents on average 7.5% of the daily energy need of one pair of breeding Great skuas.26 We ensured that the laying date did not differ significantly between groups (t test, t = −1.18, p = 0.25) by assigning pairs with similar hatching dates to each group. The hatching success did not differ significantly between groups (t test, t = −0.87, p = 0.39) with at least one egg hatching in each nest. A posteriori, we ensured that adults from both groups were caught and blood sampled at a comparable incubation stage (t test, t = 0.95, p = 0.35) and that the proportion of birds that spent the winter in Africa or Europe was well balanced between groups (11 out of 17 control adults overwintered in Africa versus 9 out of 16 in the experimental group; Chi-square = 0.13, p = 0.72). In total, data are presented for 16 experimental adults (11 females and 5 males; 16 territories) and 17 control adults (12 females and 5 males; 15 territories). 2.3. Chicks. From hatching, chicks were weighed and body measures (tarsus and wing lengths) were recorded every 5 days up to day 20 when they reach the inflection point of mass growth.27 We kept track of the hatching rank, that is, the first hatched chick of the brood being referred to as the A-chick while the second hatched chick was referred to as the B-chick. At 20 days old, a blood sample was collected from chicks (see below point 3). The daily chick growth rate (g/day) was calculated between hatching and final sampling (20 days). Since chicks hide when a threat is approaching, they could not always be retrieved and measured at the designated stages. Nevertheless, whenever successfully located, the ages at which chicks were sampled did not differ significantly for any of the stages between experimental and control groups (GLMMs, 0.72 < p < 0.83). In total, data are shown for 23 experimental chicks (7 pairs of siblings; 15 A-chicks and 8 B-chicks) and 17 control chicks (2 pairs of siblings; 13 A-chicks and 4 B-chicks) aged 20 days. 3. Blood Sampling and Plasma Analysis. Blood was sampled from the tarsal vein (under Home Office licenses), collected in heparinized syringes and centrifuged within 2 h (2000 rpm for 5 min), with plasma and RBCs frozen and stored at −20 °C. Adult and chick plasma samples were subsequently used to assess organochlorine concentrations and RBCs were used to measure δ15N and δ13C, the turnover half-life of RBC stable isotopes being around 15 days.28 Adults and chicks were sexed from RBCs after DNA extraction and PCR amplification
■
EXPERIMENTAL SECTION 1. Study Species and Study Site. Great skuas are large top predator seabirds, with a female-biased sexual size dimorphism, that breed at colonies in the North-East Atlantic. Females lay up to two eggs and undertake most of the 29 days of incubation although both sexes contribute. Great skua chicks, semi altricial birds fed with regurgitated food provided by both parents but predominantly the male, fledge about 40 days after hatching. 2. Sampling Protocol and Experimental Design. 2.1. Adults. In 2009, 31 pairs of Great skuas were followed from incubation throughout chick rearing in Foula, Shetland, UK (60°08′N, 2°05′W). All pairs laid and incubated two eggs. Adult birds of either sex were caught on their nests while incubating using remote controlled noose traps, under appropriate licenses. Laying date and incubation stage (in days), at which birds were caught, was subsequently calculated using egg density25 and/or hatching dates. One adult per nest was caught for 29 pairs while both adults of the same pair were caught for two nests (N = 33 adults). Birds were caught on average on the 22th day of incubation (21.6 ± 0.9 days, N = 33 adults). After capture, birds were blood sampled (see below point 2.3) and body mass (±0.1 g), tarsus and wing lengths were recorded. Finally, birds were ringed with both metal and plastic color rings before release. The wintering quarters of 32 adults were inferred from primary feather stable isotopes14 and geolocation data-loggers (that record light levels twice per day) deployed on a few birds of the Shetland colony.24 In the current study, 37% of the 5964
dx.doi.org/10.1021/es400442q | Environ. Sci. Technol. 2013, 47, 5963−5970
Environmental Science & Technology
Article
Table 1. Influence of Supplementary Feeding (i.e., Treatment: Control versus Experimental), Sex and Wintering Area on Biological Profiles, Carbon and Nitrogen Red Blood Cell Stable Isotopes (RBC δ13C and δ15N) and Plasma Organochlorines (∑OCs) of Great Skua (Stercorarius skua) Adults during Incubation (i.e., Prior to Supplementary Feeding)a mean values ± SE control dependent variables during incubation tarsus length (mm) wing length (mm) body mass (g) RBC δ13C (‰) RBC δ15N (‰) ∑OCsb (μg/kg wet weight)
(N = 17) 70.3 418.9 1382.3 −18.4 12.8 312.9
± ± ± ± ± ±
0.5 1.6 24.0 0.1 0.1 79.4
results of GLMMs
experimental
treatment
(N = 16)
df
F value
p value
df
F value
p value
df
F value
p value
28.00 28.00 28.00 26.48 27.41 27.22
0.00 0.90 0.42 0.13 0.20 0.14
0.97 0.35 0.52 0.72 0.66 0.71
28.00 28.00 28.00 24.29 22.76 27.39
0.99 12.66 44.06 0.17 0.11 2.50
0.33 0.001
