- Email: [email protected]
Evaluation of different contour feather types for biomonitoring lead exposure in Northern goshawk (Accipiter gentilis) and tawny owl (Strix aluco)
Ecotoxicology and Environmental Safety 85 (2012) 115–119
Contents lists available at SciVerse ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Evaluation of different contour feather types for biomonitoring lead exposure in Northern goshawk (Accipiter gentilis) and tawny owl (Strix aluco) Sofı´a Debe´n, J. A´ngel Ferna´ndez, Jesu´s R. Aboal n, Alejo Carballeira Area de Ecologı´a, Departamento de Biologı´a Celular y Ecologı´a, Facultad de Biologı´a, Universidad de Santiago de Compostela, Spain
a r t i c l e i n f o
abstract
Article history: Received 15 December 2011 Received in revised form 20 July 2012 Accepted 2 August 2012 Available online 10 September 2012
The concentration of Pb was determined in feathers (contour feathers: mantle, pectoral, ventral, and primary- and secondary-coverts) of two sedentary species of raptors in Galicia (NW Spain): the tawny owl (Strix aluco) and the Northern goshawk (Accipiter gentilis). A high degree of intraindividual variability was observed in all types of feathers, with coefficients of variation exceeding 100 percent. The correlations between feather types were too low to enable use of a single type of body feather to predict the concentration of Pb in the other feathers. The number of body feathers required to differentiate individuals on the basis of the concentration of Pb was extremely high, in some cases higher than the number of the particular type of feather in the bird. All of this provides clear evidence that the contour and covert feathers of the raptor species considered cannot be used to biomonitor contamination by Pb, at least in this sample where the overall feather concentration were fairly uniform. & 2012 Published by Elsevier Inc.
Keywords: Raptors Pollution Goshawk Tawny owl Contamination
1. Introduction Environmental contamination by heavy metals is a worldwide problem, and represents a serious threat to the quality of the environment. The need to determine the degree of exposure and the effects of contaminants has led to numerous biomonitoring studies. Many of such studies are carried out with birds, which are particularly sensitive to anthropogenic contamination (see e.g. Burger and Gochfeld, 2009; Dauwe et al., 2000; Hanson and Jones, 1968; Jenkins, 1975). Raptor species are often used because the position of these birds at the top of the trophic chain leads to biomagnification, resulting in high tissue concentrations of contaminants (Dauwe et al., 2003; Denneman and Douben, 1993). Moreover, many raptors are highly territorial, non-migratory and are long lived, so that the levels of several heavy metals in different tissues may accurately reflect the chemical contamination that the birds have been exposed to throughout their lives (Hermoso de Mendoza et al., 2006). The use of feathers to measure contaminants such as heavy metals has several advantages over the use of other tissues (e.g. liver, kidneys, etc.) (Castro et al., 2011). Feathers contain high concentrations of some contaminants and the proportion in feathers, relative to the body load, is fairly constant (Burger and Gochfeld, 2009; Monteiro et al., 1998; Movalli, 2000). Furthermore, sampling
n
Corresponding author. Fax.: þ 34 981 59 69 04. E-mail address: [email protected] (J.R. Aboal).
0147-6513/$ - see front matter & 2012 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.ecoenv.2012.08.005
of feathers is simple and non-invasive; feathers can be collected from live birds, which is particularly important in the case of rare and declining species. Feathers can also be easily stored for decades as they do not require refrigeration (Gochfeld et al., 1996). In most previous studies, flight feathers (both primary remiges and rectrices) have been used (see e.g. Burger and Gochfeld, 2009; Denneman and Douben, 1993; Dmowski and Golimowski, 1993), although asymmetrical and incomplete moult of flight feathers may lead to problems in the interpretation of results (Jaspers et al., 2007). Moreover, the removal of flight feathers would lead to reduced fitness (Vagasi et al., 2011). On the other hand, some authors have indicated the potential utility of body feathers (Burger and Gochfeld, 2009) although information on its use as indicators of contamination is scarce. These feathers can be plucked from live birds with almost no effect, so that large samples can be obtained (Furness and Greenwood, 1993), but probably the problem of incomplete moulting could also affect. Body feathers from individual specimens of seabirds are often used to represent the levels of contaminants in the colonies under study (see e.g. Burger and Gochfeld, 2000, 2009; Movalli, 2000). Nevertheless when raptors are studied, the data are referred to single individuals instead to colonies, so that the samples should adequately represent the body levels of contaminants in each specimen. Lead is one of the most toxic heavy metals and may reach the environment in many forms, mainly as a result of human activities related to industry, mining and hunting. In the case of raptors, the main route is the ingestion of lead shot present in unretrieved, killed or wounded ‘game’. The persistence of lead in
116
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
the environment, its accumulation in organisms and some degree of biomagnification in trophic chains implies that the raptor species are continually exposed to low doses of this metal (Locke and Friend, 1992; Martı´nez-Lo´pez et al., 2004). The aim of the present study was to establish whether the body feathers (mantle, pectoral, ventral, and primary and secondary coverts) of two species of raptors (Strix aluco and Accipiter gentilis) can be used to biomonitor contamination by Pb. For this purpose, the representativeness of these feathers in relation to metal uptake in individual birds was determined and the optimal sample size (number of feathers) was established by study of the variability in the concentrations of Pb in each type of feather.
2. Material and methods 2.1. Obtaining and processing the feathers The birds used in the study of adult tawny owls (S. aluco) (n¼ 20) and adult Northern goshawks (A. gentilis) (n¼ 10) were donated by Wildlife Recovery Centres (WRCs) in Galicia (NW Spain). These species are considered suitable for biomonitoring studies as they are usually sedentary within the region and capture a wide variety of prey (Pe´rez-Lo´pez et al., 2008). The birds had been sent to the WRCs with different injuries (never by lead shot) and had died as a result of these or had been sacrificed because they were not expected to recover. None of the birds sampled had been held in the WRCs for more than one month. The dead birds were stored in the WRCs in hermetically sealed plastic bags at 30 1C, to protect the feathers from attack by insects, secondary contamination and decomposition (Hahn et al., 1993). The bodies were transported to the laboratory in portable cool boxes and stored again at 30 1C until analysis. For analysis, the birds were defrosted at room temperature, and the body feathers were removed; any growing (immature) or incomplete feathers were discarded as they do not provide a representative picture of the final metal uptake in the feather (Denneman and Douben, 1993). All of the primary- and secondarycovert feathers were removed from the left wing of each bird, whereas samples of contour feathers like pectoral, mantle and ventral feathers were collected from all over the corresponding body part. The feathers were washed in an ultrasound bath (Branson 5200) with bidistilled water and 5 mL of detergent (Triton X-100), for 15 min. (Dmowski and Golimowski, 1993; Hudges et al., 1997), to remove any substances adhering to the surface (dust, parasites, etc.). The feathers were then soaked in bidistilled water to remove the detergent, and dried in a forced air oven at 45 1C for 24 h. Barbs were removed from the primary- and secondary-covert feathers for analysis, and the rachis was discarded. The down feathers from the basal zone of the ventral feathers were discarded, and the remaining part was analysed. Whole pectoral and mantle feathers were analysed.
2.2. Chemical analysis Each sample comprised two feathers of each type, in order to obtain the minimum weight required to reach the limit of quantification of the Pb content. For ventral and covert feathers ten samples of each type were analysed for each individual, whereas for pectoral and mantle feathers twenty samples were analysed. Each sample was digested in 2 mL of HNO3 (65 percent, analytical quality) and were placed on a heating plate at 90 1C until no recognisable material remained (approx. 48 h). Once the samples were totally digested, they were allowed to cool for 1 h, then 8 mL of bidistilled water was added. The Pb contents of the extracts were determined by graphite chamber atomic adsorption spectrometry (Perkin Elmer AAnalyst 600). To control the analytical quality of the process, blanks and certified reference material (GBW 07601, human hair—Institute for Geophysical and Geochemical Exploration, Langfang, China) were analysed, once every ten samples. To control the effectiveness of the acid digestion procedure, the percentage recovery of the reference material was calculated (111 percent). The limit of quantification (LQ) was 3.87 ng g 1; those samples with concentrations of metals below this limit (12 percent of the total) were assigned a value corresponding to half of the limit.
2.3. Statistical analysis The Shapiro–Wilk test was applied to test the normality of the data. The Spearman rho correlation was used to determine the association between the concentrations of Pb in different types of feathers. In order to estimate the minimum number of body feathers required to differentiate individuals on the basis of the concentration of Pb, a priori power analysis was carried out (Cochran and Cox, 1957; Zar, 1984). For all possible pairs
of individuals of the same species and for each type of feather, the number of samples (n) required to determine the minimal detectable difference that enables differentiation of the mean concentrations of Pb in each type of feather sampled from each individual was calculated. These values were calculated for differences in concentration of Pb of 1, 2, 3,y,8 mg g 1. Once the n values were obtained, the mean values and corresponding confidence intervals (95 percent) were calculated for each species and type of feather. Finally, the number of samples (pairs of feathers) of each type that must be collected from an individual for a level of 10 percent accuracy in estimating the corresponding mean value was calculated (Morris, 1954).
3. Results The mean values and coefficients of variation for each individual bird specimen and type of feather are shown in Table 1, in which the concentrations of Pb in each type of feather and individual that belong to normal distributions are underlined. The number of cases in which the data were normally distributed was very similar in both species, except when considering the primary covert feathers. For pectoral and mantle feathers, very few of the data sets were normally distributed (approx. 10 and 20 percent respectively), whereas for the other types of feathers, 50–60 percent of the data sets were normally distributed, except for those corresponding to the primary feathers in A. gentilis (26 percent). For all types of feather, the mean concentration of Pb (mg g 1) was higher in S. aluco than in A. gentilis: 5.35 vs. 4.56 mg g 1 respectively for pectoral feathers, 4.74 vs. 4.47 for mantle feathers, 4.98 vs. 2.80 for ventral feathers, 6.74 vs. 3.64 for primary coverts, and 7.67 vs. 3.30 for secondary coverts. The coefficients of variation (CV) for each individual and type of feather reflect a high degree of variability in the concentrations of Pb in both S. aluco and A. gentilis. The mean CVs were higher in Table 1 Mean concentrations of Pb (mg g 1) and coefficients of variation (%) for each individual bird specimen and type of feather analysed (n¼10 for ventral, primary and secondary covert feathers and n¼ 20 for pectoral and mantle feathers). CV: coefficient of variation; A: Accipiter gentilis; C: Strix aluco. Values that are underlined belong to a normal distribution. Type of feather Pectoral individual Mean CV
Mantle Mean CV
Ventral Mean CV
Primary Mean CV
Secondary Mean CV
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 A1 A2 A3 A4 A5 A6 A7 A8 A9 A10
7.05 6.96 6.50 7.57 8.04 5.95 2.91 3.97 3.80 2.93 1.90 5.65 4.08 3.73 3.56 6.80 1.65 3.13 5.61 2.97 3.35 3.70 5.50 3.12 3.24 2.80 – 12.0 3.12 3.36
6.97 9.41 2.14 7.85 6.52 3.48 2.95 12.0 4.45 3.32 1.90 3.95 – 3.04 4.86 3.58 6.63 4.13 4.34 3.04 1.11 4.28 1.98 2.85 3.16 2.36 2.87 – 4.44 2.13
6.81 23.0 2.85 4.13 29.9 5.10 3.87 4.75 5.69 2.16 2.85 6.73 – 2.73 2.80 9.42 2.90 4.98 2.98 4.41 2.80 5.04 2.19 4.75 3.24 – 2.27 2.81 6.78 2.85
12.1 31.1 3.05 5.20 41.4 3.09 1.93 1.12 3.65 5.20 3.21 2.67 3.19 3.48 10.2 3.51 5.15 2.36 – 4.12 1.26 4.67 1.27 3.07 6.67 4.04 2.81 3.02 3.84 2.34
8.07 7.00 5.92 5.37 9.55 3.99 2.90 3.48 4.51 2.02 1.95 – 6.69 3.72 9.51 5.43 6.68 5.45 6.58 2.83 3.36 4.56 4.80 3.09 4.74 – 2.46 9.83 – 3.60
44 32 62 49 129 47 8 50 64 67 62 – 177 43 118 93 49 58 65 17 60 14 69 136 80 – 13 100 – 24
28 54 152 49 64 66 7 115 25 70 40 127 56 46 29 31 104 39 133 44 81 33 178 6 42 10 – 70 14 13
23 41 32 46 51 28 10 93 21 92 37 32 – 10 60 28 134 47 67 27 31 79 41 19 52 19 54 – 41 43
65 62 39 50 49 28 54 111 98 26 17 78 – 16 25 40 18 35 20 47 42 78 38 48 48 – 32 25 120 28
60 25 84 48 40 11 17 18 11 102 45 3 16 50 17 30 48 18 – 21 14 52 24 1 77 91 34 36 63 61
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
1500
117
800
Primary-coverts
Secondarycoverts
600
1000
400
Number of samples
500
200
0
0
1600
3000
Ventral
Pectoral
1200
2000
800 1000
400 0
0
0
2
4
6
8 1200
Mantle 800
400
0 0
2
4
6
8
-1
Pb concentration difference ( g g ) 1200 900
Number of samples
200
Primarycoverts
Secondarycoverts
150
600
100
300
50
0
0 200
1000
Ventral
Pectoral 750
150
500
100
250
50
0
0 0
2
4
6
8 1200
Mantle
900 600 300 0 0
4
2
6
8
-1
Pb concentration difference ( g g ) Fig. 1. (a). Number of samples (n) of each type of feather required to differentiate the concentrations of Pb in pairs of individual specimens of Strix aluco. The dashed arrows indicate the n values corresponding to the mean (crosses), the lower limit (circles) and the upper limit (triangles) of the confidence interval for the modal value (solid arrow on the x axis) of the distribution of the differences in the mean concentrations of Pb between each individual, for each pair of feathers analysed. (b). Number of samples (n) of each type of feather required to differentiate the concentrations of Pb in pairs of individual specimens of Accipiter gentillis. The dashed arrows indicate the n values corresponding to the mean and the upper limit of the confidence interval for the modal value (solid arrow on the x axis) of the distribution of the differences in the mean concentrations of Pb between each individual, for each pair of feathers analysed.
118
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
S. aluco for all types of feather, except the covert feathers, whereas the highest CV in both species corresponded to the pectoral feathers. There was a significant correlation between the concentrations of Pb in the pectoral feathers and the mantle feathers for both A. gentilis (r ¼0.786n) and S. aluco (r ¼0.546n). In the case of S. aluco there were also significant correlations for the following comparisons: pectoral–ventral feathers (r ¼0.569nn), pectoral– secondary feathers (r¼ 0.498n) and mantle–primary coverts feathers (r ¼0.651nn). In A. gentilis significant correlations were found for: ventral–primary coverts feathers (r ¼0.762n), ventral– secondary coverts feathers (r ¼0.783nn) and primary–secondary coverts feathers (r ¼0.767n). Fig. 1a and b shows, for S. aluco and A. gentilis respectively, the result of the power analysis carried out to determine the number of samples (pairs of feathers) of each type that are needed to differentiate the concentrations of Pb corresponding to the pairs of individuals of the same species. In these figures, for each type of feather, arrows and dashed lines are used to indicate the n value corresponding to the mean value and the upper limit of the confidence interval for the difference in concentration most frequently observed (indicated with a solid arrow on the x axis). This most frequently observed concentration was selected using the modal value of the distribution of the differences in the mean concentrations of Pb in each pair of feathers analysed was selected. The minimum number of feathers is not shown because in most cases it was less than one. For S. aluco, an average of 150 pairs of primary covert feathers, 70 pairs of secondary covert feathers, 245 pairs of pectoral feathers, 320 pairs of ventral feathers and 380 pairs of mantle feathers were required, whereas for A. gentilis, 320 pairs of primary coverts feathers, 55 pairs of secondary coverts feathers, 230 pairs of pectoral feathers, 51 ventral feathers and 240 pairs of mantle feathers were required. As regards the number of samples required for a 10 percent level of accuracy in estimating the mean concentration of Pb in each type of feather, in the case of S. aluco, this was 432 for pectoral feathers, 360 for mantle feathers, 612 for ventral feathers, 3460 for primary coverts feathers and 6419 secondary coverts feathers. In A. gentilis, the number of feathers required was 645 for pectoral feathers, 1014 for mantle feathers, 219 for ventral feathers, 351 for primary coverts and 408 for secondary coverts feathers. The number of pairs of pectoral and mantle feathers required was higher in A. gentili, whereas more ventral, primary and secondary coverts feathers were required in S. aluco.
Falco biarmicus jugger (1.56 mg g 1) and those obtained by Battaglia et al. (2005) on analysing the mantle feathers of the Little Owl Athene noctua and the Common Buzzard Buteo buteo (19.1 and 8.87 mg g 1 respectively). The fact that the specimens were donated by Wildlife Recovery Centres enabled us to work with a much large number of specimens than in the previously cited studies. Measuring the concentrations of contaminants in a non invasive manner in wild animals is a problem that has not been fully resolved. Given the ethical and legal impediments associated with the hunting and killing of raptor species, the best option for obtaining samples is by donations of dead specimens from Wildlife Recovery Centres (Battaglia et al., 2005; Castro et al., 2011; Pe´rezLo´pez et al., 2008). The use of body feathers is another possible solution to this problem, as removal of these feathers does not injure the birds, so that they could be obtained from birds captured for this purpose then released, or from birds held in Wildlife Recovery Centres. As Burger and Gochfeld (2009) have recently indicated in reference to seabirds, body feathers and pectoral feathers are especially useful for establishing spatial and temporal patterns in bird populations and for evaluating heavy metal contamination in species that are threatened or in danger of extinction. However, in the present study, the coefficients of variation for the concentrations of Pb in pectoral feathers were higher than for those in the other body feathers analysed: 65 percent in S. aluco and 63 percent in A. gentilis. The variability in the concentrations in all types of feathers in both S. aluco and A. gentilis was enormous, despite the fact that: (i) each sample comprised two feathers, which reduces some of the variability between feathers of the same type; (ii) for some samples a value of half the limit of quantification was assigned, which also reduces the intra and inter individual variability, and (iii) in some cases the distributions were not normal, so that the power analysis may underestimate the number of samples (pairs of feathers) required (Gonza´lez et al., 2006). Moreover, if the body feathers were a good indicator of the levels of Pb in the blood, there should be a close correlation between the concentrations in the different types of feather, as all are related to the levels in blood. Although some of the correlations were significant, the regression coefficients were generally low, so that we can reject the possibility of carrying out regression with one type of feather as a predictor of the concentrations of Pb in the other type of feathers.
5. Conclusions 4. Discussion Information about the levels of heavy metals in contour and coverts feathers of birds is generally scarce, and mainly refers to seabirds. Very little data has been published as regards to the levels of Pb in the body feathers of raptor species. Comparison of the concentrations of Pb in the pectoral feathers of the birds examined in the present study (Table 1) with those reported in studies with seabirds, shows that the levels of Pb in the raptor species is much higher. Burger and Gochfeld (2000, 2009) reported mean concentrations of Pb of less than 2 mg g 1 in seabirds, and Gochfeld et al. (1996) reported concentrations of 0.003 mg g 1 in pectoral feathers in the Laughing gull Larus atricilla. These differences may be attributed to the different trophic levels of the species considered. Studies carried out by Garcı´a-Ferna´ndez et al. (1996) indicate higher tissue concentrations of Pb in carnivorous birds (nocturnal and diurnal raptors) than in fish-eating, insectivorous, omnivorous and carrion-eating birds, probably because some of these species live in areas where there is little human activity. The results obtained in the present study are in a range intermediate between the concentrations obtained by Movalli (2000) in pectoral feathers of the falcon
Because of the high intraindividual variability in the concentrations of Pb in each type of feather, with coefficients of variation of up to 100 percent, an extremely large number of feathers was required for accurate estimation of the mean concentration, and in some cases was even greater than the number of that particular type of feather in the birds. In order to differentiate the mean concentrations in a particular type of feather between different birds, the number of feathers required was again extremely high, and would be impossible to obtain (either because the birds do not have sufficient numbers of a particular type of feather or because of the stress that would be generated by removing a large number of feathers). All of this provides clear evidence that the contour and coverts feathers of the raptor species considered cannot be used to biomonitor contamination by Pb.
Acknowledgments We thank all personnel at the Wildlife Recovery Centres involved, and also Jesu´s Santamarina and Marta Prieto Rodrı´guez from the Biodiversity Conservation Service, Xunta de Galicia, for
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
their help in obtaining samples. Thanks are due to Michael Gochfeld and one anonymous reviewer for their value comments and suggestions. This study was partly funded by projects awarded by the Xunta de Galicia ‘‘Banco de Especı´menes Ambiena tales de Galicia, 3 % Fase’’ and the ‘‘INCITE 2009/PX166’’. References Battaglia, A., Ghidini, S., Campanini, G., Spaggiari, R., 2005. Heavy metal contamination in little owl (Athene noctua) and common buzzard (Buteo buteo) from northern Italy. Ecotoxicol. Environ. Saf 60, 61–66. Burger, J., Gochfeld, M., 2000. Metal levels in feathers of 12 species of seabirds from Midway Atoll in the northern Pacific Ocean. Sci. Total Environ. 257, 37–52. Burger, J., Gochfeld, M, 2009. Comparison of arsenic, cadmium, chromium, lead, manganese, mercury and selenium in feathers in bald eagle (Haliaeetus leucocephalus), and comparison with common eider (Somateria mollissima), glaucous-winged gull (Larus glaucescens), pigeon guillemot (Cepphus columba), and tufted puffin (Fratercula cirrhata) from the Aleutian chain of Alaska. Environ. Monit. Assess 152, 357–367. Castro, I., Aboal, J.R., Ferna´ndez, J.A., Carballeira, A., 2011. Use of raptors for biomonitoring of heavy metals: gender, age and tissue selection. Bull. Environ. Contam. Toxicol. 86, 347–351. Cochran, W.G., Cox, G.M., 1957. Experimental Designs. Wiley, New York. Dauwe, T., Bervoets, L., Blust, R., Pinxten, R., Eens, M., 2000. Can excrement and feathers of nestling songbirds be used as biomonitors for heavy metal pollution? Arch. Environ. Contam. Toxicol 39, 541–546. Dauwe, T., Bervoets, L., Blust, R., Pinxten, R., Eens, M., 2003. Variation of heavy metals within and among feathers of birds of prey: effects of molt and external contamination. Environ. Pollut. 124, 429–436. Denneman, W.D., Douben, P.E., 1993. Trace metals in primary feathers of the barn owl (Tyto alba guttatus) in the Netherlands. Environ. Pollut 82, 301–310. Dmowski, K., Golimowski, J., 1993. Feathers of the magpie (Pica pica) as a bioindicator material for heavy metal pollution assessment. Sci. Total. Environ 139–140, 251–258. Furness, R.W., Greenwood, J.J.D., 1993. Birds as Monitors of Environmental Change. Chapman and Hall, London. Garcı´a-Ferna´ndez, A.J., Motas-Guzma´n, M., Navas, I., Marı´a-Mojica, P., Luna, A., Sa´nchez-Garcı´a, J.A., 1996. Environmental exposure and distribution of lead in four species of raptors in Southeastern Spain. Arch. Environ. Contam. Toxicol. 33, 76–82. Gochfeld, M., Belant, J.L., Shukla, T., Benson, T., Burger, J., 1996. Heavy metals in laughing gulls: gender, age and tissue differences. Environ. Toxicol. Chem. 15, 2275–2283.
119
Gonza´lez, X.I., Aboal, J.R., Ferna´ndez, J.A., Carballeira, A., 2006. Considerations on the sample size of wood mice used to biomonitor metals. Sci. Total Environ. 366, 910–914. Hahn, E., Hahn, K., Stoeppler, M., 1993. Bird feathers as bioindicators in areas of the German Environmental Specimen Bank: bioaccumulation of mercury in food chains and exogenous deposition of atmospheric pollution with lead and cadmium. Sci. Total Environ 139–140, 259–270. Hanson, H.C., Jones, R.L., 1968. Use the Feathers Minerals as Biological Tracers to Determine the Breeding and Moulting Ground of Wild Geese. State of Illinois Dept Reg Educ Nat Hist Surv Div, Biological Note, 60 pp. Hermoso de Mendoza, M., Soler Rodrı´guez, F., Herna´ndez Moreno, D., Gallego Rodrı´guez, M.E., Lo´pez Beceiro, A., Pe´rez Lo´pez, M., 2006. Comparative study of heavy metal and metalloid hepatic levels in diurnal raptors from Galicia and Extremadura. Rev. Toxicol 23, 138–145. Hudges, K.D., Ewins, P.J., Clark, K.E., 1997. A comparison of mercury levels in feathers and eggs of osprey (Pandion haliaetus) in the North American Great Lakes. Arch. Environ. Contam. Toxicol 33, 441–452. Jaspers, V.L., Covaci, A., Van den Steen, E., Eens, M., 2007. Is external contamination with organic pollutants important for concentrations measured in bird feathers? Environ. Int. 33, 766–772. Jenkins, C., 1975. Utilisation du pigeon biset (Columba livia Gm.) comme temoin de la pollution atmospherique par le plomb. C.R. Acad. Sci. Ser. D 281, 1187–1189. Locke, L.N., Friend, M., 1992. Lead poisoning of avian species other than waterfowl. In: Pain, D.J. (Ed.), Lead Poisoning in Waterfowl. The International Waterfowl and Wetland Research Bureau, Slimbridge, UK, pp. 19–22. Martı´nez-Lo´pez, E., Martı´nez, J.E., Marı´a-Mojica, P., Pen˜alver, J., Pulido, M., Calvo, J.F., Garcı´a-Ferna´ndez, A.J., 2004. Lead in feathers and d-aminolevulinic acid dehydratase activity in three raptors species from an unpolluted Mediterranean Forest (Southeastern Spain). Arch. Environ. Contam. Toxicol. 47, 270–275. Monteiro, L.R., Granadeiro, J.P., Furness, R.W., 1998. Relationship between mercury levels and diet in Azores seabirds. Mar. Ecol. Prog. Ser. 166, 259–265. Morris, R.F., 1954. A sequential sampling technique for spruce budworm egg surveys. Can. J. Zool. 32, 302–313. Movalli, P.A., 2000. Heavy metal and other residues in feathers of laggar falcon Falco biarmicus jugger from six districts of Pakistan. Environ. Pollut. 109, 267–275. Pe´rez-Lo´pez, M., Hermoso de Mendoza, M., Lo´pez Beceiro, A., Soler Rodriguez, F., 2008. Heavy metal (Cd, Pb, Zn) and metalloid (As) content in raptor species from Galicia (NW Spain). Ecotoxicol. Environ. Saf 70, 154–162. Vagasi, C.I., Pap, P.L., Tokolyi, J., Szekely, E., Barta, Z., 2011. Correlates of variation in flight feather quality in the Great Tit Parus major. Ardea 99, 53–60. Zar, J.H., 1984. Bioestatistical analysis, fourth ed. Prentice-Hall International Editions, London.
Contents lists available at SciVerse ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Evaluation of different contour feather types for biomonitoring lead exposure in Northern goshawk (Accipiter gentilis) and tawny owl (Strix aluco) Sofı´a Debe´n, J. A´ngel Ferna´ndez, Jesu´s R. Aboal n, Alejo Carballeira Area de Ecologı´a, Departamento de Biologı´a Celular y Ecologı´a, Facultad de Biologı´a, Universidad de Santiago de Compostela, Spain
a r t i c l e i n f o
abstract
Article history: Received 15 December 2011 Received in revised form 20 July 2012 Accepted 2 August 2012 Available online 10 September 2012
The concentration of Pb was determined in feathers (contour feathers: mantle, pectoral, ventral, and primary- and secondary-coverts) of two sedentary species of raptors in Galicia (NW Spain): the tawny owl (Strix aluco) and the Northern goshawk (Accipiter gentilis). A high degree of intraindividual variability was observed in all types of feathers, with coefficients of variation exceeding 100 percent. The correlations between feather types were too low to enable use of a single type of body feather to predict the concentration of Pb in the other feathers. The number of body feathers required to differentiate individuals on the basis of the concentration of Pb was extremely high, in some cases higher than the number of the particular type of feather in the bird. All of this provides clear evidence that the contour and covert feathers of the raptor species considered cannot be used to biomonitor contamination by Pb, at least in this sample where the overall feather concentration were fairly uniform. & 2012 Published by Elsevier Inc.
Keywords: Raptors Pollution Goshawk Tawny owl Contamination
1. Introduction Environmental contamination by heavy metals is a worldwide problem, and represents a serious threat to the quality of the environment. The need to determine the degree of exposure and the effects of contaminants has led to numerous biomonitoring studies. Many of such studies are carried out with birds, which are particularly sensitive to anthropogenic contamination (see e.g. Burger and Gochfeld, 2009; Dauwe et al., 2000; Hanson and Jones, 1968; Jenkins, 1975). Raptor species are often used because the position of these birds at the top of the trophic chain leads to biomagnification, resulting in high tissue concentrations of contaminants (Dauwe et al., 2003; Denneman and Douben, 1993). Moreover, many raptors are highly territorial, non-migratory and are long lived, so that the levels of several heavy metals in different tissues may accurately reflect the chemical contamination that the birds have been exposed to throughout their lives (Hermoso de Mendoza et al., 2006). The use of feathers to measure contaminants such as heavy metals has several advantages over the use of other tissues (e.g. liver, kidneys, etc.) (Castro et al., 2011). Feathers contain high concentrations of some contaminants and the proportion in feathers, relative to the body load, is fairly constant (Burger and Gochfeld, 2009; Monteiro et al., 1998; Movalli, 2000). Furthermore, sampling
n
Corresponding author. Fax.: þ 34 981 59 69 04. E-mail address: [email protected] (J.R. Aboal).
0147-6513/$ - see front matter & 2012 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.ecoenv.2012.08.005
of feathers is simple and non-invasive; feathers can be collected from live birds, which is particularly important in the case of rare and declining species. Feathers can also be easily stored for decades as they do not require refrigeration (Gochfeld et al., 1996). In most previous studies, flight feathers (both primary remiges and rectrices) have been used (see e.g. Burger and Gochfeld, 2009; Denneman and Douben, 1993; Dmowski and Golimowski, 1993), although asymmetrical and incomplete moult of flight feathers may lead to problems in the interpretation of results (Jaspers et al., 2007). Moreover, the removal of flight feathers would lead to reduced fitness (Vagasi et al., 2011). On the other hand, some authors have indicated the potential utility of body feathers (Burger and Gochfeld, 2009) although information on its use as indicators of contamination is scarce. These feathers can be plucked from live birds with almost no effect, so that large samples can be obtained (Furness and Greenwood, 1993), but probably the problem of incomplete moulting could also affect. Body feathers from individual specimens of seabirds are often used to represent the levels of contaminants in the colonies under study (see e.g. Burger and Gochfeld, 2000, 2009; Movalli, 2000). Nevertheless when raptors are studied, the data are referred to single individuals instead to colonies, so that the samples should adequately represent the body levels of contaminants in each specimen. Lead is one of the most toxic heavy metals and may reach the environment in many forms, mainly as a result of human activities related to industry, mining and hunting. In the case of raptors, the main route is the ingestion of lead shot present in unretrieved, killed or wounded ‘game’. The persistence of lead in
116
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
the environment, its accumulation in organisms and some degree of biomagnification in trophic chains implies that the raptor species are continually exposed to low doses of this metal (Locke and Friend, 1992; Martı´nez-Lo´pez et al., 2004). The aim of the present study was to establish whether the body feathers (mantle, pectoral, ventral, and primary and secondary coverts) of two species of raptors (Strix aluco and Accipiter gentilis) can be used to biomonitor contamination by Pb. For this purpose, the representativeness of these feathers in relation to metal uptake in individual birds was determined and the optimal sample size (number of feathers) was established by study of the variability in the concentrations of Pb in each type of feather.
2. Material and methods 2.1. Obtaining and processing the feathers The birds used in the study of adult tawny owls (S. aluco) (n¼ 20) and adult Northern goshawks (A. gentilis) (n¼ 10) were donated by Wildlife Recovery Centres (WRCs) in Galicia (NW Spain). These species are considered suitable for biomonitoring studies as they are usually sedentary within the region and capture a wide variety of prey (Pe´rez-Lo´pez et al., 2008). The birds had been sent to the WRCs with different injuries (never by lead shot) and had died as a result of these or had been sacrificed because they were not expected to recover. None of the birds sampled had been held in the WRCs for more than one month. The dead birds were stored in the WRCs in hermetically sealed plastic bags at 30 1C, to protect the feathers from attack by insects, secondary contamination and decomposition (Hahn et al., 1993). The bodies were transported to the laboratory in portable cool boxes and stored again at 30 1C until analysis. For analysis, the birds were defrosted at room temperature, and the body feathers were removed; any growing (immature) or incomplete feathers were discarded as they do not provide a representative picture of the final metal uptake in the feather (Denneman and Douben, 1993). All of the primary- and secondarycovert feathers were removed from the left wing of each bird, whereas samples of contour feathers like pectoral, mantle and ventral feathers were collected from all over the corresponding body part. The feathers were washed in an ultrasound bath (Branson 5200) with bidistilled water and 5 mL of detergent (Triton X-100), for 15 min. (Dmowski and Golimowski, 1993; Hudges et al., 1997), to remove any substances adhering to the surface (dust, parasites, etc.). The feathers were then soaked in bidistilled water to remove the detergent, and dried in a forced air oven at 45 1C for 24 h. Barbs were removed from the primary- and secondary-covert feathers for analysis, and the rachis was discarded. The down feathers from the basal zone of the ventral feathers were discarded, and the remaining part was analysed. Whole pectoral and mantle feathers were analysed.
2.2. Chemical analysis Each sample comprised two feathers of each type, in order to obtain the minimum weight required to reach the limit of quantification of the Pb content. For ventral and covert feathers ten samples of each type were analysed for each individual, whereas for pectoral and mantle feathers twenty samples were analysed. Each sample was digested in 2 mL of HNO3 (65 percent, analytical quality) and were placed on a heating plate at 90 1C until no recognisable material remained (approx. 48 h). Once the samples were totally digested, they were allowed to cool for 1 h, then 8 mL of bidistilled water was added. The Pb contents of the extracts were determined by graphite chamber atomic adsorption spectrometry (Perkin Elmer AAnalyst 600). To control the analytical quality of the process, blanks and certified reference material (GBW 07601, human hair—Institute for Geophysical and Geochemical Exploration, Langfang, China) were analysed, once every ten samples. To control the effectiveness of the acid digestion procedure, the percentage recovery of the reference material was calculated (111 percent). The limit of quantification (LQ) was 3.87 ng g 1; those samples with concentrations of metals below this limit (12 percent of the total) were assigned a value corresponding to half of the limit.
2.3. Statistical analysis The Shapiro–Wilk test was applied to test the normality of the data. The Spearman rho correlation was used to determine the association between the concentrations of Pb in different types of feathers. In order to estimate the minimum number of body feathers required to differentiate individuals on the basis of the concentration of Pb, a priori power analysis was carried out (Cochran and Cox, 1957; Zar, 1984). For all possible pairs
of individuals of the same species and for each type of feather, the number of samples (n) required to determine the minimal detectable difference that enables differentiation of the mean concentrations of Pb in each type of feather sampled from each individual was calculated. These values were calculated for differences in concentration of Pb of 1, 2, 3,y,8 mg g 1. Once the n values were obtained, the mean values and corresponding confidence intervals (95 percent) were calculated for each species and type of feather. Finally, the number of samples (pairs of feathers) of each type that must be collected from an individual for a level of 10 percent accuracy in estimating the corresponding mean value was calculated (Morris, 1954).
3. Results The mean values and coefficients of variation for each individual bird specimen and type of feather are shown in Table 1, in which the concentrations of Pb in each type of feather and individual that belong to normal distributions are underlined. The number of cases in which the data were normally distributed was very similar in both species, except when considering the primary covert feathers. For pectoral and mantle feathers, very few of the data sets were normally distributed (approx. 10 and 20 percent respectively), whereas for the other types of feathers, 50–60 percent of the data sets were normally distributed, except for those corresponding to the primary feathers in A. gentilis (26 percent). For all types of feather, the mean concentration of Pb (mg g 1) was higher in S. aluco than in A. gentilis: 5.35 vs. 4.56 mg g 1 respectively for pectoral feathers, 4.74 vs. 4.47 for mantle feathers, 4.98 vs. 2.80 for ventral feathers, 6.74 vs. 3.64 for primary coverts, and 7.67 vs. 3.30 for secondary coverts. The coefficients of variation (CV) for each individual and type of feather reflect a high degree of variability in the concentrations of Pb in both S. aluco and A. gentilis. The mean CVs were higher in Table 1 Mean concentrations of Pb (mg g 1) and coefficients of variation (%) for each individual bird specimen and type of feather analysed (n¼10 for ventral, primary and secondary covert feathers and n¼ 20 for pectoral and mantle feathers). CV: coefficient of variation; A: Accipiter gentilis; C: Strix aluco. Values that are underlined belong to a normal distribution. Type of feather Pectoral individual Mean CV
Mantle Mean CV
Ventral Mean CV
Primary Mean CV
Secondary Mean CV
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 A1 A2 A3 A4 A5 A6 A7 A8 A9 A10
7.05 6.96 6.50 7.57 8.04 5.95 2.91 3.97 3.80 2.93 1.90 5.65 4.08 3.73 3.56 6.80 1.65 3.13 5.61 2.97 3.35 3.70 5.50 3.12 3.24 2.80 – 12.0 3.12 3.36
6.97 9.41 2.14 7.85 6.52 3.48 2.95 12.0 4.45 3.32 1.90 3.95 – 3.04 4.86 3.58 6.63 4.13 4.34 3.04 1.11 4.28 1.98 2.85 3.16 2.36 2.87 – 4.44 2.13
6.81 23.0 2.85 4.13 29.9 5.10 3.87 4.75 5.69 2.16 2.85 6.73 – 2.73 2.80 9.42 2.90 4.98 2.98 4.41 2.80 5.04 2.19 4.75 3.24 – 2.27 2.81 6.78 2.85
12.1 31.1 3.05 5.20 41.4 3.09 1.93 1.12 3.65 5.20 3.21 2.67 3.19 3.48 10.2 3.51 5.15 2.36 – 4.12 1.26 4.67 1.27 3.07 6.67 4.04 2.81 3.02 3.84 2.34
8.07 7.00 5.92 5.37 9.55 3.99 2.90 3.48 4.51 2.02 1.95 – 6.69 3.72 9.51 5.43 6.68 5.45 6.58 2.83 3.36 4.56 4.80 3.09 4.74 – 2.46 9.83 – 3.60
44 32 62 49 129 47 8 50 64 67 62 – 177 43 118 93 49 58 65 17 60 14 69 136 80 – 13 100 – 24
28 54 152 49 64 66 7 115 25 70 40 127 56 46 29 31 104 39 133 44 81 33 178 6 42 10 – 70 14 13
23 41 32 46 51 28 10 93 21 92 37 32 – 10 60 28 134 47 67 27 31 79 41 19 52 19 54 – 41 43
65 62 39 50 49 28 54 111 98 26 17 78 – 16 25 40 18 35 20 47 42 78 38 48 48 – 32 25 120 28
60 25 84 48 40 11 17 18 11 102 45 3 16 50 17 30 48 18 – 21 14 52 24 1 77 91 34 36 63 61
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
1500
117
800
Primary-coverts
Secondarycoverts
600
1000
400
Number of samples
500
200
0
0
1600
3000
Ventral
Pectoral
1200
2000
800 1000
400 0
0
0
2
4
6
8 1200
Mantle 800
400
0 0
2
4
6
8
-1
Pb concentration difference ( g g ) 1200 900
Number of samples
200
Primarycoverts
Secondarycoverts
150
600
100
300
50
0
0 200
1000
Ventral
Pectoral 750
150
500
100
250
50
0
0 0
2
4
6
8 1200
Mantle
900 600 300 0 0
4
2
6
8
-1
Pb concentration difference ( g g ) Fig. 1. (a). Number of samples (n) of each type of feather required to differentiate the concentrations of Pb in pairs of individual specimens of Strix aluco. The dashed arrows indicate the n values corresponding to the mean (crosses), the lower limit (circles) and the upper limit (triangles) of the confidence interval for the modal value (solid arrow on the x axis) of the distribution of the differences in the mean concentrations of Pb between each individual, for each pair of feathers analysed. (b). Number of samples (n) of each type of feather required to differentiate the concentrations of Pb in pairs of individual specimens of Accipiter gentillis. The dashed arrows indicate the n values corresponding to the mean and the upper limit of the confidence interval for the modal value (solid arrow on the x axis) of the distribution of the differences in the mean concentrations of Pb between each individual, for each pair of feathers analysed.
118
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
S. aluco for all types of feather, except the covert feathers, whereas the highest CV in both species corresponded to the pectoral feathers. There was a significant correlation between the concentrations of Pb in the pectoral feathers and the mantle feathers for both A. gentilis (r ¼0.786n) and S. aluco (r ¼0.546n). In the case of S. aluco there were also significant correlations for the following comparisons: pectoral–ventral feathers (r ¼0.569nn), pectoral– secondary feathers (r¼ 0.498n) and mantle–primary coverts feathers (r ¼0.651nn). In A. gentilis significant correlations were found for: ventral–primary coverts feathers (r ¼0.762n), ventral– secondary coverts feathers (r ¼0.783nn) and primary–secondary coverts feathers (r ¼0.767n). Fig. 1a and b shows, for S. aluco and A. gentilis respectively, the result of the power analysis carried out to determine the number of samples (pairs of feathers) of each type that are needed to differentiate the concentrations of Pb corresponding to the pairs of individuals of the same species. In these figures, for each type of feather, arrows and dashed lines are used to indicate the n value corresponding to the mean value and the upper limit of the confidence interval for the difference in concentration most frequently observed (indicated with a solid arrow on the x axis). This most frequently observed concentration was selected using the modal value of the distribution of the differences in the mean concentrations of Pb in each pair of feathers analysed was selected. The minimum number of feathers is not shown because in most cases it was less than one. For S. aluco, an average of 150 pairs of primary covert feathers, 70 pairs of secondary covert feathers, 245 pairs of pectoral feathers, 320 pairs of ventral feathers and 380 pairs of mantle feathers were required, whereas for A. gentilis, 320 pairs of primary coverts feathers, 55 pairs of secondary coverts feathers, 230 pairs of pectoral feathers, 51 ventral feathers and 240 pairs of mantle feathers were required. As regards the number of samples required for a 10 percent level of accuracy in estimating the mean concentration of Pb in each type of feather, in the case of S. aluco, this was 432 for pectoral feathers, 360 for mantle feathers, 612 for ventral feathers, 3460 for primary coverts feathers and 6419 secondary coverts feathers. In A. gentilis, the number of feathers required was 645 for pectoral feathers, 1014 for mantle feathers, 219 for ventral feathers, 351 for primary coverts and 408 for secondary coverts feathers. The number of pairs of pectoral and mantle feathers required was higher in A. gentili, whereas more ventral, primary and secondary coverts feathers were required in S. aluco.
Falco biarmicus jugger (1.56 mg g 1) and those obtained by Battaglia et al. (2005) on analysing the mantle feathers of the Little Owl Athene noctua and the Common Buzzard Buteo buteo (19.1 and 8.87 mg g 1 respectively). The fact that the specimens were donated by Wildlife Recovery Centres enabled us to work with a much large number of specimens than in the previously cited studies. Measuring the concentrations of contaminants in a non invasive manner in wild animals is a problem that has not been fully resolved. Given the ethical and legal impediments associated with the hunting and killing of raptor species, the best option for obtaining samples is by donations of dead specimens from Wildlife Recovery Centres (Battaglia et al., 2005; Castro et al., 2011; Pe´rezLo´pez et al., 2008). The use of body feathers is another possible solution to this problem, as removal of these feathers does not injure the birds, so that they could be obtained from birds captured for this purpose then released, or from birds held in Wildlife Recovery Centres. As Burger and Gochfeld (2009) have recently indicated in reference to seabirds, body feathers and pectoral feathers are especially useful for establishing spatial and temporal patterns in bird populations and for evaluating heavy metal contamination in species that are threatened or in danger of extinction. However, in the present study, the coefficients of variation for the concentrations of Pb in pectoral feathers were higher than for those in the other body feathers analysed: 65 percent in S. aluco and 63 percent in A. gentilis. The variability in the concentrations in all types of feathers in both S. aluco and A. gentilis was enormous, despite the fact that: (i) each sample comprised two feathers, which reduces some of the variability between feathers of the same type; (ii) for some samples a value of half the limit of quantification was assigned, which also reduces the intra and inter individual variability, and (iii) in some cases the distributions were not normal, so that the power analysis may underestimate the number of samples (pairs of feathers) required (Gonza´lez et al., 2006). Moreover, if the body feathers were a good indicator of the levels of Pb in the blood, there should be a close correlation between the concentrations in the different types of feather, as all are related to the levels in blood. Although some of the correlations were significant, the regression coefficients were generally low, so that we can reject the possibility of carrying out regression with one type of feather as a predictor of the concentrations of Pb in the other type of feathers.
5. Conclusions 4. Discussion Information about the levels of heavy metals in contour and coverts feathers of birds is generally scarce, and mainly refers to seabirds. Very little data has been published as regards to the levels of Pb in the body feathers of raptor species. Comparison of the concentrations of Pb in the pectoral feathers of the birds examined in the present study (Table 1) with those reported in studies with seabirds, shows that the levels of Pb in the raptor species is much higher. Burger and Gochfeld (2000, 2009) reported mean concentrations of Pb of less than 2 mg g 1 in seabirds, and Gochfeld et al. (1996) reported concentrations of 0.003 mg g 1 in pectoral feathers in the Laughing gull Larus atricilla. These differences may be attributed to the different trophic levels of the species considered. Studies carried out by Garcı´a-Ferna´ndez et al. (1996) indicate higher tissue concentrations of Pb in carnivorous birds (nocturnal and diurnal raptors) than in fish-eating, insectivorous, omnivorous and carrion-eating birds, probably because some of these species live in areas where there is little human activity. The results obtained in the present study are in a range intermediate between the concentrations obtained by Movalli (2000) in pectoral feathers of the falcon
Because of the high intraindividual variability in the concentrations of Pb in each type of feather, with coefficients of variation of up to 100 percent, an extremely large number of feathers was required for accurate estimation of the mean concentration, and in some cases was even greater than the number of that particular type of feather in the birds. In order to differentiate the mean concentrations in a particular type of feather between different birds, the number of feathers required was again extremely high, and would be impossible to obtain (either because the birds do not have sufficient numbers of a particular type of feather or because of the stress that would be generated by removing a large number of feathers). All of this provides clear evidence that the contour and coverts feathers of the raptor species considered cannot be used to biomonitor contamination by Pb.
Acknowledgments We thank all personnel at the Wildlife Recovery Centres involved, and also Jesu´s Santamarina and Marta Prieto Rodrı´guez from the Biodiversity Conservation Service, Xunta de Galicia, for
S. Debe´n et al. / Ecotoxicology and Environmental Safety 85 (2012) 115–119
their help in obtaining samples. Thanks are due to Michael Gochfeld and one anonymous reviewer for their value comments and suggestions. This study was partly funded by projects awarded by the Xunta de Galicia ‘‘Banco de Especı´menes Ambiena tales de Galicia, 3 % Fase’’ and the ‘‘INCITE 2009/PX166’’. References Battaglia, A., Ghidini, S., Campanini, G., Spaggiari, R., 2005. Heavy metal contamination in little owl (Athene noctua) and common buzzard (Buteo buteo) from northern Italy. Ecotoxicol. Environ. Saf 60, 61–66. Burger, J., Gochfeld, M., 2000. Metal levels in feathers of 12 species of seabirds from Midway Atoll in the northern Pacific Ocean. Sci. Total Environ. 257, 37–52. Burger, J., Gochfeld, M, 2009. Comparison of arsenic, cadmium, chromium, lead, manganese, mercury and selenium in feathers in bald eagle (Haliaeetus leucocephalus), and comparison with common eider (Somateria mollissima), glaucous-winged gull (Larus glaucescens), pigeon guillemot (Cepphus columba), and tufted puffin (Fratercula cirrhata) from the Aleutian chain of Alaska. Environ. Monit. Assess 152, 357–367. Castro, I., Aboal, J.R., Ferna´ndez, J.A., Carballeira, A., 2011. Use of raptors for biomonitoring of heavy metals: gender, age and tissue selection. Bull. Environ. Contam. Toxicol. 86, 347–351. Cochran, W.G., Cox, G.M., 1957. Experimental Designs. Wiley, New York. Dauwe, T., Bervoets, L., Blust, R., Pinxten, R., Eens, M., 2000. Can excrement and feathers of nestling songbirds be used as biomonitors for heavy metal pollution? Arch. Environ. Contam. Toxicol 39, 541–546. Dauwe, T., Bervoets, L., Blust, R., Pinxten, R., Eens, M., 2003. Variation of heavy metals within and among feathers of birds of prey: effects of molt and external contamination. Environ. Pollut. 124, 429–436. Denneman, W.D., Douben, P.E., 1993. Trace metals in primary feathers of the barn owl (Tyto alba guttatus) in the Netherlands. Environ. Pollut 82, 301–310. Dmowski, K., Golimowski, J., 1993. Feathers of the magpie (Pica pica) as a bioindicator material for heavy metal pollution assessment. Sci. Total. Environ 139–140, 251–258. Furness, R.W., Greenwood, J.J.D., 1993. Birds as Monitors of Environmental Change. Chapman and Hall, London. Garcı´a-Ferna´ndez, A.J., Motas-Guzma´n, M., Navas, I., Marı´a-Mojica, P., Luna, A., Sa´nchez-Garcı´a, J.A., 1996. Environmental exposure and distribution of lead in four species of raptors in Southeastern Spain. Arch. Environ. Contam. Toxicol. 33, 76–82. Gochfeld, M., Belant, J.L., Shukla, T., Benson, T., Burger, J., 1996. Heavy metals in laughing gulls: gender, age and tissue differences. Environ. Toxicol. Chem. 15, 2275–2283.
119
Gonza´lez, X.I., Aboal, J.R., Ferna´ndez, J.A., Carballeira, A., 2006. Considerations on the sample size of wood mice used to biomonitor metals. Sci. Total Environ. 366, 910–914. Hahn, E., Hahn, K., Stoeppler, M., 1993. Bird feathers as bioindicators in areas of the German Environmental Specimen Bank: bioaccumulation of mercury in food chains and exogenous deposition of atmospheric pollution with lead and cadmium. Sci. Total Environ 139–140, 259–270. Hanson, H.C., Jones, R.L., 1968. Use the Feathers Minerals as Biological Tracers to Determine the Breeding and Moulting Ground of Wild Geese. State of Illinois Dept Reg Educ Nat Hist Surv Div, Biological Note, 60 pp. Hermoso de Mendoza, M., Soler Rodrı´guez, F., Herna´ndez Moreno, D., Gallego Rodrı´guez, M.E., Lo´pez Beceiro, A., Pe´rez Lo´pez, M., 2006. Comparative study of heavy metal and metalloid hepatic levels in diurnal raptors from Galicia and Extremadura. Rev. Toxicol 23, 138–145. Hudges, K.D., Ewins, P.J., Clark, K.E., 1997. A comparison of mercury levels in feathers and eggs of osprey (Pandion haliaetus) in the North American Great Lakes. Arch. Environ. Contam. Toxicol 33, 441–452. Jaspers, V.L., Covaci, A., Van den Steen, E., Eens, M., 2007. Is external contamination with organic pollutants important for concentrations measured in bird feathers? Environ. Int. 33, 766–772. Jenkins, C., 1975. Utilisation du pigeon biset (Columba livia Gm.) comme temoin de la pollution atmospherique par le plomb. C.R. Acad. Sci. Ser. D 281, 1187–1189. Locke, L.N., Friend, M., 1992. Lead poisoning of avian species other than waterfowl. In: Pain, D.J. (Ed.), Lead Poisoning in Waterfowl. The International Waterfowl and Wetland Research Bureau, Slimbridge, UK, pp. 19–22. Martı´nez-Lo´pez, E., Martı´nez, J.E., Marı´a-Mojica, P., Pen˜alver, J., Pulido, M., Calvo, J.F., Garcı´a-Ferna´ndez, A.J., 2004. Lead in feathers and d-aminolevulinic acid dehydratase activity in three raptors species from an unpolluted Mediterranean Forest (Southeastern Spain). Arch. Environ. Contam. Toxicol. 47, 270–275. Monteiro, L.R., Granadeiro, J.P., Furness, R.W., 1998. Relationship between mercury levels and diet in Azores seabirds. Mar. Ecol. Prog. Ser. 166, 259–265. Morris, R.F., 1954. A sequential sampling technique for spruce budworm egg surveys. Can. J. Zool. 32, 302–313. Movalli, P.A., 2000. Heavy metal and other residues in feathers of laggar falcon Falco biarmicus jugger from six districts of Pakistan. Environ. Pollut. 109, 267–275. Pe´rez-Lo´pez, M., Hermoso de Mendoza, M., Lo´pez Beceiro, A., Soler Rodriguez, F., 2008. Heavy metal (Cd, Pb, Zn) and metalloid (As) content in raptor species from Galicia (NW Spain). Ecotoxicol. Environ. Saf 70, 154–162. Vagasi, C.I., Pap, P.L., Tokolyi, J., Szekely, E., Barta, Z., 2011. Correlates of variation in flight feather quality in the Great Tit Parus major. Ardea 99, 53–60. Zar, J.H., 1984. Bioestatistical analysis, fourth ed. Prentice-Hall International Editions, London.