Elevated concentrations of naturally occurring ... - Wiley Online Library

2 downloads 0 Views 308KB Size Report
To assess the effects of naturally occurring heavy metals on wild birds, we compared reproductive success and heavy metal contents in feathers of Kagu ...
Ibis (2017), 159, 580–587

doi: 10.1111/ibi.12474

Elevated concentrations of naturally occurring heavy metals inversely correlate with reproductive output and body mass of the Kagu Rhynochetos jubatus € JORN THEUERKAUF,1* TOKUSHI HANEDA,2† YUJI OKAHISA,3 NOZOMU J. SATO,4 SOPHIE ROUYS,5 6 HENRI BLOC, KEISUKE UEDA,3 IZUMI WATANABE,2 RALPH KUEHN7,8 & ROMAN GULA1 1 Museum and Institute of Zoology, Polish Academy of Sciences, Wilcza 64, 00-679 Warsaw, Poland 2 Laboratory of Environmental Toxicology, Tokyo University of Agriculture and Technology, 3-5-8 Saiwa-cho, Fuchu, Tokyo 183-8509, Japan 3 Department of Life-Sciences, Rikkyo University, 3-34-1 Nishi-Ikebukuro, Toshima, Tokyo 171-8501, Japan 4 Japan Bird Research Association, 1-29-9 Sumiyoshi-cho, Fuchu, Tokyo 183-0034, Japan 5  a Cedex, New Caledonia Conservation Research New Caledonia, B.P. 2549, 98846 Noume 6  Syndicat mixte des Grandes Fougeres, BP 10, 98881 Farino, New Caledonia 7 €t Mu€ nchen, Unit of Molecular Zoology, Chair of Zoology, Department of Animal Science, Technische Universita Hans-Carl-von-Carlowitz-Platz 2, 85354 Freising, Germany 8 Department of Fish, Wildlife and Conservation Ecology, New Mexico State University, Box 30003, MSC 4901, Las Cruces, NM 88003-8003, USA

To assess the effects of naturally occurring heavy metals on wild birds, we compared reproductive success and heavy metal contents in feathers of Kagu Rhynochetos jubatus living on ultramafic (rich in heavy metals) soil with those of Kagu living on non-ultramafic soil. From 2003 to 2016, we monitored breeding of 19 Kagu families by radiotracking and video-monitoring, and collected rump down feathers from 69 wild Kagu. The metal concentrations in Kagu feathers correlated with the concentrations in the soil. The mean numbers of eggs laid and fledglings per year of Kagu families on non-ultramafic soil were about four times higher, and home-ranges three times smaller, than those of Kagu on ultramafic soil. Mass of eggs and the proportion of eggs that developed to fledglings were similar in the two areas, whereas the mass of adult Kagu on non-ultramafic soil was nearly 10% higher than that of adult Kagu living on ultramafic soil. The impact of naturally occurring heavy metals on Kagu breeding productivity and body mass appears to act through their effects on food supply rather than being caused directly by metal toxicity. The results imply that conservation of Kagu might be more effective in non-ultramafic areas, as populations can recover much faster on these soils and Kagu can then recolonize and bolster populations in ultramafic areas. Keywords: bird, breeding success, suboptimal habitat, ultramafic soil. Lack of essential or trace elements such as calcium (Ca), chromium (Cr), zinc (Zn), iron (Fe), copper (Cu) and manganese (Mn) may lead to physiological deficiency in vertebrates (Underwood 1977, Alloway 2012). However, high levels of metals †

Present address: Kogane Junior High School, 2-16-11, Matsudo 270-0032, Japan. *Corresponding author. Email: [email protected]

© 2017 British Ornithologists’ Union

can reduce breeding success (e.g. Eeva et al. 2009). In this study, we consider a flightless, cooperatively breeding endemic bird, the Kagu Rhynochetos jubatus, whose global population is restricted entirely to the rainforests of New Caledonia (Letocart & Salas 1997). Here, a third of the land surface is covered with ultramafic soil rich in heavy metals, where some plants have evolved specific adaptations to withstand high levels of heavy metals (Jaffre et al. 1976, Amir et al. 2007).

Heavy metal concentration and reproductive output of Kagu

Through video-monitoring of nests (Gula et al. 2010), we know that earthworms (Oligochaeta) are the most important component of Kagu diet. Earthworms are accumulators of heavy metals (e.g. Hobbelen et al. 2006), which is likely to be the reason feathers of Kagu living on ultramafic soil have very high heavy metal contents compared with other bird species (Theuerkauf et al. 2015). The Kagu was thought to be a species characteristic of rainforests growing on ultramafic soil (Ekstrom et al. 2002), but their reproductive output in this environment is only about half a fledgling per family per year (Theuerkauf et al. 2009). Ultramafic soil might therefore not be an optimal habitat for Kagu, even though they can tolerate such conditions. Birds excrete heavy metals by feather moulting (Chatelain et al. 2014), so heavy metal contents in feathers typically reflect the consumption of these elements since the last moult. To assess the potential effects of heavy metals on wild Kagu populations, we compared reproductive success and heavy metal contents in feathers of a population of Kagu living on ultramafic soil with those of a population living on non-ultramafic soil approximately 100 km away, while also considering other potentially influential factors, such as habitat quality, predation and family structure (Theuerkauf et al. 2009). METHODS From 2003 to 2012, we monitored 12 Kagu families living in rainforest of the Parc Provincial de la Riviere Bleue (PPRB, 22°30 –22°120 S, 166°330 –166°460 E) on average for 6 years (sum 53 breeding attempts). From 2011 to 2016, we monitored seven Kagu families living in rainforest of the Parc Provincial des Grandes Fougeres (PGF, 21°300 –21°390 S, 165°390 – 165°500 E) on average for 4 years (74 breeding attempts). We additionally measured heavy metal concentrations in feathers of captive Kagu from the Yokohama Preservation and Research Center (Yokohama, Japan, 35°500 S, 139°510 E) as an outlier group. The PPRB is a 90-km2 reserve on ultramafic soil with a mean annual rainfall of about 2900 mm and mean annual temperature of 21 °C (data from Meteo France). We worked in rainforest at altitudes of 150–350 m asl. Soils in this area contain high concentrations of Fe (320 000–335 000 mg/ kg), Cr (35 300 mg/kg), nickel Ni (7300–8300 mg/ kg), Mn (6400–6700 mg/kg) and cobalt Co (700– 880 mg/kg) (Jaffre & Veillon 1990, Amir et al.

581

2007). The PGF is a 45-km2 reserve on schist soil, with a mean annual rainfall of about 1600 mm and a mean annual temperature of 20 °C (data from Direction des Affaires Veterinaires, Alimentaires et Rurales, New Caledonia (DAVAR) and Meteo France). We worked in rainforest at altitudes of 300–500 m asl. Although the PGF is on non-ultramafic soil, the nearest nickel mines are only 20 km to the north and 40 km to the east. Soils in the PGF contain lower concentrations of Fe (50 400–84 400 mg/kg), Cr (500 mg/kg), Ni (800 mg/kg), Mn (410–2300 mg/kg) and Co (100 mg/kg) (Latham et al. 1978, Jaffre & Veillon 1995). The average monthly Southern Oscillation Index during fieldwork in PPRB was 2.0 compared with 2.0 during fieldwork in PGF (data by Australian Bureau of Meteorology, http://www. bom.gov.au/climate/current/soihtm1.shtml), causing 7% less rainfall during the period of fieldwork in PGF compared with the period we worked in PPRB. We fitted 118 immature to adult Kagu (56 in the PPRB and 62 in the PGF) with harnessmounted radio-transmitters (Sirtrack, Havelock North, New Zealand). Adult birds of both sexes incubate the single egg of a breeding attempt (Theuerkauf et al. 2009). We found Kagu nests by locating tagged breeding adults at least once a month (which ensured finding successful nests, as the incubation length is 5 weeks) at night, when they were either roosting or on a nest. Additionally, we used radio-transmitters with activity sensors that indicate when Kagu remained inactive for longer than 5 min (in most cases, this is when they are on a nest). Because mortality might also play a role in breeding success, we compared annual mortality rates of radiotracked adult Kagu of both study areas by dividing casualties by the cumulative length of radiotracking in each respective area (Heisey & Fuller 1985). We either video-monitored (see Gula et al. 2010 for details) or regularly checked (one to three times per week) the nests if Kagu were still incubating. We sexed all individuals (nestling to adult) by the molecular genetic method described in detail in Stoeckle et al. (2012). We re-captured radiotracked Kagu on average once a year by hand, usually at night while they were roosting. We weighed Kagu (every time when captured) and eggs (after finding a nest) with spring scales. As we did not know the exact age of the egg when we found a nest, and eggs lose mass with age (Rahn & Ar 1974), we corrected egg mass for water loss

© 2017 British Ornithologists’ Union

582

J. Theuerkauf et al.

during incubation. First, we estimated whenever possible the egg-laying time as the day of hatching minus the average incubation duration of 34.5 days (Letocart & Salas 1997). Then, we conducted a linear regression analysis of all breeding attempts for which we knew the hatching date: egg mass (g) = 0.37 age of egg (days) + 77 (n = 57, R2 = 0.284). Finally, we corrected egg mass for the difference in age considering the average mass loss of 0.37 g per day. After the chick fledges, at the age of 3 days on average, Kagu feed their chick during the day and brood it at night for about 2 months (Letocart & Salas 1997). To assess whether the chick survived this 2-month fledgling period, we regularly located radiotagged adult birds to find the chick. We considered a breeding attempt successful if we observed the chick alive at the age of 2 months, by which time it had reached about two-thirds of adult body mass and was roosting alone at night. Like other carnivorous bird species in New Caledonia, Kagu have no well-defined breeding season (Barre et al. 2013). The number of breeding attempts per year is very variable, ranging from zero to seven (authors’ unpubl. data). We therefore assessed the annual breeding output of each family as the total number of eggs laid (equal to number of breeding attempts) and the total number of chicks that survived the fledgling period divided by the years of monitoring of a given family. We estimated the home-range of a family as the 100% minimum convex polygon of locations (Gula & Theuerkauf 2013) of the breeding birds as a proxy for habitat quality and assumed that its size is inversely related to food availability (McLoughlin & Ferguson 2000), as shown, for example, in Wild Turkeys Meleagris gallopavo (Badyaev et al. 1996). We used families as sample units for all reproductive parameters, and thus we calculated first one value for each family, and then means and 95% confidence intervals based on variation among families. From 2007 to 2013, we collected about 10 rump down feathers per individual from wild Kagu in the PPRB (11 females and 27 males of 12 families) and PGF (nine females and 22 males of seven families), and Yokohama Preservation and Research Centre (seven females and seven males). Feathers were stored at 4 °C in a polyethylene bag to prevent cross-contamination. Birds can eject heavy metals through eggshells (Burger 1994), so we also compared heavy metal contents in the shell of one egg from ultramafic soil (PPRB) and one egg from non-

© 2017 British Ornithologists’ Union

ultramafic soil (PGF). We analysed heavy metal contents as described in Theuerkauf et al. (2015). Feather samples were washed in an ultrasonic bath with 0.3% polyoxyethylene laurel ether and rinsed with distilled and deionized water. After washing, all samples were dried at 90 °C for 16 h. Approximately 0.05–0.10 g of dried sample was digested with a solution of 2.0 mL of nitric acid and 23 mL of Milli-Q water using a double-sealed decomposition Teflon vessel (P-25; San-ai Science Co. Ltd, Nagoya, Japan) and a microwave oven. We determined concentrations of 27 elements (lithium Li, magnesium Mg, Ca, vanadium V, Cr, Mn, Fe, Co, Ni, Cu, Zn, gallium Ga, arsenic As, selenium Se, rubidium Rb, strontium Sr, molybdenum Mo, silver Ag, cadmium Cd, indium In, tin Sn, antimony Sb, caesium Cs, barium Ba, thallium Tl, lead Pb, bismuth Bi) by inductively coupled plasma-mass spectrometry (ICP-MS) with an HP7500cx (Agilent Technologies Inc., Santa Clara, CA, USA). We used the following settings: RF power of 1600 W, carrier gas (argon, Ar) flow rate of 1.12 L/min, make up gas (Ar) flow rate of 0.1 L/min, peristaltic pump speed of 0.1 rps. For mercury (Hg), we determined concentrations by the cold vapour technique using a mercury analyser (Hiranuma HG300, Hitachi Ltd, Tokyo, Japan). We determined the accuracy of our analyses using bovine liver (NIST 1577b) as standard reference material (Zeisler et al. 2008). Relative standard deviations by sextuplicate analyses of NIST 1577b were < 10%, except for Ga (12%), Ag (19%) and In (15%). The recovery of 18 tested elements ranged from 56 to 130%. We calculated means and 95% confidence intervals using individual feather samples as replicates. We derived average values in soil from published data (Latham et al. 1978, Jaffre & Veillon 1990, 1995, Amir et al. 2007). RESULTS The metal concentrations in Kagu feathers were correlated to the concentrations in soil, with the exception of Ca and Mg (Fig. 1). Feathers of Kagu in the PPRB contained higher levels of elements typical for ultramafic soil, such as V, Cr, Mn, Fe, Co, Ni, Hg, Se and Sn, than of Kagu on non-ultramafic soil (Fig. 2). The concentrations of Cr (31 times), Co (nine times), Ni (seven times) and Fe (five times) were much higher in Kagu feathers collected on ultramafic soil than on non-ultramafic soil. In contrast, concentrations of Cu, Zn, Ba and

Heavy metal concentration and reproductive output of Kagu

Figure 1. Relationship between concentration of metal ele& ments in soil (published data from Latham et al. 1978, Jaffre Veillon 1990, 1995, Amir et al. 2007) and dry down feathers of Kagu on ultramafic soil (PPRB, filled dots) and on non-ultramafic soil (PGF, open dots). The regression model (exponential) does not include Ca and Mg.

Bi were higher in the PGF than in the PPRB. The levels of Ca and Mg in Kagu feathers were not different between the two study areas (Table 1), although the concentrations of Ca (16 times) and Mg (two times) differed between the two soil types (Fig. 1). The outlier group from Japan had generally lower values of elements typical for New Caledonian ultramafic soil, even lower than samples from non-ultramafic soil (Fig. 1) but had higher values of heavy metals related to anthropogenic pollution, such as Cd and Ag. Metal concentrations in the eggshell sample from each study area were for most elements proportional to concentrations in feathers from that study area (Table 1). The breeding output (both eggs and fledglings) of Kagu on non-ultramafic soil was about four times higher than that of Kagu on ultramafic soil, but this was almost entirely due to Kagu on nonultramafic soil laying more eggs per season than those on ultramafic soil (Table 2). The mass of eggs, family size and number of fledglings per laid egg were similar between soil types. The mass of reproductive Kagu (both sexes, as there was no difference between females and males, respectively, 910  38 and 926  29 g, average values of 52 Kagu) on non-ultramafic soil was also nearly 10% higher than that of Kagu living on ultramafic soil. This difference in mass (77 g) corresponded to the average mass of an egg (78 g). However, just

583

before laying an egg, the mass of females on ultramafic soil was not different from the mass of females on non-ultramafic soil. The shortest time interval between egg-laying (replacement egg in cases in which the first breeding attempt was unsuccessful) was 2 weeks on non-ultramafic soil, whereas it was 2 months on ultramafic soil. Homeranges were three times larger on ultramafic soil than on non-ultramafic soil (Table 2), indicating a much lower habitat quality on ultramafic soil. Annual mortality rates of adult Kagu were comparable in the PPRB (4.1%, n = 194 Kagu years) and PGF (5.7%, n = 122), of which at least half was caused by dog predation in both study areas. During 12 500 h of nest video-monitoring in the PPRB and 12 500 h in the PGF, we did not record any predation of eggs or chicks. DISCUSSION Kagu had higher heavy metal uptake on ultramafic than on non-ultramafic soil, whereas their reproductive output was much lower, primarily as a result of reduced annual egg production. Because the Kagu is an endangered species, we could not expose birds experimentally to high doses of heavy metals to verify whether low breeding success was due to heavy metal toxicity or to the low productivity of soil on ultramafic rock resulting in low prey abundance. If heavy metal toxicity was the major reason for poor reproduction, we would rather expect a higher hatching failure and smaller eggs. However, the proportion of fledglings per egg was equal in the two regions, whereas parameters probably related to food availability, such as egg production, home-range size and body mass, indicated a lower food supply on the poorer ultramafic soil. Our observation that females of both study areas had a similar mass just before egg-laying also indicates that females are likely to need to reach a critical mass before being able to lay. The four times greater egg production by Kagu on nonultramafic soil corresponds to the interval between egg-laying, which was also four times longer on ultramafic soil. We therefore conclude that food shortage on ultramafic soil is the most likely reason for the reduced frequency of egg production. High heavy metal concentrations in soil cause denitrification and a reduction in soil fertility (Liu et al. 2016) and thus lead to lower abundance of soil invertebrates (e.g. Filzek et al. 2004, Lev^eque et al. 2015).

© 2017 British Ornithologists’ Union

584

J. Theuerkauf et al.

Figure 2. Concentrations of metal elements (with 95% confidence intervals) in dry down feathers of Kagu from an ultramafic area (PPRB), a non-ultramafic area (PGF), from captive birds (Yokohama), and, when available (published data from Latham et al. 1978,  & Veillon 1990, 1995, Amir et al. 2007), in soil of the PPRB and PGF. Jaffre

Table 1. Concentrations (ppm) of 12 selected elements in one eggshell from ultramafic soil and one eggshell from non-ultramafic soil compared with concentrations in dry down feathers (with 95% confidence intervals) from the same Kagu populations (PPRB: n = 38, PGF: n = 31). Non-ultramafic (PGF) Element Ca Fe Mg Sr Ni Mn Ba Cu Se Zn Cr Co

Ultramafic (PPRB)

Feathers

Egg

Egg

           

162 000 4460 2110 187 7.3 3.6 7.9 1.7 0.6 0.5 0.003 0.93

102 000 3260 960 363 80.0 17.7 0.5 1.7 0.6 0.5 4.96 2.36

2038 129 488 7 1.6 12.0 1.8 11.6 4.6 150.7 0.38 0.15

210 31 52 1 0.3 2.1 0.4 0.7 0.5 7.8 0.05 0.03

Values with non-overlapping confidence intervals are shown in bold.

© 2017 British Ornithologists’ Union

Feathers 1982 522 521 8 10.7 16.3 0.6 9.3 11.0 137.7 11.16 1.36

           

233 77 60 1 1.6 2.2 0.1 0.5 0.9 6.7 2.01 0.21

Heavy metal concentration and reproductive output of Kagu

Table 2. Mean values (with 95% confidence interval) of reproductive parameters and home-range size measured as minimum convex polygon of Kagu families living on ultramafic soil and non-ultramafic soil (PPRB: n = 12, PGF: n = 7).

Eggs (breeding attempts) per year Fledglings per year Fledglings per laid egg Egg mass, corrected for age (g) Mass of reproductive Kagu (g) Mass of female before egg-laying Adult individuals in family Home-range size (km²)

Ultramafic (PPRB)

Non-ultramafic (PGF)

P

0.72  0.16

2.71  0.47

< 0.001

0.35  0.16

1.46  0.52

0.005

0.50  0.22

0.54  0.16

0.722

77.5  2.6

78.1  1.3

0.536

886  23

963  30

< 0.001

915  49

962  38

0.305

3.8  0.8

4.1  0.5

0.431

0.250  0.055

0.087  0.028

< 0.001

Significant differences (t-test) are presented in bold.

At the same time, we could exclude family size and predation as confounding factors influencing breeding output of Kagu, as they were comparable between the two study areas. We furthermore think it is unlikely that the breeding output was affected by the small difference of 1 °C in the mean annual temperature due to the difference in altitude of 250 m. Rainfall, however, does influence the timing of breeding, as in both study areas we observed more breeding attempts during wetter periods (authors’ unpubl. data). As the study area on ultramafic soil had nearly twice as much annual rainfall as the study area on non-ultramafic soil, this factor might have reduced what would otherwise have been an even larger difference in breeding output between the two study areas. Differences in vegetation between the two sites are unlikely to have been influential, as Kagu are ground feeders and do not depend on a specific vegetation composition, provided that there is a dense canopy cover and litter layer where they can find prey. We therefore conclude that, contrary to former assumptions (Ekstrom et al. 2002), ultramafic soil is a suboptimal habitat for Kagu. The nearly exclusive occurrence of Kagu on ultramafic soil in the past was likely to be related to lower predation risk. Since their introduction in

585

the 19th century, dogs have been the main predators of Kagu (Hunt et al. 1996), but in New Caledonia dogs do not seem able to establish sustainable populations in the wild (Rouys & Theuerkauf 2003) and are therefore restricted to the surroundings of human settlements. Ultramafic regions are generally uninhabitable for humans due to the poor soil, rich in heavy metals, and dogs are therefore rare in these areas. As a consequence, Kagu found refuge in areas far from human settlements (Hunt 1996), most often on ultramafic soil. In the last two decades, Kagu have recolonized many non-ultramafic areas, such as the PGF, possibly because there is a decrease in the number of hunters going far into the mountains with dogs (Floret 2012). The fact that mortality of Kagu in non-ultramafic areas is no longer higher than in ultramafic areas supports this interpretation. Refuge habitats have also played an important role in the survival of endangered species on other islands. In New Zealand, invasive predators drove the critically endangered Kakapo Strigops habroptilus and the endangered Takahe Porphyrio mantelli into alpine habitats (Bunin & Jamieson 1995, Powlesland et al. 2006). As these habitats were suboptimal and could not ensure survival of populations over the long-term, secondary refuges were created as a conservation measure on predator-free islands (e.g. Bunin et al. 1997, Elliott et al. 2006). In New Caledonia, there are no islands with similar habitats as on the mainland, but ultramafic areas might serve as refuge for other endangered bird species. Nearly the entire population of the critically endangered Crow Honeyeater Gymnomyza aubryana is limited to ultramafic soil (Okahisa et al. 2016). Carter and Bright (2002) showed that artificially created refuge habitats can reduce predation of invasive mammals on native species. In contrast, it is possible that ultramafic soil in New Caledonia provides to some extent a natural protection from invasive species. We do not have direct evidence for an effect of heavy metals through food supply on breeding output of Kagu, but the observed correlations support this interpretation. Having excluded other potential factors influencing breeding success, we conclude that the main reason for lower breeding productivity in the ultramafic area is likely to be a poorer food supply. The results also imply that conservation of Kagu might be more effective in non-ultramafic areas because populations can

© 2017 British Ornithologists’ Union

586

J. Theuerkauf et al.

recover much faster and Kagu can then recolonize or restock populations in ultramafic areas. Because Kagu had comparatively high concentrations of heavy metals in feathers (Theuerkauf et al. 2015), even on non-ultramafic soil, it is unlikely that genetic admixture among populations from ultramafic and non-ultramafic areas will reduce the ability of Kagu to cope with heavy metals. This study was financed by the Polish National Science Centre (grant NCN 2011/01/M/NZ8/03344), by the Polish Ministry of Science and Higher Education (grant 2P04F 001 29), Conservation des Especes et Populations Animales (France), La Fondation Nature et Decouvertes (France), and a doctoral award from Province Sud (to S.R.). The Province Sud (New Caledonia) issued all permits (2322-2002, 358-2003, 4081-2004, 2506-2005, 1691-2006, 201-2007, 1007-2008, 10029-2009, 3582010, 2017-2011, 2425-2012, 1142-2013, 60-2015). We thank Meteo France and DAVAR for providing data on rainfall; the Yokohama Preservation and Research Centre for providing feather samples; M. Broersen, C. Chatreau, P. de Pous, D. Dingemans, S. Duijns, N. Emura, P. Guichard, B. Michielsen, E. Minnema, L. Nijdam, N. Petit, H. Theuerkauf, J. van Dijk, M. van Opijnen, J. Wardenaar and many others for their help during fieldwork; and anonymous reviewers for providing valuable comments.

REFERENCES Alloway, B.J. 2012. Heavy metals and metalloids as micronutrients for plants and animals. In Alloway, B.J. (ed.) Heavy Metals in Soils – Trace Metals and Metalloids in Soils and their Bioavailability. Environ. Pollut. 22: 195–209.  , T. 2007. Amir, H., Perrier, N., Rigault, F. & Jaffre Relationships between Ni-hyperaccumulation and mycorrhizal status of different endemic plant species from New Caledonian ultramafic soil. Plant Soil 293: 23–35. Badyaev, A.V., Etges, W.J. & Martin, T.E. 1996. Ecological and behavioral correlates of variation in seasonal home ranges of Wild Turkeys. J. Wildl. Manage. 60: 154–164. , N., Tron, F., Chartendrault, V., Okahisa, Y., Sato, N., Barre Legault, A. & Theuerkauf, J. 2013. Breeding seasons of landbirds in New Caledonia. Wilson J. Ornithol. 125: 384– 389. Bunin, J.S. & Jamieson, I.G. 1995. New approaches toward a better understanding of the decline of Takahe (Porphyrio mantelli) in New Zealand. Conserv. Biol. 9: 100–106. Bunin, J.S., Jamieson, I.G. & Eason, D. 1997. Low reproductive success of the endangered Takahe Porphyrio mantelli on offshore island refuges in New Zealand. Ibis 139: 144–151. Burger, J. 1994. Heavy metals in avian eggshells: another excretion method. J. Toxicol. Environ. Health 41: 207–220. Carter, S.P. & Bright, P.W. 2002. Habitat refuges as alternatives to predator control for the conservation of endangered Mauritian birds. In Veitch, C.R. & Clout, M.N.

© 2017 British Ornithologists’ Union

(eds) Turning the tide: the eradication of invasive species: 71–78. IUCN SSC Invasive Species Specialist Group. Gland: IUCN. Chatelain, M., Gasparini, J., Jacquin, L. & Frantz, A. 2014. The adaptive function of melanin-based plumage coloration to trace metals. Biol. Lett. 10: 20140164. Eeva, T., Ahola, M. & Lehikoinen, E. 2009. Breeding performance of Blue Tits (Cyanistes caeruleus) and Great Tits (Parus major) in a heavy metal polluted area. Environ. Pollut. 157: 3126–3131. Ekstrom, J.M.M., Jones, J.P.G., Willis, J., Tobias, J., Dutson, , N. 2002. New information on the distribution, G. & Barre status and conservation of terrestrial bird species in Grande Terre, New Caledonia. Emu 102: 197–207. Elliott, G.P., Eason, D.K., Jansen, P.W., Merton, D.V., Harper, G.A. & Moorhouse, R.J. 2006. Productivity of Kakapo (Strigops habroptilus) on offshore island refuges. Notornis 53: 138–142. Filzek, P.D.B., Spurgeon, D.J., Broll, G., Svendsen, C., Hankard, P.K., Parekh, N., Stubberud, H.E. & Weeks, J.M. 2004. Metal effects on soil invertebrate feeding: measurements using the bait lamina method. Ecotoxicology 13: 807–816.  conomique et analyse Floret, A. 2012. ICONE. Etude socio-e ne fices: Conservation International, unpublished couts/be report. Gula, R. & Theuerkauf, J. 2013. The need for standardization in wildlife science – home range estimators as an example. Eur. J. Wildl. Res. 59: 713–718. Gula, R., Theuerkauf, J., Rouys, S. & Legault, A. 2010. An audio/video surveillance system for wildlife. Eur. J. Wildl. Res. 56: 803–807. Heisey, D.M. & Fuller, T.K. 1985. Evaluation of survival and cause-specific mortality rates using telemetry data. J. Wildl. Manage. 49: 668–674. Hobbelen, P.H.F., Koolhaas, J.E. & van Gestel, C.A.M. 2006. Bioaccumulation of heavy metals in the earthworms Lumbricus rubellus and Aporrectodea caliginosa in relation to total and available metal concentrations in field soil. Environ. Pollut. 144: 639–646. Hunt, G.R. 1996. Environmental variables associated with population patterns of the Kagu Rhynochetos jubatus of New Caledonia. Ibis 138: 778–785. Hunt, G.R., Hay, R. & Veltman, C.J. 1996. Multiple Kagu Rhynochetos jubatus deaths caused by dog attacks at a high-altitude study site on Pic Ningua, New Caledonia. Bird Conserv. Int. 6: 295–306. , T. & Veillon, J.-M. 1990. Etude floristique et Jaffre ^ ts dense humides sur roches structurale de deux fore donie. Bull. Mus. Natl Hist. ultrabasiques en Nouvelle-Cale r. 12: 243–273. Nat., 4e Se , T. & Veillon, J.-M. 1995. Structural and floristic Jaffre characteristics of a rain forest on schist in New Caledonia: a comparison with an ultramafic rain forest. Bull. Mus. Natl r. 17, B, Adansonia 3-4: 201–226. Hist. Nat., 4e Se , T., Brooks, R.R. & Reeves, R.D. 1976. Sebertia Jaffre acuminata: a nickel accumulating plant from New Caledonia. Science 193: 573–580.  Latham, M., Quantin, P. & Aubert, G. 1978. Etude des sols  donie. Carte pe  dologique de la de la Nouvelle-Cale  donie a  1 : 1 000 000. Paris: ORSTOM. Nouvelle-Cale Notice Explicative 78.

Heavy metal concentration and reproductive output of Kagu

 tocart, Y. & Salas, M. 1997. Spatial organisation and Le  re Bleue breeding of Kagu Rhynochetos jubatus in Rivie Park, New Caledonia. Emu 97: 97–107.  ve ^que, T., Capowiez, Y., Schreck, E., Mombo, S., Mazzia, Le C., Foucault, Y. & Dumat, C. 2015. Effects of historic metal (loid) pollution on earthworm communities. Sci. Total Environ. 511: 738–746. Liu, Y., Liu, Y., Zhou, H., Li, L., Zheng, J., Zhang, X., Zheng, J. & Pan, G. 2016. Abundance, composition and activity of denitrifier communities in metal polluted paddy soil. Sci. Rep. 6: 19086. McLoughlin, P.D. & Ferguson, S.H. 2000. A hierarchical pattern of limiting factors helps explain variation in home range size. Ecoscience 7: 123–130. Okahisa, Y., Legault, A., Angin, B., Chartendrault, V., , N., de Franceschi, C., Duval, T., Tron, F., Gula, R. Barre & Theuerkauf, J. 2016. Using playback and habitat models to estimate the distribution and population size of the critically endangered Crow Honeyeater (Gymnomyza aubryana) in New Caledonia. Emu 116: 41–47. Powlesland, R.G., Merton, D.V. & Cockrem, J.F. 2006. A parrot apart: the natural history of the Kakapo (Strigops habroptilus), and the context of its conservation management. Notornis 53: 3–26. Rahn, H. & Ar, A. 1974. The avian egg: incubation time and water loss. Condor 76: 147–152.

587

Rouys, S. & Theuerkauf, J. 2003. Factors determining the distribution of introduced mammals in nature reserves of the southern province, New Caledonia. Wildl. Res. 30: 187–191. Stoeckle, B.C., Theuerkauf, J., Rouys, S., Gula, R., Lorenzo, A., Lambert, C., Kaeser, T. & Kuehn, R. 2012. Identification of polymorphic microsatellite loci for the endangered Kagu (Rhynochetos jubatus) by high-throughput sequencing. J. Ornithol. 153: 249–253. riot, J.M. & Gula, R. 2009. Theuerkauf, J., Rouys, S., Me Group territoriality as a form of cooperative breeding in the flightless Kagu of New Caledonia. Auk 126: 371–375. Theuerkauf, J., Haneda, T., Sato, N.J., Ueda, K., Kuehn, R., Gula, R. & Watanabe, I. 2015. Naturally high heavy metal concentrations in feathers of the flightless Kagu Rhynochetos jubatus. Ibis 157: 177–180. Underwood, E.J. 1977. Trace Elements in Human and Animal Nutrition, 4th edn. New York: Academic Press. Zeisler, R., James, W.D., Mackey, E.A., Spatz, R.O. & Greenberg, R.R. 2008. NAA characterization of the new bovine liver SRM. J. Radioan. Nucl. Chem. 278: 783–787. Received 27 June 2016; revision accepted 2 March 2017. Associate Editor: Ben Smit.

© 2017 British Ornithologists’ Union