Portal de Periódicos da CAPES

Variation in the extra-pair mating systems of Acadian Flycatchers and Wood Thrushes in forest fragments in southern Ontario

2009; Association of Field Ornithologists; Volume: 80; Issue: 2 Linguagem: Inglês

10.1111/j.1557-9263.2009.00216.x

1557-9263

Melissa L. Evans, Bonnie E. Woolfenden, Lyle Friesen, Bridget J. M. Stutchbury,

Avian ecology and behavior

ABSTRACT The application of molecular tools to studies of avian mating systems has revealed that most songbird species engage in extra-pair matings. However, little is known about the possible effects of habitat fragmentation on extra-pair mating systems. During the breeding seasons of 2002 and 2003, we quantified the frequency of extra-pair matings in Acadian Flycatchers (Empidonax virescens) and Wood Thrushes (Hylocichla mustelina) nesting in forest fragments in southern Ontario. Acadian Flycatchers are at the limit of their northern range in southern Ontario and occurred at low densities (0.005–0.015 males/ha). Across forest fragments, 14% of young Acadian Flycatchers were the result of extra-pair fertilizations. In contrast, Wood Thrushes were common in forest fragments, with breeding densities ranging from 0.37 to 1 males/ha. Extra-pair mating was common among Wood Thrushes, with 40% of young the result of extra-pair matings. Compared to populations studied in less fragmented habitats in Pennsylvania, rates of extra-pair paternity in Ontario were lower for Acadian Flycatchers and higher for Wood Thrushes. Our results demonstrate that rates of extra-pair mating can vary across landscapes. However, the extra-pair mating systems of Acadian Flycatchers and Wood Thrushes appear to respond differently to fragmentation. We suggest that low breeding densities on fragments may be mediating the low rates of extra-pair mating observed in Acadian Flycatchers in southern Ontario, whereas changes to the behavioral tactics of male and female Wood Thrushes in southern Ontario may explain the high levels of extra-pair mating. La aplicación de instrumentos moleculares para estudiar los sistemas de apariamiento de pájaros ha revelado que la mayoría de las aves canoras se envuelven en actividades sexuales con individuos que no son su pareja (extra-parejas). Sin embargo, se conoce muy poco sobre los posibles efectos de la fragmentación de hábitat en especies con conducta sexual extra-parejas. Durante la temporada reproductiva de 2002 y 2003, cuantificamos la frecuencia de copulaciones extra-parejas en un papamoscas (Empidonax virescens) y un zorzal (Hylocichla mustelina) que anidaron en fragmentos de bosque en el sur de Ontario, Canada. Los papamoscas de Ontario, se encuentran en el límite norte de su distribución y sus densidades son bajas (0.005–0.015 machos/ha). A lo largo de fragmentos de bosque el 14% de los papamoscas jóvenes fueron el resultado de conducta extra-parejas. Encontraste, los zorzales resultaron comunes en los fragmentos de bosques estudiados con densidades entre 0.37–1.0 machos/ha. La conducta extra-pareja fue común en estos, con un 40% de los juveniles productos de dicha estrategia reproductiva. Comparados con poblaciones estudiadas en áreas menos fragmentadas en Pennsylvania, la tasa de paternidad producto de la conducta sexual extra-parejas en Ontario resultó más bajas para el papamoscas y más altas para el zorzal. Nuestros resultados demuestran que la tasa de copulación extra-parejas pueden variar de un lugar a otro. Sin embargo, la conducta reproductiva extra-parejas del papamoscas y el zorzal parece responder de forma diferente a la fragmentación. Sugerimos que la baja densidad de individuos reproductivos en áreas fragmentadas puede mediar en la baja tasa de conducta extra-parejas obsevadas en el papamoscas, mientras que cambios en las tácticas reproductivas en las hembras y machos de los zorzales pudiera explicar los altos niveles de conducta extra-pareja en las aves estudiadas en Ontario. Most Neotropical migratory songbirds engage in extra-pair matings and in some species >30% of young are sired by extra-pair males (Griffith et al. 2002, Stutchbury et al. 2005a). Many factors potentially influence the rate of extra-pair paternity (EPP) in a species, including breeding density, breeding synchrony, and the extra-pair behavior of males and females (reviewed by Griffith et al. 2002, Westneat and Stewart 2003). Comparisons among populations reveal that breeding density is positively related to the frequency of EPP in some songbird species (Lifjeld et al. 1991, Gelter and Tegelström 1992, Yezerinac et al. 1999), but not others (Hasselquist et al. 1995, Bjørnstad and Lifjeld 1997, Fridolfsson et al. 1997), and breeding synchrony is generally thought to be positively correlated with EPP (Griffith et al. 2002). Few investigators have examined how and why extra-pair behavior (e.g., off-territory forays, mate guarding, and solicitation of extra-pair mates) varies between populations. However, the differential expression of these behaviors may also influence rates of EPP (Westneat 1993, Gray 1996, Westneat and Stewart 2003). Habitat fragmentation may affect the social landscape and the frequency of EPP if the number of potential social and extra-pair mates is reduced in habitat fragments and the surrounding landscape (Morton 1992). This, in turn, may increase the costs of seeking extra-pair matings (Morton 1992, Debinski and Holt 2000, Ahlering and Faaborg 2006, Banks et al. 2007). To date, the only comprehensive study of the effects of forest fragmentation on extra-pair mating behavior was conducted with Hooded Warblers (Wilsonia citrina; Stutchbury et al. 1997, 2005b, Stutchbury 1998, Norris and Stutchbury 2001, 2002). In a contiguous forest where breeding density was high, EPP was common (54% young) and both males and females made frequent, short forays to neighboring territories in pursuit of extra-pair copulations (Neudorf et al. 1997, Stutchbury 1998, Chiver et al. 2008). In contrast, forest fragments supporting only one or two breeding pairs had fewer (EPY; 22% young; Stutchbury et al. 2005b) and females breeding on fragments curtailed off-territory forays whereas foray effort and distance increased for males (Norris and Stutchbury 2001, 2002). These studies indicate that extra-pair behavior and EPP may vary across Hooded Warbler populations, potentially in response to forest fragmentation. Clearly, there is a need to better understand how forest fragmentation may influence opportunities for extra-pair mating. We examined the extra-pair mating systems of Acadian Flycatchers (Empidonax virescens) and Wood Thrushes (Hylocichla mustelina) breeding in forest fragments in southern Ontario. The extra-pair mating systems of these species have been studied previously in a relatively large forest (180 ha) in northern Pennsylvania (Woolfenden et al. 2005, Evans et al. 2008). The study site in Pennsylvania is located 150 km and 200 km southeast of the Acadian Flycatcher and Wood Thrush populations in southern Ontario, respectively (Fig. 1). Acadian Flycatchers in Pennsylvania exhibit high levels of EPP (40% young) and males make long-distance off-territory forays for extra-pair matings (up to 1500 m away). Wood Thrushes breeding at the same site exhibit relatively low levels of EPP (6% young), despite the fact that both males and females make off-territory forays (Evans et al. 2008). Radio-tracking suggests that intensive male mate guarding, even off-territory, limits opportunities for extra-pair matings by female Wood Thrushes (Evans et al. 2008). Location of Wood Thrush (black circles) and Acadian Flycatcher (white circles) study sites in southern Ontario. The site where both species were studied in Pennsylvania is indicated by a black star. In southern Ontario, percent forest cover ranges between 3 and 33% (Burke et al. 2004; Friesen 2007), with many areas exhibiting relatively low forest cover compared to the 40% forest cover in northwestern Pennsylvania (Fraser and Stutchbury 2004). Wood Thrushes are common in southern Ontario (Friesen 2007), but Acadian Flycatchers are at the northern limit of their range and so uncommon that they are listed as an endangered species in Canada (Friesen et al. 2000). We predicted that levels of EPP in Wood Thrush would be comparatively low in both Ontario and Pennsylvania, but that the low density of Acadian Flycatchers in southern Ontario would result in a relatively low frequency of extra-pair fertilizations. Acadian Flycatchers and Wood Thrushes are both sexually monomorphic, migratory songbirds (Roth et al. 1996, Whitehead and Taylor 2002). Both species are socially monogamous, with males and females forming pairs and raising offspring jointly (Roth et al. 1996, Whitehead and Taylor 2002). Acadian Flycatchers (∼12 g) are area sensitive and typically nest in large tracts (>100 ha) of mature forest. The breeding range of Acadian Flycatchers covers most of the eastern United States and extends as far north as southern Ontario (Whitehead and Taylor 2002). Wood Thrushes (∼50 g) are widely distributed throughout the eastern United States and southeastern Canada (Roth et al. 1996). Although reported to be area sensitive throughout their breeding range (Roth et al. 1996), these thrushes readily inhabit forest fragments as small as 2 ha in southern Ontario (Friesen 2007). During the breeding seasons of 2002 and 2003, we monitored Acadian Flycatchers at six sites and Wood Thrushes at four sites (Fig. 1). Acadian Flycatcher study sites were located within 90 km of each other, whereas the Wood Thrush study sites were located within 30 km of each other (Fig. 1). For both species, we located singing males and nesting pairs to map breeding territories in forest fragments. Adults were captured in mist nests and banded with USGS aluminum bands and a unique combination of plastic color bands. A blood sample (25–100 μl) was taken via brachial vein puncture. We visited territories several times during the breeding season to verify occupancy, mated status of the male, and the identities of resident birds. We searched the territories of mated males to find nests and recorded nest locations in the field using hand-held GPS receivers (GPSMap76S, Garmin International, Olathe, Kansas). Nests were visited twice weekly to determine nest initiation date, clutch size, hatch date, and fledging success. Egg-laying dates ranged from late May to mid-July for both Acadian Flycatchers and Wood Thrushes. When nestlings were at least 5-d old, we banded and took a blood sample (25 μl) from each nestling. The size of each forest fragment was determined using ARCVIEW GIS 3.3 (ESRI, California) and database layers from the NRVIS database provided by the Ministry of Natural Resources (Aylmer, ON, Canada). We calculated percent forest cover within a 5-km radius of each study site using GIS (MapInfo; Canadian Topographic System 1:50,000; Riley and Mohr 1994). Breeding density in each forest fragment was calculated by dividing the number of males defending territories on each site by the total area (ha) of the fragment. For Acadian Flycatchers, we captured, banded, and obtained a blood sample from 85 ± 13% (range = 67–100%) of the males in each forest fragment and monitored 47 nests on 21 territories. The mean brood size for Acadian Flycatchers was 2.6 ± 0.6 (range = 1–4) nestlings. We monitored 107 Wood Thrush nests on 73 territories and sampled, on average, 74 ± 10% (range = 60–86%) of the males on each forest fragment in each year. The mean brood size was 2.7 ± 0.9 (range = 1–5) nestlings. Genetic analysis Blood samples were stored in 500 μl of Queen's cell lysis buffer at 4°C. DNA was extracted from whole blood using either a standard ethanol precipitation technique or Qiagen DNeasy Kit (Qiagen, Valencia, California). We used microsatellite loci EMIZ 1, EMIZ 27 (Tarof et al. 2002), SAP 22, SAP 32, and SAP 53 (Watson et al. 2002) to examine genetic variation and parentage of Acadian Flycatchers. Microsatellite loci Dpμ01 (Dawson et al. 1997), Cuμ 02, Cuμ 05, Cuμ 28, and Cuμ 32 (Gibbs et al. 1999) were used to characterize genetic variation of Wood Thrushes. All microsatellite loci were amplified using a standard PCR amplification protocol in 10 μl reactions containing 50 ng DNA, 1 μl 10 × PCR buffer (Biobasic, Markham, ON, Canada), 3.0 μl of 20 mM MgSO4, 0.2 μl of 10 mM reverse primer, 0.2 μl of fluorescently tagged (WellRED dye, Beckman Coulter, Mississauga, ON, Canada) 10 mM forward primer, 0.2 μl of 10 mM dNTP (New England Biolabs, Pickering, ON, Canada), and 0.25 U Tsg (Biobasic). Details of PCR protocols, including primer-specific annealing temperatures and cycle times, are reported elsewhere (Woolfenden et al. 2005, Evans et al. 2008). PCR products were visualized using the CEQ 8000 genetic analysis system (Beckman Coulter, Mississauga, ON, Canada). We genotyped 75 Acadian Flycatchers (46 nestlings, 15 females, and 14 males) at five microsatellite loci (Table 1). The cumulative exclusion probability, P(E) (Chakraborty et al. 1988), for the five loci was P(E) < 0.0001, indicating that these markers could identify extra-pair young (EPY) and their fathers with a high degree of certainty. We also genotyped 219 Wood Thrushes, including 116 nestlings, 57 females, and 46 males. The cumulative exclusion probability, P(E) (Chakraborty et al. 1988), for the five Wood Thrush loci was P(E) = 0.01. We identified social male parents using field observations of nestling defense and feeding. The genotype of each nestling was then compared to the social parents to determine if nestlings were sired by extra-pair males. Males were excluded as putative genetic parents if the male and nestling mismatched at two or more loci. Mismatches at one locus may result from PCR amplification errors, errors in fragment sizing, and mutations, so when a male and offspring mismatched at a single locus, we calculated the “cumulative probability of resemblance” (PRaCum; Ibariguchi et al. 2004). This statistic represents the probability that two individuals have their shared alleles in common (across all loci) by random chance. If PRaCum < 0.005 (less than 0.5% chance that the social parent and offspring share their common alleles by random chance), then the male was not excluded as the genetic parent. Because nestlings of both species exhibit low site philopatry ( 0.005 (range = 0.032–0.081), indicating that the social male was not genetically related to the offspring. Statistical analysis Observed and expected heterozygosity and tests for deviations from Hardy–Weinberg equilibrium for microsatellite loci were calculated in Genepop version 3.4 (http://genepop.curtin.edu.au/). All means are reported ± 1 SD and statistical analyses were performed using SPSS version 16 (SPSS, Chicago, Illinois). The number of Acadian Flycatcher territories ranged from 1 to 6 per site in forest fragments in southwestern Ontario, and breeding densities ranged from 0.005 to 0.015 males/ha (Table 2). Acadian Flycatchers suffered high levels of nest failure (60%), so we determined rates of extra-pair fertilization for only 14 broods from 12 pairs. Two families represent second broods. Four of 29 (14%) nestlings were the result of extra-pair fertilizations and 3 of 12 (25%) pairs had EPY. We found no EPY in the three forest fragments with a single breeding pair (Table 2). We identified the extra-pair sire for three of the four EPY. In all three cases the extra-pair male held a territory in the same forest fragment as the nest in which he sired EPY (mean distance between territories = 500 ± 250 m). Forest fragments supported 2–12 Wood Thrush territories per site and breeding densities ranged from 0.37 to 1 males/ha (Table 2). We found that 45 of 112 nestlings (40%) were EPY, and 67% (24 of 36) of pairs had EPY (Table 2). Most of these broods contained a single EPY (56%, 14 of 24). Despite sampling an estimated 75% of males per site each season, we identified the sires of only two EPY. Both extra-pair sires had territories adjacent to those with the nests containing their EPY. We found differences in the frequency of extra-pair matings between populations of Acadian Flycatchers and Wood Thrushes in southern Ontario and Pennsylvania. Acadian Flycatchers nesting in forest fragments in southern Ontario exhibited lower rates of EPP (14% nestlings) than those in Pennsylvania (40% nestlings; Woolfenden et al. 2005). In contrast, Wood Thrushes in Ontario exhibited higher rates of extra-pair matings (40% nestlings) than those in Pennsylvania (6% nestlings; Evans et al. 2008). Because forests are more highly fragmented in Ontario than in Pennsylvania, opportunities for extra-pair matings may differ for birds in these populations because of differences in breeding densities or behavioral tactics. Lower breeding densities in fragmented habitats may affect extra-pair mating opportunities by decreasing the proximity and availability of potential extra-pair mates (Banks et al. 2007). Our data suggest that the distribution of occupied fragments in southern Ontario is a likely constraint on extra-pair mating opportunities for Acadian Flycatchers and that opportunities for extra-pair matings may be limited to the few pairs sharing a forest fragment. Wood Thrushes in Ontario exhibited higher rates of extra-pair mating than those in Pennsylvania. However, breeding densities for populations in Ontario and Pennsylvania were similar (0.32 males/ha in Ontario vs. 0.30 males/ha in Pennsylvania; unpubl. data), suggesting that other factors mediate extra-pair mating opportunities in this species. Metapopulation studies indicate that the relationship between breeding density and EPP is not consistent across species but instead suggest that the interaction between density and the behavioral tactics exhibited by a species better explain differences in rates of EPP (Griffith et al. 2002, Westneat and Stewart 2003). Variation in the rates of EPP may also be influenced by the degree to which males and females engage in mating tactics such as off-territory forays or mate guarding (Westneat and Stewart 2003). Woolfenden et al. (2005) found that male Acadian Flycatchers in Pennsylvania moved off territories a mean distance of 526 m for extra-pair matings. In our study, male Acadian Flycatchers in Ontario occupied territories at a mean distance of 500 m from nests where they sired extra-pair young, suggesting that long-distance off-territory forays occur in both Pennsylvania and Ontario. Despite potentially similar off-territory foray tactics, the lower breeding densities in the fragmented forests of southern Ontario may reduce opportunities for extra-pair matings and contribute to the relatively low levels of EPP observed in Acadian Flycatchers in our study. In Hooded Warblers, males move greater distances and make more off-territory forays in fragmented habitats. However, this increased effort did not translate into increased rates of EPP (Norris and Stutchbury 2001, 2002, Stutchbury et al. 2005b). Studies of eastern and western populations of Red-winged Blackbirds (Agelaius phoeniceus) have also shown that mating tactics vary across populations, with male-driven extra-pair mating effort dominating in the east and female-driven extra-pair mating effort dominating in the west (Westneat 1993, Gray 1996). Although our study indicates that male Acadian Flycatchers obtained extra-pair matings at similar distances from their home territories in Ontario and Pennsylvania, further study is needed to determine if differences in mating tactics might have contributed to differences in EPP rates between the two populations. Male and female Wood Thrushes make off-territory forays in Pennsylvania, but rates of EPP are low, likely due to intense mate guarding by males (Evans et al. 2008). Thus, one factor potentially contributing to the higher levels of EPP in our study is reduced mate-guarding intensity by male Wood Thrushes. We are not aware of any studies in which variation in mate guarding across populations of a single species has been examined. However, the behaviors of birds nesting on habitat fragments may differ from those in more contiguous forests (Banks et al. 2007). Previous studies have indicated that populations on fragments exhibit a higher proportion of younger birds (Holmes et al. 1996, Zanette 2001, but see Burke and Nol 2001) and individuals in relatively poor condition (Rodewald and Shustack 2008) than do populations located in more contiguous forests. Young or lower quality males may not be able to guard mates as effectively as higher quality males (Estep et al. 2005, Charmantier and Blondel 2003, Bouwman and Komdeur 2005, Kokko and Morrell 2005). Alternatively, males in forest fragments may need to seek out resources across multiple habitat patches if food is scarce, thereby facing an implicit time and energetic trade-off between paternity assurance strategies, such as mate guarding, and foraging (Estep et al. 2005, Hinsley 2000, Fort and Otter 2004). Female Wood Thrushes in Ontario may be more likely to seek out higher quality males for extra-pair matings if habitat fragments provide fewer high-quality mating opportunities (Charmantier and Blondel 2003). Additional study will be needed to confirm whether differences in the observed rates of extra-pair mating across the landscape are due to differences in extra-pair mating tactics or paternity assurance behaviors of Wood Thrushes. Our study is the first to examine extra-pair mating by songbirds in a heavily fragmented habitat. From a mating system perspective, habitat fragmentation may increase the cost of seeking out extra-pair mates and limit the overall number of extra-pair mates available (Banks et al. 2007). We found differences in the degree to which two species of songbirds engaged in extra-pair matings in forest fragments in Ontario. Moreover, our results, in combination with those of other studies of Acadian Flycatchers and Wood Thrushes, indicate that rates of extra-pair mating can vary over a relatively short geographic distance, potentially due to increasing habitat fragmentation. Our results also indicate that, when developing conservation strategies to best maintain population function and viability, managers should evaluate the possible effects of habitat fragmentation on the mating systems of different species of birds. Funding was provided by the Canadian Wildlife Service, Environment Canada Ontario Region, World Wildlife Fund Canada (Endangered Species Recovery Fund), and an NSERC Discovery Grant and Canada Research Chair Award to BJMS. Field assistance was provided by B. Beveridge, D. Martin, A. Walpole, J. Wright, C. Zantinge, and H. Britton. Logistic support was provided by the ACFL/HOWA Recovery Team, Bird Studies Canada, D. Badinski, D. Burke, K. Elliot, D. Martin, P. Read, and S. Read. We thank the numerous private landowners, Catfish Creek Conservation Authority, Long Point Region Conservation Authority, Lower Thames Conservation Authority, and Middlesex County for granting permission to work on their land.