Using stable-hydrogen isotopes to reveal immigration in an Arctic-breeding songbird population
Knowledge of immigration and emigration rates is crucial for full understanding of population dynamics, yet we know little about those rates in songbirds, especially arctic birds. In a German population of blackcaps Sylvia atricapilla in which all breeding adults and all fledged young were color-banded, 56 % of the breeding birds in the subsequent year were immigrants from other populations [7]. Other studies estimated immigration rates solely of yearling birds. In a Swedish population of collared flycatchers Ficedula albicollis, estimated immigration rates of yearling birds in unmanipulated control plots was ca 16 % [11]. In yellow-headed blackbirds Xanthocephalus xanthocephalus, Ward [12] recorded a proportion of yearling males of 14 % in Illinois and 24 % in South Dakota, respectively, and he cited several other North American studies of songbirds with annual percentages of yearling males ranging between 16 and 65 %.
For northern wheatears, our minimum estimate for all immigrants was at least 38 %, so at least 40 of the 104 adult males we captured were immigrants. Hatch-year birds made up 56.7 % of captured males, so 23 (about 22 %) of immigrant males were yearlings. These estimates fall within the range of values for temperate songbirds.
There is limited and often anecdotal data on breeding site fidelity of arctic songbirds. In a 6-year study of Lapland longspurs (Calcarius lapponicus) at Barrow, Alaska cumulative return of banded birds for all years yield 11.6 % for males and 23.5 %) for females [38]. At Sarcpa Lake, Nunavut, 47 of 86 (55 %) adult Lapland longspurs were detected as returnees in a subsequent year and 31 adults from an unknown number banded at McConnell River, Nunavut, returned the next year [39]. In a 3-year study of Eastern yellow wagtails (Motacilla tschutschensis) at Cape Romanzof, Yukon Delta National Wildlife Refuge, Alaska, at least 12 of 23 (52 %) banded males returned in the following year but none of 23 banded females returned, for an overall adult return rate of 26 % [40]. Fragmentary data on snow buntings (Plectrophenax nivalis) appear to indicate considerably lower breeding site fidelity: at Sarcpa Lake, Nunavut, 2 of 9 banded males and zero of 10 banded females returned the next year; on Devon Island, Nunavut, of 32 adults banded over a 3-year period, one male and two females returned the following year [41]; and at Iqaluit, Nunavut, only two (1 male and 1 female) of 22 adults fitted with geolocators in 2013 were found in 2014 (DJTH, pers. obs.). These figures suggest about 10 % return rate for adult snow buntings; still twice the rate of 5 % we found for northern wheatears by banding.
Although the use of stable isotopes provides a promising approach for estimating immigration and emigration in a population [17, 18, 20–23], there are several potential limitations to this approach. First, in most landscapes, stable isotopes do not provide the level of spatial accuracy to detect short distance (~ 100 km) movements, even when multiple isotopes are combined [23]. Thus, in our case, stable isotopes could underestimate the true number of immigrants. Second, our approach relies on population-specific values from known-origin tissues of a relatively large sample. Acquiring known-origin tissue can involve intensive fieldwork, as it typically requires locating and monitoring nests or catching juveniles soon after they leave the nest. A less intensive approach would be to characterize local ?2H distribution based on interpolated precipitation values [27] and then offset this distribution with a diet-tissue discrimination factor [29, 30]. We chose not to take this approach because we did not have an estimated diet-tissue discrimination factor for Arctic-breeding populations of this species or an ecologically similar species. However, obtaining such an estimate would, in theory, also allow us to assign the geographic origin of immigrants with a certain degree of confidence. Third, different diets of juveniles versus adult birds could have influence on the isotopic signatures of feathers. However, there is no evidence that adults and juvenile wheatears differ in their diets at the breeding grounds (e.g. [42–44]). Metabolism is certainly different in growing birds as compared to adults. But Storm-Suke et al. [45] provide experimental evidence in quail that metabolic rate does not influence diet-tissue discrimination in hydrogen isotope values. Moreover, the consistency in average similarity in the isotopic signatures between juvenile and adult wheatear tail feathers hints on no substantial effects of age.
Immigration and emigration are crucial factors driving demography and dynamics of bird populations but not much is known about their magnitude or their annual variation. The use of stable isotopes may help fill this gap. In our study, isotope measurements of the feathers of northern wheatears indicated a high rate of immigration into the breeding population, which is consistent with low return rates of banded breeding adults as well as implying high emigration rates of local breeders. If emigration of adults is high throughout their breeding lives, calculations of apparent survival may be inaccurate or impossible, and may not closely reflect true survival. However, in songbirds with high site-faithfulness once they become breeders [1], apparent survival of adult birds as revealed by return rates may not be far off of true survival.