Environ. Sci. Technol. 2001, 35, 1845-1850
A Stable-Isotope Approach to Delineate Geographical Catchment Areas of Avian Migration Monitoring Stations in North America LEONARD I. WASSENAAR* AND KEITH A. HOBSON† National Water Research Institute, Environment Canada, 11 Innovation Boulevard., Saskatoon, SK, Canada, S7N 3H5
Migration monitoring stations (MMSs) were established to provide information on population trends of North American passerines. However, apart from inferring general origins of birds, there has been no way to delineate geographical catchment areas sampled by MMSs. The ability to resolve MMS catchment areas would greatly enhance our ability to link and constrain population declines to specific geographical areas and thereby focus conservation efforts. Here, we utilize stable-hydrogen (δD) and carbon (δ13C) isotope values of rectrices of fall hatchyear (HY) and spring adult after-second year (ASY) Swainson’s Thrush (Catharus ustulatus) moving through two MMSs in Canada to determine natal and molting catchment areas for those stations. Stable-carbon isotope signatures provided no information on geographical origins of birds. Conversely, δD signatures provided good latitudinal geographical control, and this was also supported by analysis of feathers from two other species, Graycheeked Thrush (Catharus minimus) and Harris’ Sparrow (Zonotrichia querula), also known to breed and molt in northern regions. The δD values of fall HY birds moving through Delta Marsh MMSs in Manitoba were more negative than those moving through Long Point MMSs in southern Ontario, clearly indicating more northwesterly origins. However, based on δD values alone, both MMSs sampled birds from broad catchment regions of the Canadian Boreal forest. The interpolated catchment area of the Delta MMS encompassed northwest Manitoba to northwestern Alberta. The interpolated catchment area of the Long Point MMS extended from north central Ontario and Quebec and into western Canada. The δD values of birds moving northward in spring suggested that most ASY Swainson’s Thrush molted at more southerly locations than their breeding sites. We show the stable-hydrogen isotope approach may be employed to significantly improve conventional observational techniques for avian population monitoring, and that MMSs provide a reliable means of associating population productivity with regional conservation issues.
Introduction Conservation of migrant songbirds breeding in North America has been the focus of considerable research in recent years in response to population declines observed over the past decades (1-5). A crucial component of avian conservation methodology is the interpretation of population trend 10.1021/es0010204 CCC: $20.00 Published on Web 03/29/2001
2001 American Chemical Society
analyses from observational monitoring and tag recovery data. Such observational methods include the North American Breeding Bird Survey (BBS) and the use of dedicated migration monitoring stations (MMSs) (6-8). MMSs that employ standardized constant-effort protocols (e.g., ref 6) offer a unique opportunity to sample birds derived from large geographical catchment areas, and may prove instrumental in assessing migrant songbird population trends for the Boreal forest region, an extensive but sparsely populated area of North America that is currently not well sampled by the BBS (Figure 1). The Boreal forest region is of particular concern due to expanded forestry activities and land conversions (711). Temporal population fluctuations or long-term declines observed using the MMS approach reflect events that occur within the catchment area sampled by the monitoring station (12). However, apart from inferring destinations or origins of birds based on knowledge of their overall breeding range and distribution patterns, there are currently no satisfactory methods to resolve the extent of the geographical catchment area being sampled by MMS. Band encounter efforts have not yielded useful results since recoveries for small birds are very low, geographically biased to marking locations, and do not correspond proportionally to bird densities. For example, according to the U.S. Geological Survey Patuxent Wildlife Research Center (http://www.pwrc.usgs.gov), of 359 000 Swainson’s thrush (Catharus ustulatus) banded between 1955 and 1998, only 242 have been recovered. Most encounters do not show movement vectors that are useful for determining breeding locations of migrants. In Canada, there were only 21 encounters of banded Swainson’s Thrushes more than 100 km from the banding site between 1921 and 1995, providing only fragmentary information on the breeding location of migrants (13). Such low encounter rates are typical of many small birds and other organisms, with the lowest rates associated with smallest species (Table 1). Thus, for many species, banding and marking is not a useful tool in assessing breeding origins. It was first demonstrated by (14) that stable-hydrogen isotope ratios (δD) in feathers from six species of songbirds, grown at 14 sites over a large-scale latitudinal gradient in North America (extending from Alaska to Louisiana) could be used to provide quantitative information on geographical origins where the feathers were grown. Because many species of passerines grow feathers on their breeding grounds shortly before autumn migration, the δD values of their feathers are correlated with δD of their breeding ground diet, which is in turn controlled by predictable large-scale meteorological patterns in δD of rainfall at the breeding site (14) (Figure 1). Extending this work, ref 15 used δD and δ13C to obtain quantitative information on migratory insect origins and by effectively linking wintering and breeding areas in Canada, the U.S., and Mexico. Since then, ref 16 showed that geographical δD patterns and additional δ13C isotope data could also be used to delineate populations of Red-Winged Blackbirds (Aegolius phoeniceus) associated with wetlands across the central part of North America. Others have proposed multiple stable isotope analyses (e.g., δ13C, δD, δ87Sr) to help delineate avian populations at more localized or regional scales (17, 18). * To whom correspondence should be addressed at the National Water Research Institute. Phone: (306) 975-5747; fax: (306) 9755143; e-mail:
[email protected]. † Canadian Wildlife Service, Environment Canada, 115 Perimeter Rd., Saskatoon, SK, Canada, S7N 0X4. VOL. 35, NO. 9, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1845
FIGURE 1. Contour map corresponding to δD values for feathers grown in North America and location of Delta Marsh Bird Observatory and Long Point Bird Observatory. Shaded area corresponds to the approximate distribution of the Boreal forest. Dashed line denotes the approximate northern limit of the Breeding Bird Survey. Contours are based on ref 14 with a correction factor of -25 ‰ to account for isotopic fractionation between rainfall and feathers (see text).
TABLE 1. Band or Tag Data and Percent Recovery for Selected Long Distance Migratory Organismsa species
no. banded
no. recovered
% recovery
Canada Goose Mallard Red-Winged Blackbird Swainson’s Thrush Ruby-t Hummingbird Monarch Butterfly
2 405 858 5 679 071 739 754 359 029 49 279 27 000
562 995 848 091 12 389 242 45 5
23.4 14.9 1.7 0.07
