INTRODUCTION

Recent studies of the population dynamics of migratory songbirds have enormously increased our understanding of both local and large-scale correlates of abundance and nesting success and improved our ability to detect future population changes (e.g., Martin and Finch 1995, Maurer and Villard 1996). In general, abundance, pairing success, and nesting success have been shown to be positively correlated with tract size (reviewed in Faaborg et al. 1995). In addition, nesting success has been shown to vary at the landscape scale (e.g., percentage forest cover within a 10-km radius of study sites in the American Midwest: Robinson et al. 1995b). Pairing success and nesting success, but not necessarily abundance, have further been shown to vary locally (e.g., with distance from edges: reviewed in Paton 1994, Faaborg et al. 1995).

In many cases, observed gradients in nesting success are steep enough to suggest a mosaic of population sources and sinks (sensu Pulliam 1988) in fragmented landscapes. Productivity in population sinks is insufficient to compensate for adult mortality, whereas population sources produce a surplus of young that are then available to recolonize or "rescue" (Brown and Kodric-Brown 1977) sink habitat. Robinson (1992) and Brawn and Robinson (1996), for example, concluded that nesting success was so low in small agricultural woodlots in Illinois that populations nesting in these woodlots were almost certainly being rescued by immigrants from very distant (>100 km away) sources. Robinson et al. (1995b) further suggested that, for forest birds, the American Midwest consists of many small sink populations and only a few large source populations. This hypothesis has been partially corroborated by demographic modeling (Donovan et al. 1995a) and by at least one intensive demographic study of a forest songbird, the Wood Thrush (Hylocichla mustelina) (Trine 1988). In all of these studies, low nesting success in sink habitat resulted from the combined effects of high rates of nest predation and brood parasitism by Brown-headed Cowbirds (Molothrus ater).

Studies such as these have greatly influenced the general approach to the North American Bird Conservation Plan (NABCP). A common-sense application of these results is to protect and enhance large habitat patches that have high nesting success (presumed source populations) and to reduce edges and increase tract size where nesting success is low (presumed population sinks). Any management action that increases levels of nest parasitism or nest predation can be challenged as a potential threat to bird populations.

As we will argue later in this paper, these management recommendations offer an excellent starting point. We will also argue, however, that the long-term success of the NABCP depends upon continued intensive demographic studies of representative species, both on the breeding and wintering grounds. Until we can better estimate key demographic parameters (e.g., survival, annual productivity, number of broods per year), we cannot determine the balance of source and sink habitat necessary to maintain migrant bird populations and reverse population declines. Nor can we determine the consequences of management actions on local population viability. Some species may be relatively unaffected by substantial decreases in local nesting success, whereas others may be vulnerable even to slight increases in rates of nest predation and brood parasitism. Without better demographic data, population models such as those of Pease and Grzybowski (1995) and Donovan et al. (1995a) will be difficult to apply.

In this paper, we provide examples of insights drawn from ongoing demographic studies in the American Midwest and explore the implications of these studies for bird conservation planning. While we focus on demographic variables associated with the breeding season (survival, decision rules governing site fidelity, season-long nesting success, the consequences of single- versus double-broodedness), we acknowledge that studies of survivorship in non-breeding habitats may also reveal steep survival gradients, as argued by Rappole (1995). We then advocate and describe a research program aimed at accumulating enough demographic data to model continental and even hemispheric population dynamics of a few representative migratory songbirds such as the Wood Thrush. We conclude by reviewing examples of direct applications of demographic studies to conservation planning.

JUVENILE DISPERSAL AND THE POPULATION DYNAMICS OF MIGRATORY BIRDS

Population dynamics of most migratory birds differ markedly from those of more intensively studied year-round resident species (e.g., Parids, cooperative breeders). Migratory bird populations can be limited by events occurring on widely separated breeding and nonbreeding habitats (reviewed in Sherry and Holmes 1995). This difference is widely acknowledged and is the foundation for hemispheric conservation plans for Neotropical migrants. An equally important, but less-appreciated difference is that most, but not all, migratory songbirds disperse far from their natal areas (Weatherhead and Forbes 1994). Because so few juveniles return to their natal sites, local population dynamics and reproductive success can be uncoupled (Brawn and Robinson 1996). Models that link local population collapse with fragmentation-related declines in nesting success (e.g., Urban and Shugart 1986, Temple and Cary 1988, Thompson 1993) may not be applicable to migratory birds, which can maintain locally stable populations over the long term even in habitat patches in which essentially no young are produced (Brawn and Robinson 1996). Such populations can be rescued by dispersing birds; for this reason, migratory birds are usually characterized as having source/sink population dynamics (e.g., Donovan et al. 1995a,b). The capacity to recolonize even very poor habitat makes migratory songbirds vulnerable to "ecological traps," which are habitats that attract birds but fail to provide the conditions for successful nesting (Gates and Gysel 1978). The fact that migratory birds can maintain stable populations even in poor habitat (e.g., Brawn and Robinson 1996) makes their abundance and population trends questionable as indicators of site quality (Van Horne 1983), unless nesting success is also measured.

Understanding the population dynamics of migratory birds therefore requires answering the following questions. 1) What is the threshold of nesting productivity and survival that separates population sources and sinks? 2) What determines the "decision rules" adult birds use in making dispersal decisions? 3) Do these decision rules render species more or less vulnerable to ecological traps? 4) What cues are used by dispersing juveniles when selecting breeding sites? 5) Do the decision rules used by juvenile birds make them more or less vulnerable to ecological traps? And 6) is there density-dependent dispersal from source habitat?

THE SOURCE/SINK THRESHOLD

The identification of population sources and sinks on the breeding grounds demands data on season-long per-capita productivity and the survival of adult and juvenile birds (Pulliam 1988). The source/sink threshold is the annual productivity per pair necessary to compensate for adult mortality and is defined as

Because at least some adults return to their breeding territories, we can obtain minimal estimates of survivorship based on returns of marked birds. Adult site fidelity, however, is not absolute (Marshall et al., this volume). Many intensively studied species use a predictable set of rules when deciding whether or not to return to a site. In general, birds are more likely to return if they have nested successfully both within a season (Robinson 1985, Jackson et al. 1989) and between years (Greenwood and Harvey 1982, Robinson 1985, Bollinger and Gavin 1985, Gavin and Bollinger 1989). In some cases, these "decision rules" vary with age and sex (Robinson 1985, Trine 1996). Usually, young birds are less likely to return following reproductive failure in the previous year (Robinson 1985), but in at least one species, the Wood Thrush, older birds are less likely to return following failure (Tine 1996). Decision rules also can vary with sex; usually, males are more likely to return than females. In the Kentucky Warbler (Oporornis formosus), females have higher return rates following reproductive success regardless of age, whereas males take reproductive success into account only following their first year of breeding at the site. Reproductive success has no impact on future dispersal decisions by male Kentucky Warblers following their initial return (S. F. Morse, unpubl. data). In some species, however, there are no sex-based differences (Trine 1996).

It may be possible to improve survival estimates of songbirds that follow these predictable rules by identifying the conditions under which individuals are most likely to return, and to use these return rates as survival estimates. In southern Illinois, for example, Wood Thrushes that fledge two broods in their second year in a site return to that site at a rate of over 80% (Tine 1996). In contrast, Kentucky Warblers nesting in the same area never return at rates higher than 60%, even long-term residents that nest successfully (S. F. Morse, unpubl. data). Similarly, even the most successful Black-throated Blue Warblers (Dendroica caerulescens) only return at rates of <50% (Holmes et al. 1992, 1996). Because the most reproductively successful individuals may also be the fittest, return rates of successful birds may overestimate survival rates for all individuals in a population, or at least be less likely to be biased toward underestimating survival. For this reason, experimental manipulations randomly assigning nesting success within several populations may be necessary to control for fitness differences among individuals. Such experiments are under way for at least one Neotropical migrant, the Prothonotary Warbler (Protonotaria citrea) (J. P. Hoover, unpubl. data). In general, however, the use of probability-based survival models (Martin et al. 1995) may be preferable to using return rates for estimating survival.

Because of differences in survival rates (Martin and Li 1992, Peach 1993) the source/sink threshold varies greatly among species. Wood Thrushes, for example, may need to produce only 1.7 young/pair/season, assuming a 70% adult survival rate and a 35% juvenile survival rate (Roth and Johnson 1993). In contrast, for Kentucky Warblers (Morse 1996) and Black-throated Blue Warblers (Holmes et al. 1992) to exceed the estimated source/sink threshold, their productivity needs to be in the order of 2.7-4.0 young/year, assuming adult survival rates of 50-60% and juvenile survival rates of 25-30%. Pease and Grzybowski (1995) found that estimates of the source/sink threshold were more sensitive to adult survival than to any other variable.

The strength of sources and sinks

Most studies of the season-long productivity of migratory birds conclude either that populations are sinks (Tine 1998, Thompson and Nolan 1973) or that productivity is at or just slightly above the source/sink threshold (Holmes et al. 1992, Nolan 1978). In southern Illinois, Kentucky Warbler productivity even in the "best" habitat (old, interior forest) is at the lower end of the estimated source/sink threshold (Morse 1996; Table 1), even though nest predation and parasitism rates are as low as they are anywhere in the Midwest (Robinson et al. 1995b). Similarly, Black-throated Blue Warbler productivity is barely above the estimated source/sink threshold even in optimal habitat (presence of a dense shrub layer) in a large, unfragmented eastern forest (Holmes et al. 1992, 1996).

Table 1. Rates of Brown-headed cowbird brood parasitism, rates of nest predation, and the per-capita production of young in a banded population of Kentucky Warblers breeding in a heterogeneous forested site in the Shawnee National Forest in Union County, Illinois, from 1994-1995. The data presented here is a subset of that collected from 1992-1996 (Morse 1996). The core forest was relatively mature (70-120 years old) and farther than 1 km from a suitable cowbird feeding area in a nearby agricultural plot. Marginal forest was within 1 km of this plot and consisted of recently disturbed tracts (tree plantations, clearcuts of various ages, and a residential clearing) and a small amount of mature forest. Nest success was calculated using Mayfield’s (1975) method.

The studies cited above strongly suggest that, for many species, sources are weak even under the best local conditions. Clearly, if sources are weak, any factor that increases nest predation and brood parasitism is more likely to push populations below the source/sink threshold, and decrease the pool of potential immigrants for sink habitats. Extensive source habitat may be necessary in order to balance losses in strong sinks, especially if sink habitats act as ecological traps. If sources are typically weak, annual variation in predation rates within source habitat (e.g., for Acadian Flycatchers [Empidonax virescens] in southern Illinois: Trine et al., 1998; and Wood Thrushes in Missouri: Anders et al. 1997) also might be sufficient to create temporal sources and sinks.

If we are systematically underestimating annual survival, then our estimate of the source/sink threshold may be too high (Marshall et al., this volume). The high rates of juvenile mortality measured by Anders et al. (1997), however, suggest that mortality within the first few weeks of life alone may equal estimates commonly used to approximate juvenile mortality for an entire season. Recent modeling efforts, however, suggest that adult survival may be higher than usually estimated (Martin et al. 1995, Chase et al. 1997). Given these uncertainties, however, it seems prudent to assume that we need large areas of source habitat to buffer annual variation in productivity, and to compensate for human-induced "black hole" population sinks associated with habitat fragmentation.

Spatial scale of sources and sinks

Robinson et al. (1995b) and Donovan et al. (1995a) suggested that, for a variety of species, nest predation rates and brood parasitism levels (and thus source/sink dynamics) depend on the distribution of forest cover at the scale of the landscape (operationally defined as the area of core forest within 10 km of the focal patch). Indeed, Wood Thrush nesting success does not depend on forest tract size in the highly fragmented Shawnee National Forest: even in the largest tracts, reproduction is below the estimated source/sink threshold. By contrast, studies of Kentucky Warblers in the same region suggest that reproductive success is linked to both local (within-site) and landscape-level (e.g., tract size) habitat features (Morse and Robinson, in press). For this species, nest predation decreases with forest age, and parasitism levels decrease with increasing distance to a suitable cowbird foraging area. The resulting gradient in reproductive success within this site appears steep enough to produce source/sink dynamics at the scale of the forest tract (Morse 1996; Table 1). Black-throated Blue Warblers also show possible source/sink dynamics in relation to stand age (Holmes et al. 1996).

IMPACTS OF NEST PREDATION AND BROOD PARASITISM ON ANNUAL FECUNDITY

Methods for assessing the relative impact of brood parasitism and nest predation on annual fecundity, such as proposed by Pease and Grzybowski (1995), are likewise crucial for modeling the source/sink threshold and for designing effective conservation strategies. Clearly, reducing cowbird populations is easier than controlling the myriad predators that attack nests. Cowbird populations can be controlled locally through trapping (reviewed in Robinson et al. 1993, 1995a) or at larger spatial scales by eliminating key cowbird feeding areas. If, however, cowbird parasitism has only a minor impact on annual fecundity as argued by Martin (1992), then cowbird control may have little net impact on local bird populations.

The impacts of cowbird parasitism and nest predation on annual fecundity depend upon many variables (Pease and Grzybowski 1995). For example, single- vs. double-brooded species may respond very differently to nest predation. Even a twofold difference in predation rates near clearcuts had little effect on annual productivity in a single-brooded warbler, the Ovenbird (Seiurus aurocapillus), because pairs that lost nests to predators would renest until most were successful at least once (King et al. 1996). Prairie Warblers (Dendroica discolor) also were able to compensate for very high (>80%) rates of nest predation by frequent and rapid renesting. In contrast, Kentucky Warblers (a facultatively double-brooded bird) produced almost twice as many young in habitats with low (<60%) nest predation levels than in those with high (>60%) levels (Morse 1996; Table 1). Without the production of a second brood, Black-throated Blue Warbler populations in presumed source habitat in New Hampshire would have been sinks (Holmes et al. 1992).

Brood parasitism interacts in complex ways with nest predation and the number of broods produced (Pease and Grzybowski 1995). Heavily parasitized single-brooded species would suffer substantial losses in seasonal fecundity, because in most of these species, parasitized nests fledge few host young (May and Robinson 1985). Double-brooded species would have more opportunities to escape parasitism, especially if the second nesting attempt occurred in mid-to-late summer, when cowbirds have ceased breeding. Because at least some host young are produced from parasitized nests, producing two broods would still increase host nesting success, possibly above the source/sink threshold. Because Wood Thrushes are large enough to raise mixed broods, some double-brooded populations are close to the source/sink threshold despite high parasitism levels (100%) and intensity (2-3.5 cowbird eggs per nests) (Trine 1998). Very high nest predation rates also might reduce the impact of parasitism by destroying early season nests that often suffer higher rates of parasitism. Shrubland/edge species may produce most of their young late in the season after cowbirds have finished breeding and dense annual growth of grasses and forbs reduces nest predation rates (S. K. Robinson and S. F. Morse, unpubl. data).

These factors together create fertile conditions for the use of population models such as those proposed by Pease and Grzybowski (1995). Such models can detect threshold levels of nest predation and brood parasitism above which populations are likely to be sinks, and could help evaluate the effectiveness of efforts to control cowbirds or nest predators. Once we have better demographic data, these models eventually should enable us to make reasonable estimates of seasonal fecundity where nest predation rates and brood parasitism levels are the only data available.

DETERMINING VULNERABILITY TO ECOLOGICAL TRAPS

Three kinds of data are necessary to understand local population dynamics and the vulnerability of a species to ecological traps. First, we need to understand the behavioral decision rules governing site fidelity in adults and, if possible, natal philopatry in juveniles. Second, we need to determine the cues used by juveniles when selecting a breeding site. And third, we need population monitoring data from a wide geographic area and a number of habitats to estimate the extent to which breeding birds saturate suitable habitat. In this section, we discuss each of these research needs.

Decision rules governing adult site fidelity

As discussed above, adult birds tend to disperse within a season if their nests are destroyed early in the season (Greenwood and Harvey 1982, Robinson 1985, Jackson et al. 1989, Trine 1996) and tend to not  return the following year if they fail to produce young, or, for multiple-brooded species, if they fledge fewer than two broods (Roth and Johnson 1993, Tine 1996, but see Payne and Payne 1993, Hoover et al. 1995). Although site fidelity varies with sex and age in some species, all of these studies suggest that adults have a behavioral mechanism enabling them to avoid areas of chronically high nest predation. Viewed in this light, the well-documented area sensitivity of many migratory songbirds (e.g., Robbins et al. 1989) may represent adaptive avoidance of small tracts in which nest predation rates are generally high (reviewed in Faaborg et al. 1995). Small woodlots with low nest predation rates can contain high densities of some Neotropical migrants. Roth and Johnson (1993), for example, found high population densities and high adult site fidelity of Wood Thrushes in a small Delaware woodlot in which Wood Thrushes experienced low nest predation rates. The avoidance of small woodlots by ground-nesting warblers in the American Midwest (Blake and Karr 1987, S. K. Robinson, unpubl. data) may reflect the abundance of raccoons and resulting high nest predation rates (Brawn and Robinson 1996, E. J. Heske, unpubl. data).

Even within sites, decision rules can lead to avoidance of sink habitat. Tine (1996) found that Wood Thrushes abandoned one study site after three years of high nest predation rates. The site was not recolonized for at least 5 years; in contrast, population densities in two nearby areas with lower nest predation rates remained consistently high. In sites with low predation rates, fidelity of young Wood Thrushes was high regardless of their own nesting success, suggesting that their behavioral decisions may have been influenced by the success of their neighbors (Trifle 1996). Holmes et al. (1996) found that Black-throated Blue Warblers breeding in optimal habitat were much more likely to return the following summer than those breeding in suboptimal habitat.

Habitat saturation and population sinks

Wood Thrush population densities in southern Illinois are generally low and highly variable from year to year, even in apparently suitable habitat (S. K. Robinson, unpubl. data). In contrast, Kentucky Warblers maintain high population densities in all apparently suitable ravine-bottom habitat, and quickly establish high populations in habitat created by activities such as selective logging (Robinson and Robinson, in press). These two species show dramatically different population-level responses to spatial variation in nest predation rates. For the Wood Thrush, ravines with high predation rates can remain unoccupied for years and, within and among ravines, individuals tend to concentrate in areas with low predation rates. Kentucky Warblers, on the other hand, occur at high population densities in suboptimal habitat, even after several years of high predation rates, which suggests a saturated population. Black-throated Blue Warblers similarly show a tendency to occupy optimal habitat first and then "spill over" into suboptimal sink habitats, which have a higher proportion of young birds. Petit and Petit (1996) found a similar tendency for Prothonotary Warblers to occupy habitats with low nest predation rates first.

Black-throated Blue and Prothonotary Warblers therefore conform to Fretwell and Lucas’ (1970) models of "ideal despotic" habitat selection, in which optimal habitats are occupied first and suboptimal habitats are occupied only by surplus individuals. When populations are below saturation (e.g., Wood Thrushes in southern Illinois), few individuals are forced to nest in habitats with high predation rates. Some individual Kentucky Warblers, Black-throated Blue Warblers, and Prothonotary Warblers, on the other hand, are forced to nest in areas with high predation rates.

The decision rules governing site fidelity appear to be adaptive with respect to nest predation, at least for some species. Sink habitat created by high predation rates may be occupied only if a surplus of individuals is produced in source habitats, in which case sink habitat does not function as an ecological trap. As long as productivity in population sinks exceeds the losses incurred while breeding (as opposed to floating), sink habitat may benefit populations by providing breeding opportunities for birds that might not nest otherwise (Howe et al. 1991). Long-term persistence of regional metapopulations therefore depends more on the continued health of source populations than on the extent of sink habitat. A conservation strategy focusing on large, unfragmented tracts can guarantee high regional populations by enhancing the production of surplus individuals available to recolonize even small habitat islands (Brawn and Robinson 1996), an approach advocated in the North American Bird Conservation Plan. Decreases in source habitat caused by fragmentation could cause populations to retreat to remaining source habitat and reduce the number of surplus birds available to recolonize isolated habitat patches. In this situation, the population declines may be analogous to Wilcove and Terborgh’s (1984) melting ice cube model of range contraction, because suboptimal, marginal habitats will be abandoned first. Regional populations will continue to decline only with further loss of source populations. As long as source habitat is not reduced, an increase in the amount of sink habitat will cause no further population declines. Adaptive decision rules therefore make populations less vulnerable to population sinks created by high rates of nest predation.

Decision rules in response to nest predation and the number of broods fledged suggest a long evolutionary history of responding to predation. For many North American species, however, cowbird parasitism has become a threat only recently, as the Brown-headed Cowbird increased both its geographical range and global population (reviewed in Robinson et al. 1995a). This phenomenon is so recent (<50 years in some areas of the West) that many species may not yet have had the opportunity to evolve resistance (evolutionary "lag": Rothstein 1975). Birds of the eastern deciduous forest, in particular, seem to lack any defenses against parasitism. They tend not to eject cowbird eggs or abandon parasitized nests, nor do they select nest sites that are inaccessible to cowbirds (e.g., cavities). Many prairie and shrubland birds, in contrast, have a variety of defenses against parasitism (reviewed in Robinson et al. 1995a), which suggest a long history of coevolutionary interactions with cowbirds (Robinson et al., in press, Rothstein and Robinson, 1998).

This lack of defense against cowbird parasitism in forest birds also appears to extend to decision rules governing site fidelity. Wood Thrush populations breeding in fragmented midwestern forests, in which virtually every female is parasitized, return at rates that are as high or higher than those breeding in eastern forests where parasitism rates are low (Roth and Johnson 1994, Hoover et al. 1995, Trine 1996). Return rates of Wood Thrushes to two sites in southern Illinois that are likely population sinks (Trifle 1998) averaged 50-80% (Trine 1996). Nest predation rates were low in both of these sites, but cowbird parasitism was so intense (85-95% of nests parasitized, 2-4 cowbird eggs/nest) that annual productivity was well below the estimated source/sink threshold (Tine 1998). Return rates were identical for parasitized and unparasitized female Kentucky Warblers at another site in southern Illinois (S. F. Morse, unpub. data), where parasitism ranged from 15-70%, creating likely population sinks in some areas with low nest predation rates. While predation on Kentucky Warbler nests varied little between mature forest stands within this site, there was a twofold difference in annual productivity due primarily to differences in parasitism rates (17% vs. 50%) (Morse 1996). Prothonotary Warblers likewise show no effect of parasitism on site fidelity (J. P. Hoover, unpubl. data).

These studies from the American Midwest suggest that many forest hosts lack adaptive responses to brood parasitism. As a result, areas with high cowbird parasitism but low nest predation are likely to qualify as ecological traps. Most of Illinois appears to be an ecological trap for the Wood Thrush, which may be causing the regional population declines shown by this species (Peterjohn et al. 1995, Roth et al. 1996). The ecological trap scenario, however, is probably not responsible for the declines of Wood Thrushes in the Northeast, where cowbird parasitism is much less of a problem, and forest area is increasing (Roth et al. 1996).

These results suggest an additional, hidden cost of parasitism which might increase the benefit of efforts to reduce cowbird parasitism. Because cowbird parasitism and nest predation tend to be positively correlated both at a local scale (Temple and Cary 1988, Donovan et al. 1995b) and at the landscape/regional scale (Robinson et al. 1995b), however, species that avoid areas with high nest predation may also be, on average, avoiding areas of high parasitism.

In addition to adult site fidelity, the attractiveness of habitat patches to juveniles also will affect local population dynamics. Usually, habitat selection is thought to be based on a "niche gestalt" involving such variables as vegetation structure and food availability (James et al. 1984). An additional possibility is that dispersing birds use the presence of conspecifics to evaluate habitat quality (Stamps 1988). "Conspecific attraction" is usually considered to be a problem for avian metapopulations because once a population becomes extinct, the lack of conspecifics might prevent recolonization of the habitat patch in a negative feedback loop (Smith and Peacock 1990, Ray et al. 1991). If, however, the lack of conspecifics signals poor habitat, then conspecific attraction could be adaptive. Because adults appear to return to areas in which they have nested successfully, the presence of conspecifics should be a good predictor of habitat quality, especially in habitat patches in which predation rates are predictably low. Failure to recolonize small habitat patches in which the adult population has become extinct may be adaptive for individuals, and may benefit populations as well, if predation rates are chronically high. We do not yet know the extent to which juveniles use the presence of conspecifics when selecting nesting sites. Some species seem to form aggregations within otherwise suitable habitat (e.g., Cerulean Warbler Dendroica cerulea: Robbins et al. 1992; G. C. Vanderah, unpubl. data; Kirtland’s Warbler Dendroica kirtlandii, Probst 1986), which may result from conspecific attraction. Declining populations may benefit from conspecific attraction, because it helps individuals find mates and would tend to concentrate birds in areas of low nest predation. Experimental studies of conspecific attraction are needed to test this hypothesis.

Birds often show site fidelity to wintering grounds comparable to that shown by breeding birds (reviewed in Rappole and McDonald 1994, Rappole 1995). We do not yet know the extent to which the decision rules governing site fidelity reflect variables such as food, cover, and condition. Conway et al. (1995) and Niven (1996) found few differences in condition or return rates among habitats varying greatly in disturbance. Winker et al. (1990), on the other hand, found that return rates of Wood Thrushes varied greatly among habitats. One of the key predictions of Rappole and McDonald’s (1994) hypothesis that populations of Neotropical migrants are limited on their wintering grounds was that migrants saturate available winter habitat, which forces many birds into suboptimal habitat, and forces others to live as semi-nomadic floaters. Some of the species that do not saturate their breeding habitat, such as Wood Thrushes in southern Illinois, may be limited by winter rather than breeding-habitat availability. Clearly, we need more winter-season demographic studies of such factors as site fidelity and habitat-specific differences in site fidelity and condition.

RESEARCH RECOMMENDATIONS

As we argued at the outset of this paper, the success of the North American Bird Conservation Plan will depend upon improved spatially explicit population models for key species. These models rely on estimates of the source/sink threshold and vulnerability to ecological traps, and are very data intensive, requiring better demographic data than generally are available for most species (Pease and Grzybowski 1995). For this reason, we advocate research leading toward the creation of a library of demographically well-studied species that can be used for spatially explicit modeling, building on such large-scale projects as MAPS and BBIRD. This research would include intensive studies of color-marked populations on both the breeding and wintering grounds. Ideally, demographic studies would be long term, and conducted in several different parts of the breeding and wintering range. Radio-telemetry of fledglings on the breeding (e.g., Anders et al. 1997) and wintering grounds (e.g., Winker et al. 1990) will help improve estimates of nonbreeding survival. Such studies also will enable us to use the more powerful versions of Pease and Grzybowski’s (1995) models when the only data available are on rates of nest predation and cowbird parasitism.

The species chosen for these studies should reflect both the ease of study and their conservation significance, and should be representative of the major taxonomic groups and nesting guilds. For the eastern deciduous forest, we already have extensive data on Wood Thrushes (reviewed in Roth et al. 1996), Kentucky Warblers (S. F. Morse, V. McDonald, unpubl. data), Black-throated Blue Warblers (reviewed in Holmes et al. 1996), Acadian Flycatcher (D. Whitehead, unpubl. data), Prothonotary Warblers (Petit and Petit 1996, J. P. Hoover, unpubl. data), and American Redstarts (Setophaga ruticilla) (Sherry and Holmes 1992). There also are attempts to work with canopy species such as the Scarlet Tanager (Piranga olivacea) (Rosenberg et al. 1999) and the Cerulean Warbler (P. Hamel, pers. comm.). We currently lack good demographic data from Neotropical migrants nesting in spruce woods. Other habitats and geographical areas will require different indicators.

CONSERVATION APPLICATIONS OF DEMOGRAPHIC STUDIES

Here we provide examples of a general conservation plan based on our demographic data from the American Midwest.

1. Conservation efforts should focus first on preserving and enhancing the largest forest (and grassland) tracts within any region. Such tracts are the best candidates for providing population sources for a wide variety of species. These areas may need to be very large if source populations prove to be weak (i.e., they produce only a small per-capita surplus). The reforestation of much of the northeastern United States may provide adequate source habitat for forest birds for this region. In the American Midwest, however, regional populations may depend upon just a few large forest tracts (Robinson et al. 1995b, Donovan et al. 1995b). For preferred cowbird hosts such as midwestern Wood Thrushes, the size of tracts necessary to provide source habitat may need to be very large, given the 7-km commuting range of cowbirds between feeding and breeding areas (Thompson 1994). Because of consistently heavy parasitism, even tracts of 3,000 ha in southern Illinois are likely population sinks (Trine 1998). For this reason, we advocate focusing on tracts of at least 10,000 ha; in states such as Illinois that lack tracts of this size, maintaining Wood Thrush source populations within the state may entail extensive habitat restoration. Alternatively, a regional conservation plan focusing on source populations outside the state may be the most practical strategy for this species. Virtually all forest species, even those that are not preferred cowbird hosts, will benefit from the preservation, enhancement, and restoration of these "macrosites." Even disturbance-dependent species such as the Indigo Bunting (Passerina cyanea) appear to nest more successfully on edges within large forest tracts (Suarez et al., 1997).

2. The next priority (after focusing on the largest tracts) for conservation planning should be the largest tracts available within a landscape (or a subregion, such as southern Illinois), even if they are too small to provide source habitat for all species. The source/sink dynamics of less-preferred cowbird hosts such as the Kentucky Warbler may be largely determined by rates of nest predation, and potentially occur on a much smaller scale than that for the Wood Thrush. Although Kentucky Warblers would certainly benefit from extensive forest tracts necessary for the maintenance of Wood Thrush source populations, site management practices within smaller tracts also strongly affect Kentucky Warbler reproductive success, at least in southern Illinois. A management strategy aimed at medium-sized tracts ("mesosites"), which preserves core areas of undisturbed or lightly disturbed forest far (>1 km) from agricultural edges, might increase source habitat for species such as the Kentucky Warbler. Potential population sources exist for Kentucky Warblers in tracts of 900-3000 ha in the Shawnee National Forest. Within these tracts, nesting productivity could be enhanced further by using uneven-age rather than even-age forestry management, and by eliminating or modifying the management of specific agricultural openings (e.g., feedlots) and recreational areas (e.g., campgrounds with mowed grass) in which cowbirds feed. The "Forest Interior Management Units" management plan for the Shawnee National Forest (USDA Forest Service 1986), which emphasizes uncut "core" areas and uneven-age silviculture within tracts of at least 1100 acres, should benefit Kentucky Warbler populations. More workneeds to be done, however, to determine the source/sink dynamics of other species within these tracts before we can fully assess the value of "mesosites."

3. Relatively small tracts ("microsites") should be the lowest priority for the conservation of breeding forest birds, but then may be extremely important as migratory stopover habitat. In Illinois, small tracts (<200 ha) are characterized by maximal rates of nest predation and cowbird parasitism (Robinson et al., in press, S. K. Robinson, unpubl. data). Although some small woodlots undoubtedly contain viable populations of migrant birds (e.g., Roth and Johnson 1993), many are likely to be "black hole" population sinks (e.g., Robinson 1992, Brawn and Robinson 1996). Instead of focusing on breeding birds, management of small tracts should aim toward improving the improvement of migratory stopover habitat.

4. Within all tracts, ecosystem processes should be carefully monitored and, where possible, restored. It is easy to envision how intensive deer herbivory could destroy source habitat for Black-throated Blue Warblers (Holmes et al. 1996) and  Kentucky Warblers (McShea et al. 1995) by reducing shrub density and ground cover. The loss of tree species by selective deer browsing also could have a large effect on bird populations of northern forests (Alverson et al. 1988, Holmes and Robinson 1981). Altered hydrological cycles in floodplain forests would effect the source/sink dynamics of Prothonotary Warblers (Petit and Petit 1996, J. P. Hoover, unpubl. data), by affecting the availability of safe nest sites. We are just beginning to appreciate the extent to which habitat structure and composition affect the population dynamics of migratory birds.

LITERATURE CITED

Alverson, W. S., D. W. Waller, and S. L. Solheim. 1988. Forests too deer: Edge effects in northern Wisconsin. Conservation Biology 2:348-358.

Anders, A. D., D. C. Dearborn, J. Faaborg, and F. R. Thompson. 1997. Juvenile survival in a population of Neotropical migrant birds. Conservation Biology 11:698-707.

Blake, J. G., and J. R. Karr. 1987. Breeding birds of isolated woodlots: Area and habitat relationships. Ecology 68:1724-1734.

Bollinger, E. K., and T. A. Gavin. 1989. The effects of site quality on the breeding-site fidelity in bobolinks. Auk 106:584-594.

Brawn, J. D., and S. K. Robinson. 1996. Source-sink population dynamics may complicate the interpretation of long-term census data. Ecology 77:3-12.

Brown, J. H., and A. Kodric-Brown. 1977. Turnover rates in insular biogeography: Effect of immigration or extinction. Ecology 58:445-449.

Chase, M., N. Nur, and G. R. Geupel. 1997. Survival, population trends and productivity in a Wilson’s Warbler population. Auk 114:354-366.

Conway, C. J., G. V. N. Powell, and J. D. Nichols. 1995. Overwinter survival of Neotropical migratory birds in early-successional and mature tropical forests. Conservation Biology 9:855-864.

Donovan, T. M., R. H. Lamberson, A. Kimber, F. R. Thompson, III, and J. Faaborg. 1995a. Modeling the effects of habitat fragmentation on source and sink demography of Neotropical migrant birds. Conservation Biology 9:1396-1407.

Donovan, T. M., F. R. Thompson, III, J. Faaborg, and J. R. Probst. 1995b. Reproductive success of migratory birds in habitat sources and sinks. Conservation Biology 9:1380-1395.

Faaborg, J., M. Brittingham, T. Donovan, and J. Blake. 1995. Habitat fragmentation in the temperate zone. Pages 357-380 in T. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York.

Fretwell, S. D., and J. H. L. Lucas. 1970. On territorial behavior and other factors influencing habitat distribution in birds. I. Theoretical development. Acta Biotheoretica 19:16-36.

Gates, J. E., and L. W. Gysel. 1978. Avian nest dispersion and fledging success in field-forest ecotones. Ecology 59:871-883.

Gavin, T. A., and E. K. Bollinger. 1988. Reproductive correlates of breeding-site fidelity in bobolinks (Dolichonyx oryzivorus). Ecology 69:96-103.

Greenwood, P. J., and P. H. Harvey. 1982. The natal and breeding dispersal of birds. Annual Reviews in Ecology and Systematics 13:1-21.

Holmes, R. T., and S. K. Robinson. 1981. Tree species preferences of foraging insectivorous birds in a northern hardwood forest. Oecologia 48:31-35.

Holmes, R. T., P. P. Marra, and T. W. Sherry. 1996. Habitat-specific demography of breeding black-throated blue warblers (Dendroica caerulescens): Implications for population dynamics. Journal of Animal Ecology 65:183-195.

Holmes, R. T., T. W. Sherry, P. P. Marra, and K. E. Pitit. 1992. Multiple brooding and productivity of a Neotropical migrant, the black-throated blue warbler (Dendroica caerulescens), in an unfragmented temperate forest. Auk 109:321-333.

Hoover, J. P., M. C. Brittingham, and L. J. Goodrich. 1995. Effect of forest tract size on nesting success of wood thrushes. Auk 112:146-155.

Howe, R. W., G. L. Davis, and V. Mosca. 1991. The demographic influence of "sink" populations. Biological Conservation 57:239-255.

Jackson, W. M., S. Rohwer, and J. V. Nolan. 1989. Within-season breeding dispersal in prairie warblers and other passerines. Condor 91:233-241.

James, F. C., R. F. Johnson, N. O. Warmer, G. J. Niemi, and W. J. Boecklen. 1984. The Grinnellian niche of the Wood Thrush. American Naturalist 124:17-47.

King, D. I., C. R. Griffin, and R. M. DeGraff. 1996. Effects of clearcutting on habitat use and reproductive success of the ovenbird in forested landscapes. Conservation Biology 10:1380-1386.

Martin, T. E. 1992. Breeding productivity considerations: What are the appropriate habitat features for management? Pages 455-493 in J. M. Hagen and D. W. Johnston, editors. Ecology and Conservation of Neotropical Migrant Landbirds. Smithsonian Institute Press, Washington, D.C.

Martin, T. E., J. Clobert, and D. R. Anderson. 1995. Return rates in studies of life history evolution: Are biases large? Journal of Applied Statistics 22:863-875.

Martin, T. E., and D. M. Finch (editors). 1995. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York.

Martin, T. E. and P. Li. 1992. Life history traits of open- vs. cavity-nesting birds. Ecology 73:579-592.

Maurer, B. A., and M. A. Villard, editors. 1996. Special feature: Neotropical migratory birds. Ecology 77:1-68.

May, R. M., and S. K. Robinson. 1985. Population dynamics of avian brood parasitism. American Naturalist 126:475-494.

Mayfield, H. F. 1975. Suggestions for calculating nest success. Wilson Bulletin 87:456-466.

McShea, W. J., M. V. McDonald, E. S. Morton, R. Meier, and J. H. Rappole. 1995. Long-term trends in habitat selection by Kentucky Warblers. Auk 112:375-381.

Morse, S. F. 1996. Source-sink population dynamics of Kentucky Warblers in a complex forested landscape in southern Illinois. Unpublished Master’s thesis, University of Illinois.

Morse, S. F., and S. K. Robinson. In press. Nesting success of a Neotropical migrant in a multiple-use forested landscape. Conservation Biology.

Niven, D. K. 1996. A comparison of the ecology and population dynamics of a Neotropical migrant landbird, the hooded warbler, in winter and summer. Unpublished Ph.D. dissertation, University of Illinois.

Nolan, J. V. 1978. The ecology and behavior of the Prairie Warbler Dendroica discolor. Ornithological Monographs 26:453-482.

Paton, P. W. 1994. The effect of edge on avian nest success: How strong is the evidence? Conservation Biology 8:17-26.

Payne, R. B., and L. L. Payne. 1993. Breeding dispersal in indigo buntings: circumstances and consequences for breeding success and population structure. Condor 95:1-24.

Pease, C. M., and J. A. Grzybowski. 1995. Assessing the consequences of brood parasitism and nest predation on seasonal fecundity in passerine birds. Auk 112:343-363.

Peach, W. J. 1993. Combining mark-recapture data sets for small passerines. Pages 107-122 in J. D. Lebreton and P. M. North, editors. Marked Individuals in the Study of Bird Populations. Birkhauser, Basel.

Perrins, C. M. 1991. Constraints on the demographic parameters of bird populations. Pages 190-206 in C. M. Perrins, J. D. Lebreton, and G. J. M. Hirons (editors). 1991. Bird Population Studies: Relevance to Conservation and Management. Oxford University Press, New York.

Peterjohn, B. G., J. R. Sauer, and C. S. Robbins. 1995. Population trends from the North American Breeding Bird Survey. Pages 3-39 in T. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York.

Petit, L. J., and D. R. Petit. 1996. Factors governing habitat selection by Prothonotary Warblers: Field tests of the Fretwell-Lucas model. Ecological Monographs 66:367-387.

Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652-661.

Rappole, J. H. 1995. The ecology of migrant birds: A Neotropical perspective. Smithsonian Institution Press, Washington, D. C.

Rappole, J. H., and M. V. McDonald. 1994. Cause and effect in population declines of migratory birds. Auk 111:652-660.

Ray, C., M. Gilpin, and A. T. Smith. 1991. The effect of conspecific attraction on metapopulation dynamics. Biological Journal of the Linnaean Society 42:123-134.

Robbins, C. S., D. K. Dawson, and B. A. Dowell. 1989. Habitat area requirements of breeding forest birds of the middle Atlantic states. Wildlife Monographs 103:1-34.

Robbins, C. S., J. W. Fitzpatrick, and P. B. Hamel. 1992. A warbler in trouble: Dendroica cerulea. Pages 549-562 in J. M. Hagan and D. W. Johnston, editors. Ecology and Conservation of Neotropical Migrant Landbirds. Smithsonian Institution Press, Washington, D.C.

Robinson, S. K. 1985. Coloniality in the yellow-rumped cacique as a defense against predation. Auk 102:506-519.

Robinson, S. K. 1992. Population dynamics of breeding Neotropical migrants in a fragmented Illinois landscape. Pages 408-418 in J. M. Hagan and D. W. Johnston, editors. Ecology and Conservation of Neotropical Migrant Landbirds. Smithsonian Institute Press, Washington, D.C.

Robinson, S. K., J. A. Grzybowski, S. I. Rothstein, M. C. Brittingham, and F. R. Thompson, Ill, editors. 1993. Management implications of cowbirds parasitism on Neotropical migratory birds. USDA Forest Service General Technical Report RM-229, Fort Collins, CO.

Robinson, S. K., J. P. Hoover, J. R. Herkert, and R. Jack. In press. Cowbird parasitism in a fragmented landscape: Effects of tract size, habitat and abundances of cowbirds and hosts. In T. Cook, S. K. Robinson, S. I. Rothstein, S. G. Sealy, and J. N. M. Smith, editors. Ecology and Management of Cowbirds. University of Texas Press, Austin.

Robinson, S. K., S. I. Rothstein, M. C. Brittingham, L. J. Petit, and J. A. Grzybowski. 1995a. Ecology and behavior of cowbirds and their impact on host populations. Pages 428-460 in T. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, Oxford.

Robinson, S. K., F. R. Thompson, III, T. M. Donovan, D. R. Whitehead, and J. Faaborg. 1995b. Regional forest fragmentation and the nesting success of birds. Science 267:1987-1990.

Robinson, W. D. and S. K. Robinson. In press. Impacts of selective logging on forest bird populations in a fragmented landscape. Conservation Biology.

Rosenberg, K.V., J.D. Lowe, and A.A. Dhondt. 1999. Forest fragmentation and breeding tanagers:  a continental perspective. Conservation Biology 13:568-583.

Roth, R. R., and R. K. Johnson. 1993. Long-term dynamics of a wood thrush population breeding in a forest fragment. Auk 110:37-48.

Roth, R. R., M. S. Johnson, and T. J. Underwood. 1996. Wood Thrush (Hylocichla mustelina). A. Poole, and F. Gill, editors. The Birds of North America, No. 246. The Academy of Natural Sciences, Philadelphia, PA, and The American Ornithologists Union, Washington, D. C.

Rothstein, S. I. 1975. Evolutionary rates and host defenses against avian brood parasitism. American Naturalist 109:151-176.

Rothstein, S. I., and S. K. Robinson (editors). 1988. Parasitic Birds and their Hosts. Oxford University Press, New York.

Sherry, T. W., and R. T. Holmes. 1992. Population fluctuations in a long-distance Neotropical migrant: demographic evidence for the importance of breeding season events in the American redstart. Pages 421-442 in J. M. Hagan and D. W. Johnston, editors. Ecology and Conservation of Neotropical Migrant Landbirds. Smithsonian Institute Press, Washington, D. C.

Sherry, T. W., and R. T. Holmes. 1995. Summer versus winter limitations of populations: What are the issues and what is the evidence? Pages 85-120 in T. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York.

Smith, A. T., and M. M. Peacock. 1990. Conspecific attraction and the determination of metapopulation colonization rates. Conservation Biology 4:320-323.

Stamps, J. A. 1988. Conspecific attraction and aggregation in territorial species. American Naturalist 131:329-347.

Suarez, A. V., K. S. Pfennig, and S. K. Robinson. 1997. Effects of different edges on the nesting success of a disturbance-dependent bird. Conservation Biology 11:928-935.

Temple, S. A., and J. R. Cary. 1988. Modeling dynamics of habitat-interior bird populations in fragmented landscapes. Conservation Biology 2:340-347.

Thompson, C. F., and J. V. Nolan. 1973. Population biology of the Yellow-breasted Chat (Icteria virens L.) in southern Indiana. Ecological Monographs 43:145-171.

Thompson, F. R., III. 1993. Simulated responses of a forest-interior bird population to forest management options in central hardwood forests in the United States. Conservation Biology 7:325-333.

Thompson, F. R., III. 1994. Temporal and spatial patterns of breeding brown-headed cowbirds in the midwestern United States. Auk 111:979-990.

Tine, C. L. 1996. Mechanisms underlying population dynamics of a migratory songbird in a fragmented forest. Unpublished Ph.D. dissertation, University of Illinois.

Tine, C. L. 1998. Wood thrush population sinks and implications for the scale of regional conservation strategies. Conservation Biology 12: 576-585.

Tine, C. L., W. D. Robinson, and S. K. Robinson. 1998. Consequences of brown-headed cowbird parasitism for host population dynamics. In S. I. Rothstein and S. K. Robinson, editors. Parasitic Birds and their Hosts. Oxford University Press, Oxford, England.

Urban, D. L., and H. H. Shugart. 1986. Avian demography in mosaic landscapes: modeling paradigm and preliminary results. Pages 273-279 in J. Verner, M. L. Morrison, and C. J. Ralph, editors. Wildlife 2000: Modeling Relationships of Terrestrial Vertebrates. University of Wisconsin Press, Madison.

USDA Forest Service. 1986. Landscape and resource management plan: Eastern region.

Van Home, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47:893-901.

Weatherhead, P. J., and M. L. R. Forbes. 1994. Natal philopatry in passerine birds: Genetic or ecological influences? Behavioral Ecology 5:426-433.

Wilcove, D. S. and J. W. Terborgh. 1984. Patterns of population decline in birds. American Birds 38:10-13.

Winker, K., J. H. Rappole, and M. A. Ramos. 1990. Population dynamics of the Wood Thrush (Hylocichla mustelina) on its wintering grounds in southern Veracruz, Mexico. Condor 92:444-460.

¹Illinois Natural History Survey
  607 E. Peabody Drive
  Champaign, IL 61820