---

---

Abstract: Predation is a natural component of waterfowl population biology, but environmental alterations have changed the magnitude and importance of predation on waterfowl breeding areas. We reviewed the status of waterfowl populations, adaptations of waterfowl that minimize impacts of predation, and the impacts of predation on waterfowl populations in 2 major North American breeding areas, the Arctic and Prairie Regions. We identified the underlying factors contributing to most waterfowl predation problems to be changes in essential breeding habitats and changes in predator community composition and abundance. In the Arctic, high predation rates on waterfowl eggs and young are usually associated with predators gaining access to populations that were previously isolated. In the prairie, predation problems are often related to large-scale habitat degradation coupled with changes in predator communities. Predation problems are often symptomatic of inadequate habitat management, but we recognize that habitat management alone is not always sufficient to effectively manage predation problems. Predation management efforts should be integrated with strategies of long-term management of habitats critical to breeding waterfowl, strategies embraced by the North American Waterfowl Management Plan. Predation management must be tailored to different situations and include an element of flexibility that allows appropriate response to the dynamic nature of factors influencing survival and recruitment.

Key Words: Arctic, breeding waterfowl, Prairie Region, predation

---

---

Introduction

North American waterfowl are a resource of economic, recreational, and aesthetic value. They are appreciated as gamebirds by millions of hunters, and their colorful plumage, elaborate displays, and observability attract even larger numbers of nonconsumptive users. Each year in the United States, more than 22 million people spend millions of dollars on waterfowl-related recreational activities (Teisl and Southwick 1995). State, federal, and private wildlife management agencies have invested numerous resources in programs to enhance waterfowl recruitment. Goals of these programs are to maintain waterfowl populations at sufficient levels to sustain harvest and ensure conservation of nonhunted populations (United States Department of Interior, Environment Canada, and Desarrolo Social México 1994).

Forty-two native species of waterfowl occur commonly in North America, representing 7 taxonomic tribes. Waterfowl present complex management challenges because of their diversity, widespread distribution, and seasonal migration. Among waterfowl species there is wide variation in reproductive potential and mortality rates, as well as predation rates (see Johnson et al. 1992, Sargeant and Raveling 1992). Swans, geese, and sea ducks tend to have lesser reproductive potential (e.g., deferred sexual maturity, smaller clutch size, lesser renesting capacity, variable annual rates of breeding) than other waterfowl groups, but they also have lesser annual mortality rates for eggs and juveniles (Bellrose 1980). In contrast, dabbling ducks and bay (diving) ducks have greater reproductive potential but also greater rates of mortality.

An obvious approach to enhancing populations is to identify key biological factors that affect population size and recruitment and to influence those factors in the most efficient and effective manner (Caughley 1994). For waterfowl, reducing mortality is key to enhancing recruitment rates and populations (Johnson et al. 1992). Generally, we know little about mortality rates of individual species of waterfowl, although we have insight into causes of mortality (see Sargeant and Raveling 1992). For this discussion, we examine problems associated with mortality due to predation. We use the term "problem" but recognize that waterfowl populations are not in peril because of predation. Rather, predation is a problem because it reduces annual production of waterfowl and reduces benefits of habitat conservation activities.

The underlying causes of most predation problems are changes in habitats essential to waterfowl and changes in predator populations. These changes influence rates of predation on waterfowl. Nesting waterfowl suffer from chronic deterioration and loss of critical habitats over much of their breeding range because of encroachment by agricultural, industrial, and residential developments (Sugden and Beyersbergen 1984, Petersen and Hogan 1996). Changes likewise have occurred among predator communities and populations in many waterfowl breeding areas, which have adversely affected survival of waterfowl species (Johnson and Sargeant 1977, Sargeant et al. 1993, Sovada et al. 1995). Decline in abundance or low densities of waterfowl may eventuate a population that is less able to withstand the normal range of predation rates (Raveling 1989). Generally, more than one factor is involved when predation negatively impacts waterfowl populations.

Our purpose is to review the impacts of predation on recruitment and population size of waterfowl in 2 major North American breeding areas, the arctic tundra and coastal region (Arctic Region) and the prairie pothole region (Prairie Region). Although waterfowl breed in numerous other regions of North America, these 2 are especially important (Bellrose 1980). We emphasize species groups in major breeding areas rather than individual species. We focus on predation during the breeding period, because the role predation plays in recruitment and size of populations among waterfowl species is most evident on the breeding grounds (Johnson et al. 1992, Sargeant and Raveling 1992). Finally, we review the status of waterfowl populations, adaptations of waterfowl that minimize impacts of predation, and the underlying factors specific to the breeding areas that influence predation risk for waterfowl. We discuss effectiveness, applicability, and acceptability of predation management options. We discuss management approaches, recommendations, and tradeoffs for obtaining solutions to predation problems where they occur.

Breeding Areas

The Arctic Region is composed of the tundra and river deltas of Alaska and northern Canada. This region provides important breeding areas for most of North America's geese, swans, and sea ducks (Bellrose 1980). This region also is critical to many species of prairie ducks, especially northern pintail (Anas acuta) and American wigeon (A. americana), because it provides stable habitat in years when the prairies are drought-stricken (Derksen and Eldridge 1980, Johnson and Grier 1988, Flint and Grand 1996). Coastal tundra areas are important breeding areas for geese, tundra swans (Cygnus columbianus), and eiders (Somateria spp.), as well as many other ducks. River deltas in the region are especially productive. The Yukon-Kuskokwim Delta, Alaska, is the primary nesting area for cackling Canada geese (Branta canadensis minima), emperor geese (Chen canagicus), black brant (B. bernicla nigricans), greater white-fronted geese (Anser albifrons), and a large portion of the western tundra swan population (King and Derksen 1986). It also is a major breeding area for northern pintails; approximately 15% of the continental breeding population occurs there (Flint and Grand 1996).

The Prairie Region is composed of grassland and aspen parkland in the north-central United States and south-central Canada (Batt et al. 1989). The climate of the Prairie Region is very unstable and characterized by dramatic seasonal and annual fluctuations in precipitation that affect vegetative growth and availability and persistence of wetlands. Periods of drought can last years, the most recent occurred during 1985-1993. The Prairie Region is among the most important areas for breeding ducks in the world. It produces 50-80% of the North American harvested duck species when conditions are favorable, yet it provides only 10% of the breeding habitat on the continent (Batt et al. 1989). Twelve species of ducks commonly breed in the Prairie Region. Surveys during breeding have indicated that the majority of the mean estimated populations (1955-1985) of 8 species were found in the Prairie Region (Batt et al. 1989).

Status of Population

Most populations of swans and geese in North America have increased or remained stable in recent decades, although many populations experienced severe fluctuations in the past century (emperor geese; cackling, dusky [B. c. occidentalis], and Aleutian [B. c. leucopareia] Canada geese; and brant; Raveling 1984; King and Derksen 1986; Pamplin 1986; Hodges et al. 1996). Less is understood about the status of sea duck populations, but several populations have reached critically low levels (e.g., spectacled [Somateria fisheri] and Steller's [S. stelleri] eiders; see Petersen and Hogan 1996, Hodges et al. 1996). Predation has not been identified as a serious threat overall to populations of swans, geese, and sea ducks. Nevertheless, local populations have occasionally suffered high rates of predation, and the importance of predation in population declines should not be minimized (King and Derksen 1986, Pamplin 1986, Stehn et al. 1993, Sedinger et al. 1994, O'Briain et al. 1998).

Since 1955, duck breeding populations surveyed in the Prairie Region have varied from 34 million in 1956 to 12 million in 1991, when populations of some species were at or near record low levels (Batt et al. 1989, Austin 1998). Since then, populations of most species common in the Prairie Region, except northern pintails and lesser scaup (Aythya affinis), have recovered to levels near or above the population goals of the North American Waterfowl Management Plan (United States Fish and Wildlife Service 1998). Greenwood and Sovada (1996) discussed factors (e.g., drought cycle, habitat conditions, predator populations) contributing to the recovery of duck populations in the 1990s. Population trends among breeding ducks tend to reflect availability of ponds in spring in the Prairie Region (Batt et al. 1989, Austin 1998), but Johnson and Shaffer (1987) presented evidence of long-term population declines in mallards preceding the drought of the 1980s. Bethke and Nudds (1995) indicated that declines in mallards and northern pintails during the recent drought were more severe than predicted on the basis of variation in available wetlands. Furthermore, high predation rates may inhibit the recovery of duck populations when habitat conditions are favorable. Population declines of several species have been linked with low recruitment, primarily attributed to high rates of predation on eggs, but predation on adults and ducklings also have contributed (Cowardin et al. 1985, Klett et al. 1988, Sargeant and Raveling 1992, Greenwood et al. 1995).

Waterfowl Adaptations to Minimize Predation Risk

Species breeding in the Arctic Region have evolved with few mammalian and avian predators; nevertheless, evolution has favored reproductive strategies that minimize risk of predation. Arctic foxes (Alopex lagopus) are the principal predators of waterfowl in the Arctic Region; eggs and young are especially vulnerable to depredation by arctic foxes (Pamplin 1986, Sargeant and Raveling 1992, Ely et al. 1994). Large gulls (Larus spp.) and jaegers (Stercorarius spp.) also have been identified as important avian predators of waterfowl eggs and young (Mickelson 1975, Grand and Flint 1997, Tremblay et al. 1997, Schmutz and Hobson 1998). Other occasional predators include bears (Ursus spp.), gray wolves (Canis lupus), coyotes (C. latrans), red foxes (Vulpes vulpes), caribou (Rangifer tarandus), ravens (Corvus corax), eagles (Aquila spp.), and snowy owls (Nyctea scandiaca; Abraham et al. 1977, Sargeant and Raveling 1992).

Predation on adult swans is probably rare; published accounts are infrequent and describe only isolated killing of individual birds (see Sargeant and Raveling 1992). Similarly, adult geese of most species do not commonly fall prey to predators (Sargeant and Raveling 1992), but predation on them could be additive to mortality due to other factors (e.g., subsistence harvest, environmental contamination; King and Derksen 1986, Raveling 1989, Sargeant and Raveling 1992). Modeling suggests that in long-lived species, adult survival may be key to maintaining healthy populations (Brault et al. 1994, Goudie et al. 1994, Schmutz et al. 1997).

Species such as the cackling Canada goose (Mickelson 1975), tundra swan (Monda et al. 1994), and common eider (S. mollissima; Schamel 1977, Robertson 1995, Cornish and Dickson 1997) nest on islands, islets, or hummocks surrounded by water that improve visibility for incubating birds and limit access by mammalian predators, especially foxes. Furthermore, nest initiations on islands may be delayed until they are ice-free, inhibiting mammalian access during nesting (Schamel 1978, O'Briain et al. 1998, Spaans et al. 1998). Inland nest sites are accessible to most predators, but often these nests are dispersed and well concealed, reducing the chance of egg depredation. Petersen (1990) found taller vegetation near successful emperor goose nests than near unsuccessful ones. Nest sites are often near water and foraging areas, which probably increases survival of young and adults. Adults of most arctic nesting waterfowl are very attentive to their nests (Thompson and Raveling 1987), especially after egg laying, and eggs are concealed with down and nest material when the parents leave the nest. Among swans and geese, males and females actively defend nests and care for young (O'Briain et al. 1998). Eiders actively defend nests from avian predators (Schamel 1977, Robertson 1995). Species nesting in dense colonies may benefit from deterrent effects of nest defense by colony members and communal brood rearing (Findlay and Cooke 1982, King and Derksen 1986). Also, smaller species of geese and some ducks often nest in association with raptors or jaegers and among bird colonies such as larger geese, gulls, or terns that offer protection from predators (Schamel 1977, Robertson 1995, Tremblay et al. 1997, O'Briain et al. 1998). Among individual species breeding in the Arctic Region there is notable synchrony in nest initiation and hatching, thus the period when eggs and young are most available is remarkably short (Ryder 1967, Monda et al. 1994, Cooke et al. 1995, Grand and Flint 1997). This tremendous increase in prey abundance likely overwhelms predator species.

Breeding waterfowl in the Prairie Region, unlike the Arctic Region, are vulnerable to many predators. Twenty avian and mammalian predator species that affect waterfowl have been identified in the Prairie Region (Sargeant et al. 1993). Numerous studies have demonstrated high predation rates on adults, ducklings, and eggs (see Sargeant and Raveling 1992). Depredation of eggs is thought to be the primary factor influencing recruitment; however, impacts of depredation on hens and ducklings also are important (Johnson and Sargeant 1977, Johnson et al. 1992, Greenwood et al. 1995 ).

Natural selection among nesting ducks in the Prairie Region has favored reproductive strategies that reduce the adverse effect of predation. These include breeding as yearlings, large clutch sizes, and ability to renest if the first nest is destroyed (Bellrose 1980). Females of all species have cryptic plumage, and nests are well concealed. Nest sites in uplands are commonly located in dense vegetation. Some species tend to select nest sites over water or on islands, which provide additional security from certain mammalian predators (Sargeant et al. 1984). Some species have developed antipredator behaviors, such as injury displays to divert predators from nest sites or feigning death (Sargeant and Eberhardt 1975).

Change in Habitats and Predator Communities and the Impacts on Waterfowl Populations

Substantive depredations of waterfowl eggs and young in the Arctic Region are usually associated with predators gaining access to isolated populations. Notable is the near extinction of the Aleutian Canada goose following introduction of arctic foxes on its nesting islands (Bailey 1993). Waterfowl nesting colonies on islands also suffer severe losses when arctic foxes cross ice bridges or mudflats that periodically occur during nesting seasons (Ryder 1967, Quinlan and Lehnhausen 1982). Decrease in availability of secure nest sites following destruction of islands by storm surges and ice scouring contributed to the decline in the spectacled eider population on the Yukon-Kuskokwim Delta (Dau, unpublished, in Ely et al. 1994). A major earthquake in Alaska in 1964 uplifted the outer portion of the Copper River Delta, draining many waterways and substantially altering nesting habitat (Crow 1971). Subsequently, brown bears (Ursus arctos) and coyotes invaded the Delta and were responsible for the dramatic decline in nest success of dusky Canada geese. In some years following the earthquake, up to 80% of nests were depredated (Cornely et al. 1985, Campbell 1990).

Arctic foxes are particularly effective predators that remove and cache many eggs (Quinlan and Lehnhausen 1982, Fay and Stephenson 1989, Stickney 1991, Bantle and Alisauskas 1998). Cached eggs provide food for foxes during periods of low prey availability and contribute to fox survival and reproductive success (Fay and Stephenson 1989, Bantle and Alisauskas 1998). Several studies of arctic-nesting waterfowl have described a negative association between the magnitude of nest depredation by arctic foxes and the abundance of alternative prey species such as brown lemmings (Lemmus sibiricus), collared lemmings (Dicrostonyx torquatus), and tundra voles (Microtus oecorcomus, Pehrsson 1986, Anthony et al. 1991, Tremblay et al. 1997, O'Briain et al. 1998). Bantle and Alisauskas (1998) observed a similar diet shift among foxes denning in a snow goose colony. Nevertheless, Stickney (1991) reported waterfowl eggs to be the primary food of foxes in a brant colony, despite availability of microtines. The greatest impact on waterfowl populations often occurs when arctic fox densities are high and densities of nesting waterfowl are low. Raveling (1989) observed that the predation rates on brant nests in small colonies were greater than in large colonies. Predation is believed to have precipitated severe declines in local brant populations already diminished by excessive harvest and habitat degradation (Raveling 1989, Ward et al. 1993).

In the Prairie Region, breeding habitat has been impacted severely by conversion of grassland to cropland and wetland drainage (Sugden and Beyersbergen 1984, Miller and Nudds 1996), increasing the impact of predation on waterfowl populations (Cowardin et al. 1985, Greenwood et al. 1995, Beauchamp et al. 1996). Loss of wetlands and fragmentation of grasslands have concentrated ducks and predators in the remaining patches of suitable habitat and likely have reduced the abundance of alternative prey species (Cowardin et al. 1983, Sargeant and Raveling 1992). Because prairie ducks occupy breeding habitat based on availability of wetlands in spring, they are at great risk to predation in areas where cropland is abundant because they are forced to nest in the scant patches of nesting cover that remain. Greenwood et al. (1995) reported that nest success in the Prairie Region of Canada decreased about 4 percentage points for every increase of 10 percentage points in availability of cropland.

Habitat changes and human activities also appear to have affected the distribution and abundance of some predators in the Prairie Region (Sargeant et al. 1993, Greenwood and Sovada 1998). The replacement of larger predators (e.g., gray wolf, coyote) by smaller species, which often are more numerous, has been detrimental to nesting ducks (Johnson and Sargeant 1977, Sargeant et al. 1984, Sargeant et al. 1993). Particularly significant has been an increase in numbers of red foxes, a principal predator of nesting ducks and eggs (Johnson et al. 1989, Sovada et al. 1995). Sovada et al. (1995) demonstrated that nest success was greater in areas occupied by coyotes than in areas with foxes. Raccoons (Procyon lotor) and Franklin's ground squirrels (Spermophilus franklinii) appear to have increased numerically and expanded their range in response to changes in habitats following human settlement in the region (Sargeant et al. 1993, Greenwood and Sovada 1998). Increases in woody habitats and protection from human-caused mortality also may have contributed to changes in the avian predator community (Murphy 1993, Sargeant et al. 1993). Red-tailed hawks (Buteo jamaicensis), great-horned owls (Bubo virginianus), crows (Corvus brachyrhunchos), and magpies (Pica pica) have expanded their ranges in the Prairie Region (Houston and Bechard 1983, Murphy 1993, Sargeant et al. 1993). These changes in predator populations likely have contributed to increased rates of predation on breeding waterfowl in the Prairie Region.

Predation has a major impact on populations of ducks in the Prairie Region. Johnson and Sargeant (1977) demonstrated that predation on nesting mallard hens by red foxes may be responsible for the imbalanced sex ratio common in that species. Murphy (1993) and Greenwood et al. (1995) reported a high proportion of females among dabbling duck remains killed by several mammalian and avian predator species in the Prairie Region. Analyses of nesting studies indicate long-term nest success of ducks in the Prairie Region has declined (Beauchamp et al. 1996), likely due in part to loss and fragmentation of critical breeding habitats and changes in predator communities. Reported nest success rates of ducks (see reviews: Sargeant and Raveling 1992, Beauchamp et al. 1996) are often below the levels (15-20%) thought to be necessary for population stability of the common species (Cowardin et al. 1985, Klett et al. 1988). Most studies report that > 70% of nest failures are attributed to predation.

Predation Management

Marked differences between arctic and prairie ecosystems affect the potential utility of strategies to manage predation on nesting waterfowl. The Arctic Region, nearly uninhabited by humans, with vast undisturbed areas of mainland and island nesting habitats, is characterized by few predators of significance to waterfowl and relatively stable habitat quality and availability. Waterfowl, particularly geese, may be concentrated in mainland and island colonies, but some are dispersed throughout the open mainland landscape. The prairie, in contrast, is inhabited throughout by humans and greatly disturbed by agriculture. It is characterized by numerous predators of significance to waterfowl and very unstable habitat quality and availability. Ducks predominate among waterfowl in the Prairie Region and are dispersed widely. In localities where cropland dominates the landscape, nesting ducks often are concentrated in scattered patches of grassland that remain.

Arctic region

Available information indicates that local breeding populations in mainland and island colonies tend to be more affected by predation than populations characterized by dispersed nest sites, although few studies have documented effects of predation in landscapes with more isolated nesting sites. Nesting colonies, whether on islands or on mainland, have discrete distributions that logistically facilitate predation management. Arctic foxes, the primary predator of nesting birds, tend to occupy exclusive areas during the waterfowl nesting season (Anthony 1997), which also facilitates potential management. Elimination of foxes from some Aleutian Islands brought about the spectacular recovery of Aleutian Canada geese (Bailey 1993). Elimination of foxes from high-density mainland nesting colonies also could benefit production (Pamplin 1986, Anthony et al. 1991). Removal of arctic foxes from a brant colony on the Yukon-Kuskokwim Delta improved nest success (Anthony et al. 1991); however, fox removal in a more inland area failed to benefit the nesting success of other goose species (M. Anthony, unpublished data). Grand and Flint (1997) likewise found increased nest success in mainland colonies of spectacled eiders when mew gull (Larus canus) populations were reduced.

Although nesting habitats in the arctic are relatively undisturbed, it may be important to monitor and limit encroachment by human activities, such as oil exploration, and to maintain desirable waterfowl breeding populations (Walker et al. 1987). Human disturbance in nesting colonies is likely to increase predation of eggs and young (see King and Derksen 1986, Pamplin 1986). Furthermore, garbage and man-made structures associated with activities such as petroleum development may provide winter food and shelter sufficient to sustain increased arctic fox populations, ultimately affecting waterfowl breeding success (Sargeant and Raveling 1992, Stickney and Ritchie 1996). Quinlan and Lehnhausen (1982) warned against constructing causeways that connect the mainland to barrier islands because these could provide access to barrier islands by arctic foxes, resulting in greater predation risk for island-nesting waterfowl.

Recommendations

The relative stability, undisturbed nature, and remoteness of arctic ecosystems will likely focus future management efforts on island and mainland colonial nesting species that are most vulnerable to predation. Anthony et al. (1991) recommended that arctic fox densities and abundance of alternative prey be monitored annually to provide guidance in management decisions regarding protection of brant colonies. Raveling (1989) recommended removal of arctic foxes associated with small colonies of brant to reestablish or expand diminished colonies. Because few in-depth studies of predation in arctic and subarctic regions have been conducted, there are opportunities for research on predator populations and predator-prey relationships in nesting colonies and in habitats that support dispersed waterfowl nesting.

Prairie region

Numerous methods have been tried or suggested to reduce the effects of predation on nesting ducks in the prairie (see reviews by Clark et al. 1996, Greenwood and Sovada 1996). Management approaches fall into 3 general categories: 1) restoration and protection of habitats critical to nesting ducks, 2) isolation of nests from predators, and 3) removal of predators by lethal methods.

Predation is most severe where a large proportion of the landscape has been converted to cropland. Greenwood et a1. (1995) found that nest success in the Prairie Region is correlated positively with amount of perennial grassland in the landscape. Because of this, the first priority in managing predation should focus on protecting and restoring grassland. Areas where large tracts of grassland remain intact should be protected from conversion to cropland. Further, Reynolds et al. (1996) suggested that areas with greatest potential for management to improve success of nesting ducks are those where wetlands are most abundant.

Intensive management designed to keep predators away from nests can be grouped into methods that physically separate or conceal prey from predators, those that alter foraging behavior of predators or their food availability, and those that affect predator distribution or abundance (Greenwood and Sovada 1996). Most methods are aimed at protecting the nesting hen and eggs from mammalian predators; few focus on avian predators or enhancing survival of ducklings. Methods appropriate for the local scale include isolating mechanisms (nesting structures, barrier fences, and moats) and mechanisms that alter predator foraging (conditioned taste aversion, supplemental foods). The former are effective in increasing nest success, although they can be resource- and labor-intensive (Lokemoen 1984). The latter have not proven to be effective (Greenwood et al. 1998, also see Greenwood and Sovada 1996). Methods that influence predator distribution, such as managing canid populations to favor coyotes and reduce red fox populations, have potential to positively affect nest success over large areas of landscape (Sovada et al. 1995). Coyotes also may suppress raccoon densities (Johnson et a1. 1989). Badgers (Taxidea taxus), a species associated with large areas of grassland, may suppress striped skunk (Mephitis mephitis) populations (Johnson et al. 1989). Numerical reduction of predators with reproductive inhibitors also has been suggested, but this approach is not likely to be acceptable because of cost and difficulty in accomplishing goals (see Greenwood and Sovada 1996).

Several studies have shown lethal methods (e.g., toxicants, traps, shooting) to be effective in controlling predators and increasing duck nest success (Duebbert and Lokemoen 1980, Greenwood 1986, Sargeant et al. 1995, Garrettson et al. 1996). Use of toxicants to reduce predator numbers could be especially effective over large areas of the landscape, but effects on nontarget species are of great concern. Although predator control is effective, there are several drawbacks. Predator control often is opposed by the public, even to protect endangered species, and will be difficult to implement because of social constraints (Goodrich and Buskirk 1995). Toxicants, the most effective control method, require an extraordinarily expensive process to register compounds, deterring the development of new control agents. Other issues to consider include humane treatment of animals, ethical concerns, and anti-hunting and anti-trapping sentiment (Kirkpatrick and Turner 1985). To enhance duck production, predators would have to be controlled each spring at a time when fur has little value. Thus, there are serious ethical concerns for wildlife managers who would be forced to advocate wasting a valuable fur resource to enhance production of ducks to be harvested by other users. Information is not available regarding the long-term impact of predator control on passerines and other species; however, Dion et al. (1999) reported that in the short term passerine nest success is not affected by predator removal. Finally, there are no long-term benefits because redistribution, dispersal, and compensatory reproduction by predators require extensive control of predators every year.

Recommendations.

In the unstable and very disturbed environment of the Prairie Region, where wetland conditions change continually, the challenge is to know when, where, and how to apply existing options to manage nesting habitat and predation. To make informed decisions, managers need to know habitat conditions, predator community composition, and nest success rates at the landscape scale. Maintaining waterfowl breeding habitat is paramount to the long-term welfare of waterfowl populations. Maintaining and increasing critical breeding habitat can be accomplished through continued purchase or lease and then restoring or improving good breeding habitat, encouragement of beneficial public policy, and working with landowners to encourage beneficial farming practices (see United States Department of Interior, Environment Canada, and Desarrolo Social México 1994). All of these actions improve characteristics of the landscape for ducks and other wildlife species and likewise improve soil and water conservation.

Management to enhance wetland and upland habitat for nesting ducks is likely to be most effective when applied in areas where nest success is greatest (Greenwood et al. 1995), and this occurs generally near large tracts of grassland. In areas where nesting habitat is scant and predation rates are high, it makes little sense to attempt to attract more ducks through wetland restoration or to increase nesting habitat through purchases or leases of small tracts of land. Some landscapes are so severely altered that it would be impossible to restore enough nesting habitat to return nest success to sustainable levels. Under these conditions, it may be more appropriate to isolate areas of nesting habitat from predators or reduce predator populations through specially permitted lethal control.

Knowledge of how predators respond to managed habitats (e.g., configuration of grassland patches, wetland-grassland composition in landscapes) could direct management to create more secure habitats. It may be possible to "compose" favorable grassland-wetland patterns on a landscape scale in collaboration with farm programs. Knowledge of how predator species interact as a community could provide insight to predation management options thus far not identified.

Conclusions

The vision of the North American Waterfowl Management Plan (NAWMP) is focused clearly on conservation of biological diversity and a landscape approach to management, recognizing that the perpetuation of waterfowl populations depends on long-term protection, restoration, and management of critical habitats (United States Department of Interior, Environment Canada, and Desarrolo Social México 1994). The NAWMP is intended to provide benefits to the overall environmental fitness of affected ecosystems largely through habitat conservation. Within the NAWMP, predation issues are addressed as symptoms of inadequate habitat management, yet it is recognized that habitat improvements alone are insufficient to effectively manage predation in all situations. Intensive predator management is a warranted and important alternative when habitat management alone is inadequate.

Clearly, intensive predator management, especially extensive predator removal, can successfully enhance waterfowl productivity. However, in an era of limited resources, expending funds on intensive predator management necessarily competes with habitat acquisition, management, and enhancement activities, the emphases of the NAWMP. Such economic trade-offs must be considered and incorporated if predation management is to be applied at more than a site-specific scale. Long-term solutions transcend site-specific intensive predation management, yet we cannot dismiss the dilemma facing individuals responsible for managing local breeding populations of waterfowl with limited resources. To ensure biologically sound management decisions at site-specific and landscape scales, guidance, information, and acceptable tools must be available for managers to develop suitable strategies to abate predation problems. Strategies must be tailored to each situation, but ultimately integrated with long-term conservation of waterfowl. This means that predation management, even at a local scale is usually not a substitute for proper habitat management.

The NAWMP provides a framework to actively manage predation (United States Department of Interior, Environment Canada, and Desarrolo Social México (1994), but evaluation is an essential component needed to assess effectiveness of predation management. If the management goal is to enhance waterfowl production, evaluation should focus on changes in recruitment. Other measurements (number of predators removed, amount of nesting habitat planted, number of wetlands restored) are easy and appealing to document, but may have little relation to actual number of offspring recruited (Sargeant et al. 1995).

Although predation is a natural component of water fowl population biology, environmental and anthropogenic alterations have changed the magnitude and importance of its impact on waterfowl populations. Predation issues are controversial and complex because they raise questions involving humane treatment of animals, anti-hunting and anti-trapping sentiments, and ethical concerns about management of one harvested species in favor of another. The effects of altering predator communities must be examined if intensive predator management programs are to be defended adequately. Innovative methods of population management need to include an element of flexibility to allow decision making to respond to the dynamic nature of factors influencing breeding success. Ultimately, predation management must not only solve waterfowl recruitment shortfalls, but also be accepted by the public, whose support is critical to the maintenance of waterfowl populations, and contribute to long-term management goals.

---

Acknowledgments

  J. E. Austin, D. V. Derksen, R. J. Greenwood, R. E. Kirby, C. D. Mitchell, and T. L. Shaffer provided thoughtful comments on early drafts of this manuscript.

---

Literature Cited

NPWRC Home | Site Map | About Us | Staff | Search | Contact | Web Help | Copyright