Northern Prairie Wildlife Research Center
Female mallards prefer seasonal ponds during nesting (Krapu et al. 1997), which concentrates pairs (Johnson and Grier 1988) and nests (Greenwood et al. 1995) where spring pond densities are highest in the species' primary range in the PPR. In areas where nest success rates or renesting efforts also are high, broods will be more numerous and, based on our findings, survive at higher rates when seasonal ponds are plentiful in late spring and summer. Percent of landscape in perennial cover did not affect brood survival. However, because nest success rate varies with percent of landscape in perennial cover (Greenwood et al. 1995), brood density would be expected to vary with percent of perennial cover, if other factors are equal.
Several factors probably contribute to higher survival of mallard broods when seasonal ponds are abundant. Ducklings, when in seasonal ponds, are less vulnerable to aquatic predators that prefer permanent water. Mink (Mustela vison), the most effective predator of ducklings and other neonatal waterbirds (Sargeant et al. 1973, Talent et al. 1983, Eberhardt and Sargeant 1977, Arnold and Fritzell 1990, Korschgen et al. 1996) and the most frequently identified cause of mortality among radiomarked ducklings in our study (G. Krapu, unpublished data), are dependent on wetland-derived prey for survival (Eberhardt 1974). Mink avoid temporary and seasonal ponds and dry basins (Arnold and Fritzell 1990), reproduce poorly during drought (Eberhardt 1974), and rely on permanent water to survive severe drought (Sargeant et al. 1993). As a result, in landscapes lacking permanent water, few mink remain at the end of severe droughts, and mink populations presumably require several years to recover and become a major cause of duckling mortality. Our study was conducted during and immediately following a major drought and brood survival was exceptionally high in the wet years (>90%). It is not known for how many years high brood survival can be maintained under continued wet conditions in areas where mink are present. Current knowledge of habits of mink and mallard broods would suggest that brood survival would remain relatively high until the first summer that broods are confined largely to semipermanent and permanent water. However, further investigation is warranted to determine the extent to which mink might expand into habitats classified as seasonal that take on more permanent water regimes during extended wet periods. Because wet conditions occurred following an extreme drought on our study areas, our model might over predict brood survival when high levels of WETSEAS are present during prolonged wet periods.
Water conditions also affect food availability for mallard broods. Duckling growth and survival varies with macroinvertebrate availability (Cox et al. 1998), which increases (on a landscape scale) as seasonally flooded pond habitat becomes more abundant (Neckles et al. 1990). During the first 2 weeks after hatch, mallard ducklings feed primarily on macroinvertebrates (Chura 1961, Perret 1962). When invertebrates are scarce, ducklings spend more time feeding, move more, and have lower rates of food intake (Hunter et al. 1984, Hill et al. 1987). As a result, starvation and predation probably increase, lowering duckling and brood survival.
Higher mortality of mallard broods during rainy periods may result, in part, because young must be brooded more or risk dying from exposure. Reduced feeding time may result in more rapid depletion of energy reserves, and could lead to duckling hypothermia, starvation, or greater susceptibility to predation. Also, availability of invertebrates at the water surface where young mallard ducklings feed, and in particular, emerging chironomid larvae, a key food (Chura 1961), may be depressed during periods of adverse weather (Nelson 1989). Canvasback ducklings in northwestern Minnesota were particularly susceptible to rain during cold periods (Korschgen et al. 1996). Energy required to maintain homeothermy by ducklings during cold conditions is influenced by ambient temperature, wind velocity, and humidity among other factors (Bartholomew 1982). Rainfall may influence the relative importance of these factors, and the amount of energy required for homeothermy, if ducklings become wet. Further research is needed to understand how environmental factors interact to influence homeothermic costs to ducklings, particularly in natural situations where ducklings are brooded.
Higher survival of early hatched mallard broods probably results, in part, from greater availability of seasonal ponds during late spring and early summer. However, Rotella and Ratti (1992), Dzus and Clark (1998), and this study documented an effect of hatch date after controlling for temporal declines in ponds during the breeding season. The cause of a residual effect of hatch date on brood survival is unknown but might be caused, in part, by growing dependence on local food resources by females to meet energetic needs as fat reserves are depleted (Krapu 1981), leading to these females being less responsive to the needs of young. When water conditions are poor early in the season but much better late, mallard duckling survival is higher late in the season (Dzus and Clark 1998), presumably linked to lower predation rate and better nutrition.
Despite higher survival of early hatched mallard broods, poor success of early nests (Greenwood et al. 1995) results in disproportionately fewer early than late hatched mallard broods in the PPR. As a result, a tradeoff exists between nesting early (low nest success, high brood survival) versus late (high nest success, low brood survival). This tradeoff probably has intensified over the past 60 years as mallard nest success has declined in the PPR (Beauchamp et al. 1996) with increasing habitat fragmentation from agricultural development (Bethke and Nudds 1995, Krapu et al. 1997). Higher survival to fledging of early hatched broods probably explains, in part, why early nesting has remained adaptive in the mallard despite a lower success rate of early laid clutches.