Population dynamics of Greater Scaup breeding on the Yukon-Kuskokwim Delta, Alaska

Populations of greater scaup (Aythya marila) remained relatively stable during a period when populations of lesser scaup (A. affinis) have declined from historic levels. To assist in describing these differences in population trends, from 1991 through 2000, we studied the survival, nesting ecology, and productivity of greater scaup on the Yukon-Kuskokwim Delta (Y-K Delta), Alaska, to develop a model of population dynamics. We located nests, radio-marked females for renesting studies, estimated duckling survival, and leg-banded females to examine nest site fidelity and annual survival.

Greater scaup initiated egg laying later than other species, and most clutches (>80%) were initiated over 20 days each year. We located 1,056 nests; nest success ranged from 7 to 61 % among years. Following loss of their first clutch, 51 % of radio-tagged females attempted to renest. Duckling survival to 30 days of age was 37.5%. Our best model suggested that annual survival did not vary among years and averaged 81 %. Survival rate was positively related to structural body size. Only 8 of 214 banded individuals were reported as recovered (1 each in Maryland, Michigan, Minnesota, Washington, and Alaska and 3 in California).

Using a stochastic model, we estimated that, on average, breeding females produced 0.57 young females/nesting season. We combined this estimate of productivity with our annual estimates of adult survival and an assumed population growth rate of 1.0, then solved for an estimate of first-year survival (0.40). Under these conditions the predicted stable age distribution of breeding females (i.e., the nesting population) was 15.1% 1-year-old, 4.1% 2-year-old first-time breeders, and 80.8% 2-year-old and older, experienced breeders. We subjected this stochastic model to perturbation analyses to examine the relative effects of demographic parameters on k. The relative effects of productivity and adult survival on the population growth rate were 0.26 and 0.72, respectively. Thus, compared to productivity, proportionally equivalent changes in annual survival would have 2.8 times the effect on k. However, when we examined annual variation in predicted population size using standardized regression coefficients, productivity explained twice as much variation as annual survival. Thus, management actions focused on changes in survival or productivity have the ability to influence population size; however, substantially larger changes in productivity are required to influence population trends.