Abstract

Abstract The number of black brant Branta bernicla nigricans nests within major breeding colonies on the Yukon-Kuskowkim Delta has declined since 1992. It has been hypothesized that these declines are partially explained by increased numbers of black brant breeding outside of these colonies. To assess this hypothesis, we analyzed spatiotemporal patterns in numbers of black brant nests occurring outside major colonies. Nesting densities of black brant vary among three strata: 1) peripheral to major colonies, 2) other coastal habitats, and 3) inland habitats. We sampled some substrata within each stratum type only between 1986 and 1999 (historical strata), whereas we sampled others over the period 1986–2016 (long-term strata). We fit regression models with number of nests on a plot as a response variable, a log link, and year as the explanatory variable. We allowed each stratum (e.g., historical peripheral) to have its own intercept to account for variation in mean nest density but constrained linear and quadratic regression coefficients to be the same for strata in similar habitats (e.g., historical and long-term peripheral). We used a negative binomial distribution for nests to allow for substantial variation in nests per plot. We fit models using Markov Chain Monte Carlo methods in JAGS. Ninety-five percent credible intervals for both linear and quadratic coefficients for the peripheral and coastal strata, where most nests occurred, broadly overlapped zero, indicating modest trends in numbers of nests in these strata. We estimated there were 6,584 (95% credible interval: 4,221−11,269) dispersed nests in 1986, increasing to 11,051 (95% credible interval: 7,450−17,460) in 2016. Our results indicated that increases in dispersed nests were unable to replace declines in colony nests. Furthermore, quadratic trends indicated that potential earlier annual increases in dispersed nests have declined to zero. We conclude that total numbers of black brant nests on the Yukon-Kuskokwim Delta are likely declining, consistent with the trend in fall age ratios over the same period. Uncertainty about trends in areas not sampled since 1994 adds to the uncertainty about the precise magnitude of the decline. We recommend that the area sampled by the random plot program be expanded to include strata sampled only before 1995.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call