- The 1989 Exxon Valdez oil spill (EVOS) in Prince William Sound, Alaska provided impetus for a great deal of research into the ecosystems of the Northern Gulf of Alaska. Buried within the multitude of resulting impacts, which included hundreds of thousands of oiled seabirds and dramatic ecosystem shifts in the North Pacific Ocean, was the steady decline of pigeon guillemots (Cepphus columba) at the historically important nesting area in the Naked Island Group of central Prince William Sound. The cause of this decline, however, perplexed researchers for decades, as it both preceded and outlasted the effects attributed to the spill.
Guillemots are a seabird in the auk family (Alcidae) that nest in burrows and other natural crevices. Monitoring of guillemots nesting at the Naked Island Group in Prince William Sound began in 1979, 10 years prior to EVOS, and have continued intermittently until the present. Between 1979 and 2008, the number of guillemots nesting at the Naked Island Group declined by 95%. In 2008, when a remnant population of only about 100 guillemots was still attending the Naked Island Group, it came to light that American mink (Neovison vison), a voracious predator of ground-nesting birds, had been intentionally introduced to the Naked Island Group throughout the 1970’s to provide trapping opportunities.
In 2012, we initiated research to test whether removing mink from the Naked Island Group would allow the local guillemot population to recover. Lethal trapping efforts to remove mink from the islands were initiated in 2014 and continued through 2018. The last mink was captured in the spring of 2016, and the last mink sign (i.e. tracks or scat) was observed in the spring of 2017. By 2018, the Naked Island Group appeared to be free of mink.
Removal of introduced predators is a widespread and effective tool for restoration of island nesting birds. Despite its pervasive use, the effects of predator removal on target species for conservation remains understudied. In addition, no predator removal efforts in North America have targeted American mink. In order to evaluate the effects of mink removal on the guillemot population at the Naked Island Group, I used a Before-After, Control-Impact (BACI) study design. Specifically, I compared trends in population counts of guillemots at the Naked Island Group before and after the removal of mink to the population trends at a number of nearby mink-free islands. To augment this experiment, I measured six variables that have been linked to guillemot nesting success and productivity, due to either predation pressure or forage fish availability. I hypothesized that the abundance of guillemot nests, nestling survival rates, and guillemot use of nest sites vulnerable to land-based predators would all increase following the removal of mink. Also, I predicted that guillemot chick growth rates, diet composition, and meal delivery rates to nests would not be affected by mink removal and would vary instead due to stochastic environmental factors as they affect forage fish availability.
Results from my BACI experimental design indicated that guillemot numbers at the Naked Island Group increased dramatically following the initiation of mink removal efforts, as compared to the control sites. Prior to mink removal, guillemot counts at the Naked Island Group decreased from 146 birds in 2007 to only 69 in 2014. Following mink removal, guillemot counts immediately began to increase, reaching 167 birds by 2018. During this same time period, I observed no clear change in guillemot abundance at control islands where guillemot population counts consistently, though slowly, increased from 2008 to 2018. Concurrent with increases in the number of guillemots at the Naked Island Group following the removal of mink, I also observed increases in guillemot nest abundance, nestling survival rates, and the proportion of guillemot nest in sites vulnerable to mink predation (i.e. burrows instead of crevices in cliffs). These changes in combination suggested that the removal of mink resulted in improved quality of nesting habitat available for guillemots and the initiation of recovery of the guillemot population.
Foraging conditions for guillemots also appeared to have been favorable at the Naked Island Group during my study period, and these would not be expected to be related to mink removal. Guillemot chick growth rates and meal delivery rates were significantly higher than those observed during the period when mink were present at the Naked Island Group, and I found evidence that guillemot diet composition contained a larger percentage of high-quality prey than had been documented in earlier studies. This suggests that foraging conditions guillemots have improved compared to the period when mink predation pressure was at its peak, which coincided with the aftermath of EVOS. As EVOS has been linked to declines in the abundance of high-quality prey, such as schooling forage fish, the improved foraging conditions that I observed were likely due to the recovery of prey resources following EVOS.
All of these factors, both top-down and bottom-up, indicate recently improved conditions for guillemots nesting at the Naked Island Group and in central Prince William Sound. I expected the guillemot nesting population to grow at the Naked Island Group following the removal of an introduced mammalian predator; however, the observed rate of increase in guillemot numbers at the Naked Island Group exceeded my expectations to such an extent that it is unlikely that intrinsic growth (population growth through reproduction) alone was responsible. This suggests that immigration of guillemots from elsewhere in Prince William Sound drove much of the initial recovery in guillemot numbers at the Naked Island Group following the removal of mink.
In conclusion, I demonstrated that the number of pigeon guillemots nesting at the Naked Island Group had been limited by mink predation, and that the removal of mink initiated the recovery of this historically important sub-population. The initial rate of increase in guillemot numbers was likely primarily the result of immigration to the Naked Island Group from other breeding colonies within the Sound. In addition, currently favorable foraging conditions and positive guillemot population trends at nearby control islands suggest that conditions are conducive to the continued recovery of the sub-population of pigeon guillemots at the Naked Island Group, as well as the Prince William Sound population at large.