Confirmed mouse attacks on Gough and Marion Island seabirds

Peter Ryan tells us about the discovery of this behaviour on these seabird islands.

Grey-headed Albatross mice injuries. Photo by Ben Dilley

Grey-headed Albatross juveniles with mice injuries. Photo by Ben Dilley

“We’ve suspected that House Mice are significant predators of seabird chicks on Gough Island since 2001, when Richard Cuthbert and Erica Sommer were the first ornithologists to spend a year on the island. They witnessed the sudden disappearance of many large, seemingly healthy Tristan Albatross chicks, and suggested that mice were the likely culprits (see Cuthbert and Hilton 2004, Biol. Conserv.). Mice were confirmed to be responsible in 2004, when Ross Wanless filmed mice attacking albatross chicks. Combined with their mortality on long-lines at sea, this resulted in the Tristan Albatross being up-listed to Critically Endangered (see Wanless et al. 2009 Biol. Conserv.).

Ross also confirmed attacks on Atlantic Petrel chicks, another species almost entirely restricted to Gough Island (Wanless et al. 2012, Anim. Conserv.), and it is likely that mouse predation also is responsible for Gough Buntings being increasingly confined to the island’s coastal cliffs and alpine interior (Ryan and Cuthbert 2008, Bull. Br. Orn. Club). This kick-started a research programme to assess the feasibiity of eradicating mice from Gough Island, which including learning more about mouse biology (e.g. to determine their breeding season and seasonal population changes, Cuthbert et al. in press, J. Mammal.) and their susceptibility to poison bait (e.g. do all mice in caves forage on the surface and thus be exposed to an aerial bait drop? Cuthbert et al. 2011, Wildl. Res.). There was relatively little focus on how mice attack birds until 2013/14, when Ben Dilley and Delia Davies used a range of cameras to document the nature of mouse attacks. This was a particularly bad year for seabirds, with fewer than 10% of Tristan Albatross chicks surviving, and they were able to show that an attack by a single mouse was capable of killing a large albatross chick within a couple of nights (Davies et al. 2015, Avian Consev. Ecol.). They also showed that mice kill chicks from almost all petrel species, with smaller birds and winter-breeding species being particularly susceptible to predation (Dilley et al. 2015, Antarct. Sci.). All Atlantic Petrel chicks in filmed burrows were killed within a day of hatching, despite the presence of a parent in the burrow.

Mouse Attack. Photo by Stefan Schoombie

Mouse Attack. Photo by Stefan Schoombie

Mouse Attack. Photo by StefanSchoombie

Mouse Attack. Photo by StefanSchoombie

When mouse predation of seabird chicks was first discovered on Gough Island it was assumed that the large size of Gough mice contributed to their nasty behaviour – at up to 50 g in extreme cases, they average 50% larger than other populations of House Mice. However, Wanless et al. (2007, Biol. Letters) suggested that the fact that mice were the only introduced mammal was the main reason for the attacks. At South Africa’s Marion Island, House Mice have been the sole introduced mammal since cats were eradicated from the island in 1991. Despite these mice being of typical size, we’ve had occasional attacks on Wandering Albatross chicks on Marion Island since 2002 and in 2009 we found a few pre-fledging Sooty Albatrosses with head wounds that we could only attribute to mice (Jones and Ryan 2010, Antarct. Sci.). In 2015, the habit of attacking large mollymawk chicks suddenly exploded across the island, with up to 10% of Grey-headed, Sooty and Light-mantled Albatross fledglings attacked (Dillet et al. in press, Antarct. Sci.). The Marion mice have learned that pre-fledgling albatross chicks are vulnerable on their heads, where the contour feathers are shorter than on the body. Interestingly, although we also had a few attacks on Wandering Albatross chick heads, the rate of attacks on Wanderer chicks didn’t increase in 2015. Winter 2015 also saw the first confirmed kills of petrel chicks on Marion, although we have suspected for some time that mice are responsible for the limited recovery in burrowing petrel populations on Marion since the cats were eradicated (Cerfonteyn and Ryan 2015, Antarct. Sci.).

The good news is that it is likely that mice can be eradicated from both Gough and Marion Islands. Gough is widely regarded as the top global priority island for rodent eradication. Plans for an eradication are moving forward, with fund-raising currently underway to allow an eradication attempt in winter 2019. Until 2015, Marion hasn’t featured on the global priority list, but the widespread attacks this year have increased the urgency to tackle mice on this island. A feasibility study conducted in April-May 2015 concluded that Marion was a good target for an eradication attempt; lobbying is underway to put this on the South African Department of Environmental Affairs’ conservation agenda.

The message from Gough and Marion is that mice can be significant predators of seabird chicks and eggs when they are the sole introduced mammal on an island – as is the case on the Antipodes. Without dedicated observation effort, such predation can be easily overlooked, especially among burrow-nesting species, and the absence of direct evidence cannot be assumed to mean that there is no impact on seabirds. Given their widespread impacts on other biota (especially invertebrates), mice should be eradicated from islands wherever possible.” Read more

Click to see video

Mice as sole introduced mammalian pest species

DOC help out on Gough Island