Te Papa bird curator Colin Miskelly describes a recent (pre-COVID lockdown!) attempt to solve a mystery that he has pursued in several remote parts of Fiordland over the past four years.

The grey-backed storm petrel is a small seabird that breeds around the Southern Ocean, including known New Zealand breeding sites in the Chatham Islands and the subantarctic Auckland Islands, Antipodes Island, and Campbell Island. Evidence that grey-backed storm petrels also breed in Fiordland has accumulated since the 1960s, and was summarised in a publication by three Te Papa staff members in 2017.

The article was prompted by sightings of at least three grey-backed storm petrels during a Te Papa expedition to Dusky Sound in late 2016. We also summarised 14 previous records of at least 18 birds from Fiordland. These records included two fledglings with down still attached, indicating local breeding.

Signs of incubation

Developments since 2016 have included five birds seen in Chalky and Preservation Inlets (south of Dusky Sound) during a joint Department of Conservation and Te Papa survey in late 2017. We managed to catch one of these birds, and found that it had a bare brood patch. This naked skin on the belly is only free of down when an adult bird is incubating an egg (it allows efficient transfer of body heat) and is a sure sign that the bird has a nest nearby. I also found a stoat-killed grey-backed storm petrel while walking the Routeburn Track (200 km away) a year later.

The distance between these records is a reminder of how vast Fiordland is – and grey-backed storm petrel nests are notoriously difficult to find. I have only ever seen one nest (in 1983!), despite having visited many known breeding sites in the intervening years.

The 2017 article concluded that “Attempting to locate grey-backed storm petrel breeding colonies within Fiordland would be a daunting challenge”. Or maybe it would just require extraordinary good luck . . .

Mysterious sounds in the night

In February 2019, the Department of Conservation deployed several acoustic monitoring devices on 8,140 ha Secretary Island, at the entrance to Doubtful Sound. The devices were intended to pick up calls of tokoeka (South Island brown kiwi), and were left in place for several months. DOC staff member Maddie van de Wetering was tasked with counting the kiwi calls captured by the acoustic monitoring devices. One of the devices (installed at Secretary Lake, high on the island) had captured a series of calls that Maddie did not recognise – and nor did any of the DOC bird scientists who she shared the sound files with.

Maddie eventually forwarded the sound files to the Birds New Zealand Records Appraisal Committee as an Unusual Bird Record, which is when they came to my attention. The calls were made late at night, in early March. I did not recognise them either, but wondered whether they could be the begging calls of a grey-backed storm petrel chick, made when one of its parents returned to the nest chamber under the cover of darkness. It was the right time of year, the right open habitat (grey-backed storm petrels are not known to nest under forest) and the right time of day/night. Unfortunately, no one knows what a grey-backed storm petrel chick sounds like!

Sound file: Mystery bird call from Secretary Island, March 2019

Although it would be remarkably good luck for an acoustic monitoring device to be placed by chance near a grey-backed storm petrel nest, Secretary Island has several characteristics that make it a plausible site for these tiny, vulnerable seabirds to breed at. It is very high (it is the third highest island in New Zealand, after the South Island and the North Island), meaning that it has extensive tussock habitat above the bush-line. By good fortune, the island has never been colonised by rats or mice (it is the second largest island in New Zealand that has never had rodents, after Adams Island in the Auckland Islands). And stoats have been suppressed to near zero density by intensive trapping since 2005.

Dogged determination

The mystery calls, and knowledge of the island’s conservation history, were sufficient to entice me back to Fiordland. But knowing how difficult grey-backed storm petrel nests are to find, I enlisted the help of conservation dog-handler Jacinda Amey and her DOC-certified seabird dog Poppy. Hopefully if we could get Poppy to a site where the storm petrels were breeding, her sensitive nose would show us where the nests were.

The next challenge was getting to Secretary Island. Fortunately, we were able to join a DOC stoat-trapping trip with boat and helicopter support. DOC staff member Pete McMurtrie manages the stoat control/eradication project, and has been involved since the outset. He was as keen for nesting storm petrels to be found on the island as we were.

It was a drizzly day when we were dropped off on the shore of Secretary Island for the long walk up to Secretary Lake. We had precise instructions on where the acoustic monitoring device had been installed (near the hut we would be based in), but found nothing there.

The next day was fine and clear, allowing us to climb to the top of Mt Grono (1196 m) and search the alpine tussock fields on its slopes. Another two days were spent searching areas of low vegetation around Secretary Lake and a rocky area overlooking Thompson Sound, and we used a powerful searchlight shining into the night sky, in the hope of attracting any storm petrels returning to or departing their nests.

Nothing to see here

Despite this effort, we found no evidence of any petrel species on Secretary Island. No nests, no flying birds, no calls, no corpses, no feathers. It was a spectacular setting to explore, but we failed to find our target species.

The identity of the calls recorded in March 2019 remains a mystery. They may have been of a grey-backed storm petrel, as there are several reasons why a nest at the same site could have failed to produce a chick in 2020. We found several kea feathers at the acoustic monitoring device site when we arrived, and twice saw kea at the site at night. Kea are known to excavate other petrel chicks from their nests and to eat them (Hutton’s shearwaters in the Seaward Kaikoura Mountains). Weka (another potential predator) were common at the site, and the 2019–20 season was a plague year for stoats in Fiordland. Heavy beech seed fall in 2019 lead to an explosion in mouse numbers. This super-abundant food supply meant that stoats born in late 2019 also had very high survival rates, and stoat capture rates on several Fiordland Islands – including Secretary Island – were much higher than normal. Perhaps a predator had destroyed the nest, and any others in the vicinity.

Where to next? We re-installed an acoustic monitoring device at the same site in February 2020, which will be retrieved during the next (post-COVID-lockdown) stoat trapping trip to the island. We wait with interest whether it has picked up more of the mystery calls, or whether the calls are picked up in future seasons.

With thanks to Maddie van de Wetering for her efforts to identify the mystery calls, the Friends of Te Papa for contributing to costs, Pete McMurtrie and the Te Anau Department of Conservation team for logistic support, and of course Jacinda Amey and Poppy for their determined efforts and good company in the field.

Related blogs

• The petrels of Dusky Sound
• Seabird discoveries in remote southern Fiordland
• Fiordland’s Breaksea Sound: 30 years after the rats
• The petrels of Rangatira Island, Chatham Islands
• The call of the wild – attracting seabirds to remote Coal Island