The New Zealand Storm-petrel is not Extinct

Text by Bob Flood, photos by Bryan Thomas

On November 17, 2003, Bryan Thomas and I chartered a small fishing boat (the Assassin) from Sandspit, near Warkworth, just north of Auckland, North Island, New Zealand. Our main purpose was to watch White-faced Storm-petrels Pelagodroma marina (WFSP) close-up and over an extended period in the Hauraki Gulf. The method we employed to attract the storm-petrels replicated the one we use off the Isles of Scilly, UK, to attract British Storm-petrels Hydrobates pelagicus (BSP) and Wilson’s Storm-petrels Oceanites oceanicus (WSP) – drift for two or three hours whilst hanging over the side and just below the surface several onion bags filled with ‘rubby-dubby’ (normally pummelled fish). We left harbour at 07.30 and steamed to two kilometres north of Little Barrier Island, arriving at approximately 09.00 and started drifting with ‘rubby-dubby’ deployed. There was strong sunlight and clear blue skies. The wind was westerly Force 5-6 that, coupled with strong currents in the Hauraki Gulf, resulted in a relatively rough sea-state for a small boat and tricky viewing conditions.

Approximately 20 minutes later a storm-petrel arrived from down-wind, as expected, having smelt the ‘rubby-dubby’. However, it was a medium sized black-and-white storm-petrel, which we did not expect. We thought it was a Black-bellied Storm-petrel Fregetta tropica (BBSP) being generally black-brown with a white rump and belly and appearing big-headed with a very obvious foot projection. We saw dark markings on the belly, but these were difficult to place given our viewing conditions and the flight and feeding behaviour of the storm-petrel (see below). It certainly did not fit White-bellied Storm-petrel Fregetta grallaria (WBSP) given dark on the belly and long foot projection amongst other things. The storm-petrel fed at about 30 metres off the boat whilst continuing on to the oily slick forming up-wind and directly into the sunlight.

Over the following hour-and-a-half more of these black-and-white storm-petrels arrived from down-wind, heading straight for the slick. Brett the skipper chopped-up pilchards and threw them down-wind creating a food source away from the sunlight that encouraged the storm-petrels to feed, albeit for short periods, in improved light for photography. They mainly remained on the slick where 10 were counted at one time (three can be seen together on video footage). We estimate that overall 20 of these black-and-white storm-petrels visited the slick. At the time we thought all were BBSPs, of which we have minimal experience, although we noted that the wing structure did not appear as broad and rounded as expected for BBSP. We did not see WFSPs or any other storm-petrels for comparison.

We then changed location and drifted from 13.00-16.00 about two kilometres off Needles Point at the north end of Great Barrier Island. We saw many WFSPs, but none of the black-and-white storm-petrels. After a fantastic day watching storm-petrels and many other seabirds, we decided to head home and in the end were relieved to get ashore after a lumpy day out and a bumpy steam back to Sandspit.

Later that evening at the cottage we were renting, BT was downloading his digital images onto his laptop and checking them to evaluate the results of his efforts. He called to me in another room asking why the presumed BBSPs were streaked on the belly and lacked a central black stripe. I went to look at the images bearing in mind that field guides warn that the black central stripe is a variable feature and can be absent. However, whilst reviewing the images I was struck by an emerging pattern on all the underside shots of a number of different individuals: (1) breast-band not clear-cut, but had black-brown ‘bleeding’ projections onto a white belly; (2) white flanks and bellies streaked black-brown to varying degrees and, especially obvious on more heavily streaked birds, these formed lines from the edges of the breast to the outer thighs and then on to the undertail coverts, and the central belly was an unmarked white on all of them. This was quite different from anything we expected for BBSP. And the wing structure did not look as we expected. And the feet projection was longer than we expected. And …

On January 25, 2003, a Wrybill Birding Tours pelagic trip from Whitianga on the Coromandel Peninsular, North Island, New Zealand, encountered a storm-petrel, thought to be the NZSP, near to the Mercury Islands just south of Great Barrier Island (see trip report). We recalled Saville et al’s (2003) article in Birding World and as far as we could remember our birds showed the same characteristics as the NZSP discussed therein. However, the authors were bravely presenting the possibility that they could have seen a NZSP that was presumed extinct and known only from three skins collected in the 1800s. Dare we conclude that we had seen at least 10 and probably around 20 of them? Hold on, we thought, we had better get to read the article again and soon.

The next day we joined a pelagic trip organised by Kiwi Wildlife Tours into the Hauraki Gulf. The weather and seas had calmed. We told the organisers that we had seen a good number of BBSPs off Little Barrier Island the previous day and that we were keen to see the article by the Wrybill Birding Tours group on the NZSP (!). The preset itinerary for the day allowed for a relatively brief visit to Little Barrier Island where ‘rubby-dubby’ bags were hung over the sides; but with virtually no wind and calm seas there was little advertisement to storm-petrels of the food source. At various locations during the day, whilst steaming, we did encounter WFSPs, but no black-and-white storm-petrels.

That evening we had dinner at Te Kaura Lodge where Kiwi Wildlife Tours are based and a web version of the article on the NZSP was made available to us. The front-page photograph of the January 25, 2003 storm-petrel looked just like our ones. The photographs of the three museum skins collected in the 1800s reflected the pattern emerging on the underside images of our storm-petrels. There was only one reasonable conclusion. Gulp. It has to be that the New Zealand Storm-petrel is not extinct!


The following description is based on digital images of the NZSPs seen on November 17 (included with this submission) that were taken in strong sunlight, and takes into account field experience from November 17 where relevant. Flight and feeding behaviour also takes into account approximately five minutes of video footage (included with this submission), but naturally only covers Force 5-6 wind conditions.


A medium sized storm-petrel. Big-headed with a thick and short neck. A heavy chest. Wings of medium width, not broad; relatively long and the hand often appearing pointed with P8 and P9 longer than P7 and P10. Tail of medium length, not short; at least as long as the height of the white rump and upper-tail coverts. Legs and feet long; feet projected notably beyond the tail with perhaps two-thirds of the central toes’ length visible. A typical storm-petrel bill with size in expected proportion relative to the head and body.



Belly, flanks and under-tail coverts: White streaked with black-brown; variable between individuals, from light to medium streaked; heaviest on belly sides and tending to form lines on both sides from the edges of the breast to the outer thighs and continuing to the under-tail coverts. Central belly an unmarked white.

Breast: Brown-black with the breast band not clear-cut, but with black-brown ‘bleeding’ projections onto a white belly. Breast band curved indented towards the front end; not with a black apex pointing towards the rear end. This coupled with white belly and undertail coverts resulted in a white egg shaped underside (with streaks on all but the central region). Position of the transition of the breast-belly margin may vary from individual to individual, but always appeared behind the leading edge of the wing.

Head, neck, mantle, back and scapulars: Brown-black with faint scaling on mantle, back and scapulars resulting from narrow paler fringes to the feathers.

Rump and upper-tail coverts: White.


Primaries and secondaries: Brown-black; roughly the same tone as body feathers. Secondaries on the upperwing may have a light reflective quality.

Upper-wing coverts: Overall black-brown. The secondary greater coverts appear to have a silvery cast that is highlighted in certain light conditions. This silvery cast was accentuated on some birds by very pale fringes to secondary greater coverts. On some birds, or at certain angles, these features were clearly visible in the field as pale carpal bars, but the bars were nowhere near as obvious as in WSP.

Under-wing coverts: Thick black-brown leading edge and marginal coverts. Primary greater coverts dusky with pale fringes; primary median coverts dusky with wider pale fringes (possibly variable across individuals). White secondary greater and median coverts and axillaries (there may be light streaking on axillaries of one individual). The overall effect in the field was a broad white central under-wing that fades on the outer-wing.

Tail: Brown-black as remiges.

Moult: There was little or no evidence of wear, abrasion or bleaching suggesting that the birds were of reasonably fresh plumage. However, some images show tips to tail feathers slightly worn. Some images show abraded secondary greater coverts.

Bare parts: Black bill, eyes, legs, feet and webbings.

Flight and feeding behaviour

All the storm-petrels came from down-wind and some were seen at a distance of 100 metres or more approaching the boat with a strong, direct and purposeful low flight. Feeding flight involved steady progress into the wind, keeping low most of the time and hugging contours of the sea surface, which on the day was turbulent resulting in a slight zigzag progression according to changing contours. Wings mainly held parallel to sea surface. To maintain this relative position when sea contours rapidly changed, the storm-petrels would rapidly change body and wing position. Food was taken by dipping and surface seizing. Occasionally birds stalled for short periods over ‘rubby-dubby’ debris, pattering and dipping. Normally the wings were held with trailing edge relatively straight, similar to WSP, and thus the wings were not angular at the carpal joints, for example, as with BSP. However, flight was not as stiff-winged as WSP although gliding flight sometimes resembled that of WSP. Would often glide for long periods whilst contouring and surface dipping. Glides interspersed with periods of rapid wing beats sometimes involving just a couple of beats, other times leading to an extended flight of five seconds or more.


The New Zealand Storm-petrel is not extinct. On November 17, 2003 a group of at least 10 and possibly up to 20 NZSPs was observed, with at least four photographed and three videoed. The birds were in relatively fresh plumage. They were present in an area offering ideal breeding localities and a very rich food source that serves many breeding birds in November including WFSPs. If they were breeding locally, then this population would be comparatively remote from the nearest breeding populations of WSPs, BBSPs and WBSPs. We have published details of our sightings straight-away and have not yet discussed with seabird experts the taxonomic possibilities, which no doubt will follow this article. A distinct taxon is most likely . There is a need to refind the population of NZSPs that we encountered and to establish the months they are in the Hauraki Gulf region and, presumably, around the Mercury Islands and possibly elsewhere in New Zealand waters. There is a pressing need to find out where these birds are breeding and to conserve them.


Saville, S., Stephenson, B., & Southey, I. 2003. A possible sighting of an ‘extinct’ bird – the New Zealand Storm-petrel. Birding World 16: 173-75.

Bob Flood, Isles of Scilly, UK

New Zealand Storm-petrel, c.2km north of Little Barrier Island, North Island, New Zealand, November 2003 (Bryan Thomas).

New Zealand Storm-petrel, c.2km north of Little Barrier Island, North Island, New Zealand, November 2003 (Bryan Thomas).

Bob Flood (centre), with Bryan Thomas to the right (with camera). On the left are Higgo and Paul, with Joe the skipper stood in the cabin. Photo taken of Scilly, UK, by way of Bob Flood.

The above photographs are copyrighted. If you would like to use either, then please contact me first, and I will pass your details on to the photographer – Bryan Thomas.