Nick Lethaby (4/18/2006)
One issue I would like to raise is the reporting of Swinhoe's Storm-Petrels (Oceanodroma monorhis) off the Pacific coast. While the species does doubtlessly occur (though how frequently is unclear), my own opinion, based on the birds that I have seen, is that Tristram's (O. tristrami) is much the commoner species of dark storm-petrel on the Pacific side of Japan, off Chiba and Ibaraki prefectures, down by the Izu Islands and further south. I have never seen a bird I felt confident was a Swinhoe's, although I have seen reports by experienced observers of this species, at least some of which are presumably correct.
If one considers the main breeding range (Sea of Japan and Yellow Sea), migration and wintering areas (via Singapore to winter in the Indian Ocean) of Swinhoe's Storm-Petrel, there is not much reason for it to occur commonly off the Pacific coast of Japan.
With storm-petrels, I think it is easy to fall into the trap of guessing at the species one sees and recording what is supposed to occur. For example, on my first summer ferry trip to Hokkaido, I put down all the dark storm-petrels as Swinhoe's. However, on the return trip, I was able to compare these birds with a few clear-cut Leach's Storm-Petrels (O. leucorhoa) and it was obvious that they were much bigger. In addition, in scoping some, I could see the slightly paler grey rump of Tristram's.
The same comments apply to white-rumped storm-petrels. During that trip, I noted all down as Band-rumped (Madeiran, O. castro), although many looked to me just as likely to be Wilson's. Although I hate to leave many birds unidentified, I think the honest approach is leave many storm-petrels so. I would be interested to know if anyone has been off Chiba-ken or Ibaraki-ken in a smaller boat and been able to get good photos or take specimens of the storm-petrels there.
Sean Minns (4/19/2006)
On my one trip back from Hachijojima in April 2003, I encountered Tristram's and Swinhoe's Storm-Petrels, along with Bulwer's Petrel (Bulweria bulwerii) and small numbers of Wedge-tailed Shearwaters. I believe that Wedge-tailed and Bulwer's are present off Hachijojima in small numbers. The following are my thoughts on the ID issues of the four dark-rumped petrels that might occur. I should mention that dark Leach's is also a possibility.
Bulwer's are sufficiently larger and broader-winged, and not to be confused with Tristram's or Matsudaira's (O. matsudairae). Although the patterning on the mantle and wings does superficially resemble Swinhoe's, their size and jizz should separate them. In my limited experience, Bulwer's is stiffer-winged, more like a shearwater, while Swinhoe's is more fluttering in flight, with glides in between, like British Storm Petrel (Hydrobates pelagicus).
I have not seen Matsudaira's, as they are more southerly in distribution, so cannot really comment; but Tristram's are much longer-winged than Swinhoe's and appear to glide more, similar to Bulwer's, while having comparatively longer, slimmer wings for their body size. I think the main confusion occurs between Tristram's and Matsudaira's, and not between Tristram's and Swinhoe's, if seen well.
Views from the Hachijojima ferry are usually better than from the larger Honshu-Hokkaido ferries. Many of the darker-rumped petrels I saw near Hachijojima remained unidentified, but several I saw close enough to be certain that they were Tristram's, as they were larger, longer-winged and grey-rumped. They could be differentiated from Matsudaira's because of their forked tail and lack of white in the wing, as well as by the features mentioned previously.
I also saw two much smaller, all-dark petrels from the ferry as we neared Tokyo Bay quite far north. They were more like Band-rumped or British Storm-Petrel in size, and had the obvious paler rump and shoulder markings, similar to a dark British Storm-Petrel. I also ruled out Least (O. microsoma), which is also on the Japanese list as an accidental, based on plumage tone. This was much browner on my birds than in any of the illustrations or photographs I have seen of Least.
These are the only small dark-rumped petrels I have seen in Japanese waters, despite 4 or 5 ferry trips between Honshu and Hokkaido. The only storm-petrels I have seen from the Honshu-Hokkaido ferries have been Leach's, in large groups in summer, and several Fork-tailed Storm-Petrels (O. furcata) in winter/spring.
You are right to suggest caution, Nick, as regards identifying all-dark petrels, or indeed any seabirds. The Pterodroma group are even more complex. I have seen several Pterodromas from the Honshu-Hokkaido ferries in spring, summer and winter, but all remained unidentified. Though I have my suspicions, they do not usually linger or approach close enough to give convincing views.
However, I have to take issue with your supposition that Swinhoe's is unlikely to occur off the Pacific Coast. If Swinhoe's can be caught and ringed on a regular basis off the North Sea coast and Irish coast in the UK, as well as in the Atlantic Ocean off Portugal and in the Eastern Mediterranean, what is to preclude their being seen off the Pacific coast of Japan, with breeding areas so close in the Sea of Japan? I do not know of any birds being caught or banded, or of any studies off the Pacific coast, but they could possibly occur at as yet undiscovered breeding colonies off the Pacific coast, perhaps even with Band-rumped or Leach's Storm-Petrels farther to the north. I think the distribution and habits of Swinhoe's are still far from well known.
Some Japanese birders go into Tsuruga Bay off Shizuoka prefecture regularly in spring, but I have not heard of any such trips along the rest of the Pacific coast. Most of the species they encounter are skuas, auks and terns. They added a second record of Little Auk to the Japanese list in 1999 and also saw large numbers of Aleutian Terns in spring 2004. This makes me wonder, and I know I am not alone in this, what else is out there! I believe that one of the big tour companies hires a boat out of Sendai to look for seabirds on its February trip. It would be great if birders could experience in Japan the kind of pelagic trips that are regularly available in the UK, the US, Australia, New Zealand and Peru. It would also be great if seawatching became as popular as it has become in other parts of the world.
Nick Lethaby (4/20/2006)
I agree that Matsudaira's and Tristram's Storm-Petrels are likely subject to much confusion. Just to clarify, I am not saying Swinhoe's Storm-Petrel is unlikely to occur off the Pacific coast. I am sure that it occurs regularly. What I am taking issue with is its relative abundance compared to Tristram's. My own feelings are that Tristram's is probably common and Swinhoe's is probably scarce to uncommon.
Swinhoe's occurrence in the UK is not really relevant to how common they will be off the Pacific coast of Japan. Obviously many seabirds can occur as vagrants anywhere. The major known Swinhoe's breeding colony is off Russia in the Sea of Japan, and the Japanese population is only estimated at around 1000 pairs (at least according to Birdlife). It also breeds in the Yellow Sea off South Korea at least (and I understand that this is a big colony). It is well established that the birds migrate past Singapore in September/October and May, and that they winter in the Indian Ocean. I agree that a possible migration route would take birds down the Pacific coast of Japan. I think, given the breeding distribution, it is more probable that most birds head out via the Sea of Japan and then join up with the Yellow Sea population to migrate down the Chinese coast towards the Indian Ocean.
Another factor to take into account is temporal distribution. Given their wintering range in the Indian Ocean and their migration period past Singapore, it seems that Swinhoe's are unlikely to occur off Japan before late April/May or after September. To take one record for example: in Brazil's book, he lists a record of 200 Swinhoe's Storm-Petrels from the Tokyo-Kushiro ferry in November. Even though the observers (Per Alstrom and Urban Olsson) are exceptionally good birders, I would argue that this record is nevertheless more likely a misidentification, based on the facts that it is difficult to get adequate looks at petrels from the ferry, that most observers have limited experience of storm-petrels, that other dark storm-petrel species are in the area, and that all Swinhoe's should be thousands of miles away in a different ocean at that time.
The main point I am trying to make is that I think quite a few of the Swinhoe's Storm-Petrel records off the Pacific are likely a result of the perception that it is a common bird there and therefore many people are comfortable in claiming all dark petrels as Swinhoe's. While it may be true that it is common, I am skeptical of this. It would be nice to see some more formal studies done via high-quality photographs or specimen collection in order to get an unambiguous idea of storm-petrel status off the Pacific coast.
PS (4/21): Generally, Bulwer's Petrels are pretty straightforward to id, so I don't think this species is much subject to misidentification.
Fergus Crystal (4/22/2006)
I do not have much experience with Bulwer's Petrel and Tristram's and Swinhoe's Storm Petrels in Japan: I have observed each species on 2 or 3 occasions. I understand the difficulty of identifying these birds at speed on a passing ferry, and that distance, light, weather conditions. etc., all play an important part in identification.
I think that these three species are fairly distinct, if not in terms of plumage, at least in structure and flight pattern. I believe Nick's comments detailing the flight pattern of observed birds demonstrate the amount of thought that has to go into observing dark petrels in Japan. I think that size estimates (particularly of wingspan) are very important in the field, and size is usually possible to gauge against nearby Streaked Shearwaters in Japanese waters.
I observed Bulwer's Petrel in June 2004 off the Tokara Islands (south of Yakushima in the northern Ryukyus) and also single Swinhoe's Storm-Petrels in April and May of 2003 and 2004 off Yakushima (which is mentioned as a possible breeding site in Brazil 1990) and after typhoons off Sata in Kinko Bay, Kagoshima (singles, September and October 2004). For me, the most interesting record of Swinhoe's was of a single bird with 30 Streaked Shearwaters, noted from the Nagoya-Tomakomai ferry off Chiba/ Ibaraki-ken on 22 Dec 2003. This is a winter record from the Pacific seaboard, and I am convinced that the bird was this species. I also noted 2 Tristram's Storm-Petrels from that ferry on the same day; they were both seen singly offshore at Inubomisaki, Chiba. I also saw 2 single Tristram's in mid-May 2004 off Oita prefecture from the Kagoshima (Shibushi)-to-Osaka ferry. All sightings (with the exception of Bulwer's) were in the first 3 hours of daylight.
I have recently seen single skins of all three species in the Tiachung Museum, Taiwan. The Tristram's specimen was mislabelled as Swinhoe's Storm-Petrel. I recommended that the Tristram's specimen be relabelled, because it is or will be the first Taiwan record. All birds were picked up off the Taitung (east) coast during the summer months. In the hand, I noted Swinhoe's was easily the smallest, with a noticeably forked tail. The wing panel was dark and rather indistinct. Tristram's and Bulwer's were similar in length. The Tristram's had a fairly deep tail fork, and the palest and most distinct upperwing panel of the three. The most distinctive feature of Bulwer's was the long wings: folded, the wing-bends protrude much farther than in the other two specimens, and the wing cords appeared more 'bowed' in shape. Also, the wedge-shaped, forkless tail was distinctive. The upperwing panel was intermediate in tone between those of Swinhoe's and Tristram's.
Briefly, I have noted what I think are the characteristics of flight action and jizz for each species I have sighted:
1. Swinhoe's Storm-Petrel: Notably small-sized, hard to watch in the swell or in choppy seas from a passing boat. Flight seemed fluttery both in direct flight and in foraging flight, but in a 'feverish' way; birds seemed energetic and purposeful, rather than fluttering around 'loosely'. Birds mostly appeared very dark, but occasionally, when light shone on uppers, a slightly paler brownish panel on upperwings was noted. Tail often closed in flight, and appeared short, but fork still visible when seen close.
2. Tristram's Storm-Petrel: Noticably longer-bodied than previous species, with longer wings. Birds seen only in direct flight, not foraging, when flight appeared smoother, less 'manic' than Swinhoe's. On turns, birds soared with outstretched wings, when the wings appeared narrow and straight, almost like a shearwater. Tails appeared long, and the fork always held closed, even on turns (but fork tips still noticeable on turns). Even in darker conditions, the pale upperwing panel was quite noticeable, and of a 'milk tea colour'.
3. Bulwer's Petrel: Birds looked very dark (effectively black) alongside Streaked Shearwaters in strong summer sunlight. The flight action on moving birds was loose but powerful, banking with a distinctive 'bowed profile' (see excellent illustrations in Collins Bird Guide, Svensson et al., 1999), then pulling away from the shearwaters with a few strong flicks before banking in another long sweep. The bowed profile gave the impression of loose flight, but the compact jizz and speed of flight indicated flight power! It was almost as if an invisible puppeteer had been pulling the bird along by an invisible line during the banks.
Birds of Europe, Svensson, Grant, Mullarney & Zetterstrom, Princeton University Press, 1999)
While these notes are not detailed, they may be of interest. I have yet to go to the Ogasawaras, so I have not observed larger numbers of these smaller dark petrels off Japan, and I have not yet seen Matsudaira's Storm-Petrel. I think the fact that Swinhoe's Storm-Petrel has been observed off the UK should not be ignored. I also think that Alstrom and Olsson's observation of Swinhoe's in November should be taken seriously, as I believe they would have taken Tristram's into account when recording this observation. So few of us currently get out to sea in Japan, and there are so few opportunities for Japanese birders to go on reasonably-priced pelagics, that knowledge of birds occurring just offshore is lacking. There could easily be a population of Swinhoe's Storm-Petrels that breeds off the Kurils and migrates down the Pacific coast of Japan. Or indeed, this species could be an irregular winter wanderer in small groups in Japanese waters between September and March. It may not be outlandish to suggest that there could even be a new species of small, dark petrel that winters off Pacific Japan and looks very similar to Swinhoe's.
Yoshiki Watabe (4/26/2006)
Both Tristram's Storm-Petrel and Swinhoe's Storm-Petrel probably breed on Hachijo-ko-jima, about 7.5 km west of Hachijo-jima. No recent research has been done there, however. On Sangan-jima, Iwate prefecture, Swinhoe's Storm-Petrels stay from late May to early October, so the Swinhoe's that breed on Hachijo-kojima probably return from their wintering area in May too. Tristram's Storm Petrels, however, are winter breeders, returning to their breeding ground in November and leaving there in May or June. Thus, I think it possible that both Swinhoe's and Tristram's Storm-Petrels are among the dark-rumped petrels which are observed near Hachijo-jima in mid- and late April, though it is rather more likely Tristram's than Swinhoe's.
In my experience, Swinhoe's Strorm-Petrel is more fluttering in flight than Tristram's, while Tristram's often glides in flight, as Sean mentioned. Tristram's (and Matsudaira's) Storm-Petrels often follow boats, but Swinhoe's probably does not. Bulwer's Petrel is more similar to Puffinus shearwaters in flight and is marked with a long and thin tail. It looks like a 'flying cross'.
Nick Lethaby (4/26/2006)
Watabe-san, are there any population estimates for Tristram's and Swinhoe's Storm-Petrels in Japan, for both the Pacific and Sea of Japan populations? I saw an estimate of 1000+ pairs of Swinhoe's Storm-Petrels for Japan on BirdLife's website.
The late May arrival date makes sense to me, since I recall reading that the migration off Singapore in mostly in May (and I was surprised at how late this was). Unfortunately I cannot find a reference to the Singapore migration times.
Yoshiki Watabe (4/26/2006)
According to the Ministry of the Environment (Threatened Wildlife of Japan: Red Data Book, 2nd ed.,2002), the estimated population of Swinhoe's Storm-Petrel was 1000 individuals breeding on the islands of Japan in a 1984 study. It seems that 180 pairs bred on Chikuzen-oki-no-shima, Fukuoka Prefecture, and 500-600 pairs bred on Kutsu-shima, Kyoto. Also, breeding on Shiriya-zaki, Aomori prefecture; Sangan-jima, Iwate prefecture.; O-shima in the Nanatsu-jima Islands, Ishikawa prefecture.; and Hoshikami-jima in the Oki Islands off Shimane prefecture was confirmed. The population of Swinhoe's Storm-Petrels is presumed to be decreasing, and breeding has not been confirmed on Shiriya-zaki, Hoshikami-jima, Nanatsu-jima or O-shima in the Nanatsu-jima Islands in recent years. A small population probably breeds on Hachijo-ko-jima. Therefore, few individuals are likely to be observed near Hachijo-jima.
The population size of Tristram's Storm-Petrel is unclear in Japan, but over ten thousand pairs probably breed on Tadanae-jima and Onbase-jima in the Izu Islands. Very many pairs also breed on Kita-no-shima in the Bonin Islands. Some dozens of pairs seem to breed on Hachijo-ko-jima. Accordingly, the possibility of observing Tristram's Storm-Petrel seems to be higher than that for Swinhoe's Storm-Petrel near Hachijo-jima.
Nial Moores (4/26/2006)
Although Swinhoe's Storm-Petrel is not a species I am very familiar with (probably encountered on about100 days in total over 15 years), in South Korea it breeds at least in the East Sea (Sea of Japan) and in the Yellow Sea, in the southwest of Korea (on offshore islands), off the west coast and very likely at least as far north as off the southwest coast of DPRK. It can be encountered especially in August and early September from domestic ferries in Korean waters (personal peak count from one ferry of 159 on September 4th in the Yellow Sea, 100-150 km west of Incheon); seen also in the East Sea/Sea of Japan in mid-summer months in both Korean and Japanese waters. Birds are rarely encountered off the DPRK-Incheon coast into September, but they can still be seen reasonably regularly 400 km south at that time in southwestern waters.
Dr. Lee Ki-Seup, who studied one main breeding colony at Chilbal Island in the southwest, says that it is very vocal from late May through late June or even early July. The same breeding colony was also filmed as late as October, when fledging birds were still present.
These sketchy details suggest to me that breeding colonies are not necessarily active at exactly the same dates within this region, or even perhaps within this country, and that its presence in waters near Japan or Korea into November is therefore unlikely to be very exceptional.
Swinhoe's do sometimes follow boats (at least, commercial ferries), though it seems most often at a 100-200 m range. Their flight varies,of course, but their most typical flight in fairly calm conditions is rather powerful and graceful, with sudden swerves and banks and arcs (presumably the manic flight described above). The species type they probably most remind me of in flight is Oriental Pratincole: typically far more confident than the much more fluttering flight of either British or Wilson's Storm-Petrels, but less powerful than e.g. the larger Tristram's (a species with which I have even less experience).
There are still no records of Leach's or Tristram's in Korean waters - and still only one record of Sooty Shearwater, suggesting that seabirds in Korean waters are still far from being well known. It would be very interesting to hear more of other people's experiences with petrels in this region. I also have seen apparently very small, all-dark or blackish petrels two or three times. About 30% smaller than the Swinhoe's they were seen near, with a more fluttering flight, they were seen from a commercial ferry in waters between Fukuoka and Ikinoshima in the summer months in 1991 and 1992: very young and runty Swinhoe's, optical illusions, or something else?
EDITOR'S NOTE: There is further discussion integrated with this topic at BIRDFORUM.net.