Taimyr Gull Larus (heuglini) taimyrensis: an Update

Taimyr Gull Larus (heuglini) taimyrensis: an Update by Nial Moores


A valuable paper by van Djik et al. (2011) in Dutch Birding provides a much-needed review of the distribution of the poorly-known Taimyr Gull Larus (heuglini) taimyrensis. On present knowledge this taxon is confined (more or less) to the Taimyr Peninsula during the breeding season. Based on observations of breeding colonies and of a small number of banding recoveries, the authors make several important assertions:

  1. Genetic analysis of breeding birds shows that they represent a distinct population (= Taimyr Gull), with evidence of a measurable degree of genetic differentiation and no evidence of obvious introgression i.e. Taimyr Gull does not appear to be a hybrid population as suggested by Olsen & Larsson (2003);
  2. While there might be some degree of difference in the darkness of the saddle in adults (perhaps clinally?), and while there is variation in leg colour (with most pale yellow, but some pink) there is no indication that there are two species of gull breeding there;
  3. All six known long-distance recoveries to date were of Taimyr Gull that migrated south-eastwards from the Taimyr Peninsula towards coastal East Asia;
  4. The structure, plumage and bare parts coloration of Taimyr Gull photographed on the breeding grounds resemble closely birds that have been identified as Taimyr Gull during the non-breeding season in East Asia (including in the Republic of Korea [ROK], Japan and southern China) since at least the mid-1990s (Kennerley et al. 1995, Hoogendoorn et al.1996).


There are several older posts on the Birds Korea websites describing the status and the identification of Taimyr Gull in the ROK (including Moores 2003a, 2003b). This note aims to help clarify present understanding of migration phenology and also to support the identification of a Taimyr-type gull in Indonesia in 2008 (Trainor et al. 2011) and especially of a Juvenile>First-winter Taimyr Gull photographed in Singapore in November 2011 by Francis Yap and colleagues.


Migration and Moult

Contra Moores (2003b), there appears to be no evidence that nominate Heuglin’s Gull has been recorded in the ROK (a taxon believed to migrate largely south-westwards from breeding colonies: Olsen & Larsson 2003). However, Larus (heuglini) barabensis are now suspected to be regular or near-regular in the ROK in small numbers (Moores 2005). The vast majority of heuglini-type gulls in the ROK and elsewhere in East Asia are believed to be taimyrensis.

Taimyrensis start to depart breeding colonies on the Taimyr Peninsula in late July and strong southward migration has been noted in August (van Djik et al. 2011). Largely as stated in Moores (2003a, 2003b) Taimyr Gull start to arrive in the ROK in September; appear to peak in number in October; remain through the winter in rather smaller numbers (presumably as most migrate further south); and then increase again in March, with some remaining into May. Single-taxon flocks of 40-50+ Taimyr Gull are regular in open sea especially in March and April, with strong northward movement in late April suggested by counts of birds at sea as well as in harbours (Moores 2011). Birds are already back on breeding grounds before the beginning of June (van Djik et al. 2011).

In September, adults arrive white-headed and usually yellow-legged in the ROK. They then develop head-streaking by November; undergo outer primary moult largely in December and January; and re-attain breeding plumage (white-headed and with brightest bare parts) in late March and April, shortly before departure. First Calendar-years, which are often with flocks of adults and immatures, retain a juvenile-type plumage into December. Wear and limited moult becomes increasingly evident by March and April. Second Calendar-years at that time appear largely white-headed (though streakily so, especially on the crown, and with a darker eye-patch) and some by April also have a few darker slate-grey feathers on the mantle (i.e. they appear much more advanced than vegae and much less advanced than mongolicus of a similar age).



As with most large gull taxa, there is variation in structure, plumage and bare part coloration. Generally, Taimyr Gulls appear neat, well-proportioned, flat-backed and long-winged (when not in primary moult) and rather short-billed (though some can be obviously more bulky-looking).

Adults have a dark slate-grey saddle, cleanly darker than vegae and the vast majority of mongolicus in direct comparison. Although many observers suggest that there is much variation in the darkness of the saddle, this does not seem to be borne out by observations of birds within “pure” flocks in the ROK. It therefore seems likely that while Taimyr Gull might include darker and slightly paler-saddled adults (and that such birds might occur together in the non-breeding area), many of the perceived differences of shades of grey in birds within the ROK are most likely due to misidentification or perhaps to under-appreciation of apparent differences caused by changes in light or the relative position of birds to the observer.

In adult Taimyr Gull, the black on the primaries of the open wing is extensive (often extending onto P4) and contrasts strongly with the saddle and most of the upperwing. Taimyr Gull lack the obvious string of pearls often shown by vegae and invariably shown by adult Slaty-backed Gull Larus schistisagus. While many show a white mirror only on P10, some also show a white mirror on P9 (apparently unlike nominate heuglini). There is much variation in the extent of head-streaking shown by adults during the non-breeding season. On arrival in September and through October, most adult Taimyr Gull are white-headed. Head streaking is acquired gradually, with a few showing streaking largely confined to the nape (though with a darker eye-patch), and the majority becoming more heavily streaked – though lacking the coarser thumb-print smudges on the breast sides shown by most vegae. Unlike vegae, adult Taimyr Gull perhaps never appear densely-hooded. Adult bare part coloration is also somewhat variable. Most remain obviously pale-eyed (though usually with much brown flecking) and are also bright-billed. The bill base is bright, warm yellow, and the gonys is orangey-red, usually with some hint of red bleeding up to the cutting edge and even onto the upper mandible. The distal third of the bill is paler, and there is often a small dark mark near the gonys. Leg colour is frequently yellow in adults, most intense in March and April. However, many show pinkish tones admixed and in mid-winter a small number appear to be pink-legged (lacking obvious yellow tones). More research is needed to determine whether such birds are usually adults or rather sub-adults.

First Calendar-years and early Second Calendar-years are in some ways intermediate in appearance between vegae and mongolicus of the same calendar age. Through until early December, they are largely brown and scaly-looking on the upperparts, paler brown-washed below, with a darker mask reaching back onto the crown (often contrasting with a paler collar), and usually with dark tertials only edged with pale or a weak holly-leaf pattern. On the spread upperwing, there is a weak pale blaze on the inner primaries (much less prominent than in vegae and mongolicus), and variably darker greater coverts. By mid-December, many are starting to look more pale-headed (though with a dark eye-patch, some crown streaking and a coarse nape shawl, often tinged with rusty tones), with contrasting dark wavy bars or thin anchors on the mantle and scapulars. Usually, some blacker markings can also be seen on the scapulars, and occasionally a few dark grey feathers can appear on the mantle by February or March. However, most look bleached or worn and rather pale by this time. The bill is dark initially, showing some pink basally by mid-winter onwards. They are thus less clean-looking than mongolicus of a similar age (which is already in First-winter plumage by October, and looks largely white-headed by November), and more contrasting and paler-headed than most vegae.

Birds apparently develop more extensive adult-type dark grey on the mantle late in the Second Calendar-year, and perhaps become more adult-like in the late spring of their Third Calendar-year when identification as Taimyr Gull is rather more straightforward. Based on the description in Olsen & Larsson (2003), Taimyr Gull therefore appears perhaps to be similar to but slightly slower than barabensis and nominate heuglini in moult timing and progression.


The Singapore Gull

Most of the easily-accessible English-language literature on the identification of First and Second Calendar-year mongolicus, vegae and taimyrensis is still rather poor. Brazil (2009) for example includes “taimyrensis” in quotation marks, stating that it is “often considered an intergrade between Vega and Heuglin’s”, perhaps following Yésou (2002) and Olsen & Larsson (2003). While both adult taimyrensis and heuglini are depicted, the standing and flying drawings are apparently of First-winter nominate heuglini, as depicted in Olsen & Larsson (2003). The image of juvenile vegae shows pale-fringed almost solidly-dark tertials (when most show much patterning on the tertials) and the text for Mongolian Gull states surprisingly that “separation of 1st-/2nd winter [Mongolian] from Vega is beyond the scope” of the guide, when separation of both species at these ages is generally straightforward. This confusion in the literature continues to make the identification of out-of-range immature gulls especially challenging in eastern Asia.

A request was therefore sent out by Francis Yap and fellow birders in Singapore to help confirm the identification of a gull they found and photographed on November 20th during a pelagic trip (shown in images 1-3; images used with permission): potentially a national first record.



In the hope that this might be useful to help with discussion on this and other gulls, a shortened version of my reply (sent on November 22nd) is pasted in below:

“Allowing for all of the many caveats of identifying a lone out of range gull on a set of images (rather than in the field), I personally would identify this as a juvenile-type Taimyr Gull Larus (heuglini) taimyrensis. To arrive at that conclusion it is first necessary, of course, to eliminate most other gull taxa.

It is clearly a juvenile-type large gull. All large white-winged gulls can be excluded by e.g. the darkness of the primaries, as can Slaty-backed Gull (as e.g. typically shows a much paler underside to primaries). This gull is therefore most likely one of the three most numerous large white-headed gulls (“LWHG”: Vega, Mongolian and Taimyr Gull) or even less likely one of several extralimital taxa…

Mongolian and Caspian L. cachinnans can immediately be excluded by the state of moult and by the lack of white head and heavily marked grey saddle (shown from September onwards: we see no dark juvenile-type Mongolian after about mid-August – as all are already moulting into the much more distinctive First-winter plumage). Vega at this time of year typically look rather more blotchy than this bird, might show some more obvious wear, and would be expected to show more obvious paler bases to the inner primaries by mid-October and especially by November (as would American Herring Gull L. smithsonianus, which would also look much darker and dirtier than this bird, including on the rump). The apparently solidly dark bill and rather prominent white on the rump of the Singapore bird also helps to exclude even less expected species like Californian Gull L. californicus.

The identification thus tends towards either nominate heuglini, barabensis, taimyrensis or perhaps L. fuscus. All of these are considered on present knowledge to be extremely similar (and by some to be conspecific). Based on some literature (e.g. Olsen & Larsson 2003) and on subsequent correspondence, however, barabensis might be expected to look more white-headed and more advanced in moult by November; nominate fuscus should look rather more solidly dark on the upperwing (e.g. across the greater coverts) and nominate heuglini should also look rather darker on the upperwing, lacking such obvious barring across the rump and also lacking a hint of paler contrast on the inner primaries (typical of taimyrensis, and increasingly apparent with wear).

The combination of fairly slight structure (at least, not massive); still fresh juvenile-type plumage in mid-late November; obviously brownish head; (only) slight hint of pale on inner primaries; dark spots on scapulars; darkish bar on greater coverts; and mid-brown tones all appear to me to fit taimyrensis very well (though might expect to see a few more rusty tones somewhere in the plumage: perhaps hinted at anyway in some of the images?).

Although this might be less important than the features visible on this bird, its behaviour (found in open sea) and the timing of its occurrence would not be unexpected in taimyrensis. This tends to be the most pelagic of the three common LWHG here in the ROK at least…This taxon is also the one that regularly winters furthest south in East Asia, south to Thailand.”

Do you agree with this identification? Detailed and informed differences of opinion are of course welcome!


See also:



Brazil, M. 2009. Birds of East Asia. Helm Field Guides.

Hoogendoorn W., Moores N., & T. Morioka. 1996. The occurrence and field identification of adult “Herring Gulls” with yellow legs in Japan. Birder 4: 64-73 (in Japanese).

Kennerley, P., Hoogendoorn, W. & M. Chalmers. 1995. Identification and Systematics of Large White-headed Gulls in Hong Kong. Hong Kong Bird Report 1994: 127-156. Dec 1995.

Olsen, K. & H. Larsson. 2003. Gulls Of North America, Europe, and Asia. Princeton, NJ: Princeton University Press

Moores, N. 2003a. Status and Distribution of Gulls in South Korea, with particular reference to “Gulls of Europe, Asia and North America. http://www.birdskorea.org/Birds/Identification/ID_Notes/BK-ID-Gull-status.shtml

Moores, N. 2003b. A Consideration of the “Herring Gull Assemblage in South Korea” http://www.birdskorea.org/Birds/Identification/ID_Notes/BK-ID-Herring-Gull-Page1.shtml

Moores, N. 2005. Steppe Gull barabensis in South Korea: A Step Closer to Identification? http://www.birdskorea.org/Birds/Identification/ID_Notes/BK-ID-Steppe-Gull.shtml

Moores, N. 2011. The Distribution, Abundance and Conservation of the Avian Biodiversity of Yellow Sea habitats in the Republic of Korea. Doctoral thesis, University of Newcastle, Australia.

Trainor, C., Imanuddin & J. Walker. 2011. Heuglin’s Gull Larus heuglini on Wetar Island, Banda Sea: the first Indonesian record. Forktail 27, 90-91.

van Dijk, Kharitonov, S., Vonk, H. & B. Ebbinge. 2011. Taimyr Gulls: evidence for Pacific winter range, with notes on morphology and breeding. Dutch Birding 33: 9-21.

Yésou, P. (2002). Systematics of Larus argentatus-cachinnans-fuscus complex revisited. Dutch Birding 24 (5): 271-298.


Taimyr Gull Larus (heuglini) taimyrensis. Birds Korea Blog. Nial Moores, Dec 2011


6 comments on “Taimyr Gull Larus (heuglini) taimyrensis: an Update

  1. Thanks for the interesting article. I’d like to add these images for contrast and opinions.

    1) Presumed 2nd calendar-year Taimyrensis, Eochong Is., April 15th 2010, with more advanced/ worn individual in background.


    2) Adult taimyrensis. Eochong, April 9th 2010. Appears short-necked (?)


    3) Not really sure what this is. Taimyrensis or…? Eocheong, April 7th 2010


    • Thanks for your comments. These gulls are somewhat easier to ID when seen in the field, and in images when there are other birds present for direct comparison and most especially when clear details of the open wing can be seen. The strong holly-leaf tertial pattern on the rear gull in your top image suggests vegae to me (this would be easier to confirm either way by seeing the extent of pale on the inner primaries); respectfully, all the images look rather overexposed , making it hard to assess plumage tones; and the primaries in the middle image on one wing are cut out of the image – making (for me at least) analysis of the outer primaries more difficult (does black extend to P4 or not?). The bottom of the three images does look like taimyrensis, but again it is difficult to assess plumage tones and bare parts coloration, and we can see no details of the spread wing. I would be happier to suggest (if this helps at all?), based on the April date, to say it is not e.g. mongolicus or barabensis, as it retains the typical amount of nape-streaking that you would expect on either taimyrensis or vegae in April (and while barabensis should be largely white-headed through the winter, mongolicus are also pretty much white-headed and clean-naped by March).

      Re the Ujihara’s website: this is indeed an excellent source of images and ID information on East Asian gulls. I cannot see the date of the image that you link to, but (of course) would also identify this as a First-winter taimyrensis. It shows all of the features described in the blogpost above, with a rather more obvious blaze apparent on the inner primaries than shown by the Singapore bird. There also seems to be a little more bleaching on the coverts, yet the head is still largely dark – suggesting perhaps that the image was taken in late December or perhaps in January. Was there a date with the image?

  2. Thanks for your thoughts. The 1stW image is undated, although it says “mid-winter”. Here is the page:

    Image 3 gave the impression (to me at least^^) of being lighter built, less stocky, more round-headed and perhaps shorter-legged and flatter-backed than a typical taimyr. All of these being features ascribed to barabensis. Add in the confusion in the plumage tones to the variables. But as you say, neck streaking should not be on a spring adult.
    Looking at a lot of March photos from Oman, I get the sense of quite a different, consistent and distinctive overall look, being eg slenderer, with flatter, longer bill, more extensive black on the primaries etc. (See also the contrast with heuglini).

    If taimyrensis is a genetically distinct population (rather than made up of hybrids and intergrades), is the same true for barabensis?

    • Thanks for your comments. There are quite a few folks out there who know barabensis well: lets hope that they read this thread and decide to comment as their insights would be extremely helpful.

      Before anything else, it is first important to emphasise (as always) that there is a huge amount of variation shown by individuals of the LWHG taxa – between males and females, bigger and smaller individuals, between different populations of some of the taxa, and of course between ages. The timing of moult (and appearance of wear) can also be affected by a number of factors. This is a very tough group to identify (hence the frequent use of putative and tentative in the few ID articles on these taxa that we have had time to post).

      Identification to taxon is therefore best based on a range of characters (some of which are extremely fine in detail), and the determination of patterns within a large number of individuals, rather than on gut feeling or a rapid assessment of structure or moult alone in single or in a small number of individuals. When you try to id these gulls, it seems best to have a list of boxes in your head which can then be ticked one by one and in order. If you cannot tick every box, then either best to leave the bird unidentified, or to see this as evidence of variation within the expected, common taxon (as wiser to accept you will encounter more variation within the expected common taxon than to try to assign such variation to unexpected variation within a taxon considered to be very scarce in that region).

      So, as you write, bird 3 might show a structure that might fit with some barabensis (but which is also within the range of taimyrensis ) . However, it shows head-streaking in March which almost no barabensis should show (apparently), but which most taimyrensis do show. It also shows other taimyrensis features. Therefore, as this bird is in the ROK within the geographical range of taimyrensis, and shows no non-taimyrensis features, it is best considered to be a taimyrensis. If the same individual was in the Gulf or East Africa, it would, however, probably be wiser at this time to assume that this was variation within heuglini or barabensis, rather than an extralimital taimyrensis – unless there are enough individuals there similar to it to reassess the status of taimyrensis there, or unless it is genuinely far enough outside of the expected features range of heuglini and barabensis to make experienced observers convinced it was of a different taxon. This is something folks in those regions would be in the best position to comment on.

      The present identification of the various taxa of LWHG here in the ROK is therefore not based on the confident identification of the occasional individual on single characters, but on the determination of a range of characters that can be used (structure, plumage, behaviour and soon – lets hope – vocalisations). The vast majority of individuals fit very comfortably into these main categories. A few individuals that show marked (and perhaps consistent) differences from these main patterns can then be ascribed tentatively or otherwise to other taxa that show a different pattern of characters. Others that appear oddly intermediate or bafflingly different might (or might not) be hybrids, further confusing the process (however, IF you accept the existence of birulai, which I do, then where is the evidence of hybrid influence in taimyrensis?).

      This was the kind of process that first led to the development of ID criteria for vegae and then taimyrensis and mongolicus (with these features subsequently supported by observations on the breeding grounds and a number of marked birds in the latter two taxa). Some individuals are obviously outside of the expected range of these main taxa. Some of these (a very few) fit very well with what is known of barabensis (each box can be ticked), and after some discussion with a few folks who know barabensis, these have been ascribed as barabensis. But again, only when every box has been ticked. The same is true of the very few cachinnans that seem acceptable.

      However, there are probably (many?) dozens of cachinnans-types and barabensis-types in the ROK each winter which are still best left unidentified. We do not know enough yet to define the expected range of variation within this region for some of these more western taxa (e.g. how different are eastern cachinnans from western cachinanns; are there differences between northern-breeding and southern-breeding barabensis etc?) yet alone about whether they hybridise or not.

      As far as I understand it (and very happy to be provided with better information and papers etc), while these gulls show alot of predictability within these brackets of variation, very little actual genetic variation has yet been detected within heuglini, and barabensis and heuglini are considered to be genetically extremely close to each other (so our decision at present to consider barabensis, heuglini and taimyrensis as subspecies of Heuglin’s Gull). Personally, I have a poor understanding of genetics, but accept the logic of the argument that rapid separation of these taxa so that they could exploit different ecosystems (separated by e.g. deserts, mountain ranges or glacial periods) will have led to more variation in structure and moult timing than might be evident in part of their genetic make-up (such genetic change takes time). This is not unique to these gulls.

      Some believe that the observable differences are consistent and useful enough to recognise these taxa as different from one another (as species or subspecies); others dont. Some authors include heuglini with Lesser Black-backed Gull L. fuscus, and others still consider barabensis (and mongolicus) to be part of cachinnans.

      The Birds Korea checklist is based on a combination of our own experiences within this region and on the best expert information made available to us.

      If this is of particular interest, there is a lot of expert literature much of it accessible online on the evolution of these taxa – and a growing body of information on identification (including some great websites), mostly from better-birded areas with committees of highly-skilled birders and ornithologists. Please post links to these as you find them if you would like?

      Feel at this time, what we can best contribute from here (outside of the breeding range of most of these taxa) is better information on the observed variation within the more numerous taxa, and insights that can help to correct the popular ID literature (much of it apparently based on inadequate research or lack of familiarity with these taxa in the non-breeding range).

      Hope this is helpful, at least to explain the consistency of positions on these taxa taken in ID articles, latest bird news archives and our checklist?

  3. ..Thanks. I’m interested then, to see if you agree with my assesment of the following, photographed along the NE coast in late February 2009 (picture courtesy of birddb.com ):


    Looking at your ID note “Gull Assemblage”, it appears that the right bird is a Type A Adult winter Mongolicus.

    By contrast however the left bird appears to show:
    1) a small dark-eye
    2) small, all-white, round head
    3) unusually bright bill with dark band and pale tip
    4) subtly darker upperparts than mongolicus
    5) flattish lower mandible/ very weak gonys angle

    Which ticks several boxes for barabensis. However there are problems- barabensis should be darker, have a yellow eye, smaller white primary tips, and no neck streaking.
    Looking at the excellent illustrations of cachinnans in the Collins guide, everything begins to look good for an adult W Caspian Gull, until I read that neck streaking would be unusual, and the outermost primary should have a long white tip (apparently not shown in the picture).

    So then are we looking at a darker type C mongolicus?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.