A perplexing gull

As a starter, note the pristine white tips to the inner primaries - surely these would be far more worn if the primaries were a year old...

I've not updated the blog for a while and my next post was going to be a summary of the rather exciting birding that I've experienced over the past month - however, I feel like it's worth getting another gull essay out of the way first. The header image perfectly sums up the absolute headache in aging this Lesser Black-backed Gull at Stanton St John on 15th June. In my submission to the BBRC for the four Baltic Gulls from Appleford in 2021 I said something along the lines of "the biggest confusion when identifying 2cy fuscus is not 2cy intermedius/graellsii, but 3cy intermedius/graellsii". It seems that I may have potentially fallen for this trap myself, although I suspect this particular bird will continue to defy any attempts at identification!

The water level at Shepherd's Pits has been slowly going down over the past couple of weeks and it is now looking very attractive to waders. Being only five minutes from my house I can quickly check it on my way to and from other sites, and I've been monitoring a pair of Little Ringed Plovers that arrived a few days ago. I initially thought these must be failed breeders and were surely early migrants about to head off - but they've taken a liking to the pit and have offered some great photographic opportunities.

There were only a handful of gulls present so there seemed little point in hanging around. I was just about to leave when I saw a new gull flying in from the east, heading directly towards me. For some reason it gave off a curious impression and even some alarm bells for heuglini began ringing - I could see that the wing was fairly dark and fresh whilst the upperparts seemed paler than what I would expect for fuscus. I took a series of photos as it flew around the pit a few times before settling on the small mound in the middle next to a 3cy graellsii Lesser Black-backed Gull.

I got the scope out of the boot of the car and the bird was close enough to obtain detailed views. Now that it was on the deck it seemed much more like a fuscus, at least in terms of its attenuated structure, moulting white head with a thin, insipid-coloured, black-tipped bill and dark eye - although I couldn't get past the fact that the mantle was much too pale. Not only was it pale, but the base colour of the feathers was slaty-grey rather than brownish - lighter, in fact, than the 3cy graellsii next to it. I could see that most of the visible scapulars were bleached second-generation feathers with broad white fringes (apart from maybe a few in the middle of the mantle between the wings, visible in flight), and wondered if that could be an explanation. Indeed, if some dark third-generation feathers were beginning to come through I wouldn't have questioned my identification. The second-generation scapular and covert pattern (with a single dark line down the middle of the feather, lacking any anchor markings) is most typical of fuscus and intermedius although can sometimes be found in graellsii.

The primaries seemed blacker and fresher when compared with the 3cy next to it, and looking through the photos on the back of my camera confirmed my thoughts that these were definitely second-generation primaries, being dark and rounded. In fact, they were in great condition, with pristine white tips on some of the inner primaries. I also noted that P1 had been shed, and thus the bird was starting its third wave of primary moult. Bizarrely, there were only eight primaries present on the left wing compared with the right wing, and although the images are difficult to interpret, I'm pretty certain this is because P10 was missing on the left wing. The gap in the inner primaries where P1 was shed didn't seem to be larger on the left wing compared with the right wing, and P10 usually appears a tiny bit shorter than P9 on an extended wing, which can be seen in the right wing but not the left. All secondaries and tail feathers were moulted to second-generation, and the tail did not have any white third-generation feathers (often found in 3cy intermedius/graellsii). The underwing was whitish but not completely white - a feature that is variable in fuscus. I knew that pale fuscus can be found in the core range so if the moult was right I didn't see this as too much of a problem, so I decided to put the news out as a Baltic Gull on the local WhatsApp group - although I was still doubting myself a bit.

Digiscoped image showing a little more detail - including some fine streaking on the neck that is being moulted out

I continued watching the bird (which wasn't doing very much) and Tom Bedford soon arrived on his bike. I got him onto the bird but I was beginning to feel more concerned and voiced that the mantle was too pale. I was also beginning to wonder if my aging was correct. After just a few minutes it took off with its graellsii companion and flew southwest. We waited a while in case it returned and observed the Little Ringed Plovers displaying and copulating, but there's been no further sign of the gull - thanks to Justin and Lew for checking.

I find birds like this very interesting but they invariably cause me to waste a lot of time searching through hundreds of gull photos online. At the end of my research I'm barely the wiser about this bird, but will attempt to summarise my thoughts. Since the identification of fuscus hinges on accurate aging, this is the main area I will try and address.

Other than the bare-part colouration, dark eye and mantle condition (all of which indicate a 2cy), two things stand out to me when trying to age this bird - the state of wear in the primaries and the timing of the third moult wave. In the field, I had judged the primaries and secondaries to be in a very good condition compared with the 3cy bird next to it (as well as my experience with watching a lot of 3cy individuals recently, most of which have had obviously worn, brown second-generation primaries), but on reflection, I do wonder if the fresh primaries of a 2cy fuscus would appear even blacker and more distinct, although assessing this in harsh sunlight can be a real pain. This feature is also variable depending on exactly when a 2cy fuscus replaces its primaries, but we might expect these to be up to six months old if that were the case. In the flight photographs, the primaries don't show too much wear or bleaching, but there are small areas of damage evident on some of the feathers, and maybe a slight brownish cast. Overall, the condition of the primaries is basically in between what I would expect a typical 2cy fuscus and 3cy intermedius/graellsii to look like - distinctive, but not blindingly fresh. I can find images of both 2cy fuscus and 3cy intermedius/graellsii where the condition of the primaries match this bird.

Crop from an image above - of course a comparison between only two birds is probably of limited value, but illustrates the difference in primary condition between the odd gull (front) and the 3cy graellsii (behind), in which the feathers are considerably more worn, brown and frayed

The dropped P1 indicates that the third wave of primary moult has just started. This is very early for fuscus - I have found photos of a few individuals like this in mid-June, but usually they begin moulting in July or August. In contrast, the timing fits well with a 3cy graellsii/intermedius starting its third wave of primary moult. I find it hard to reconcile the severely retarded mantle moult of this bird with comparatively advanced wing moult, if it is indeed a fuscus - although I guess stranger things have happened. I think these two features combined are possibly more supportive of a 3cy rather than a 2cy, although I really don't think those primaries look a year old (as they should be on a 3cy graellsii/intermedius). I have never seen a 3cy bird like this and have failed to find any images of one that matches its overall appearance either. It must have a hormonal deficiency for the bare parts to be so dull at this age, and the upperparts basically look like a 2cy intermedius. Almost all 3cy graellsii/intermedius will have replaced at least some of the scapulars and coverts with plain third-generation feathers at this point.

The alternative explanation is an unusually pale 2cy fuscus or an intergrade with intermedius from one of the mixed colonies in Norway, which could explain both the colour and weird moult. Pale birds are not too infrequent in the core range and matters there are complicated by heuglini. In Western Europe, however, I doubt these pale birds are acceptable as vagrants by rarities committees. I also don't see much here to support an identification as heuglini other than the mantle colour - I don't think I can really make a case for multiple moult loci within the primaries based on a missing P10 on one wing. Whilst heuglini can follow an accelerated primary moult strategy like fuscus, I think it is more likely to show a very advanced mantle (with lots of blue-grey third-generation feathers) combined with retained juvenile primaries - i.e. the opposite of what we are seeing in this bird.

Based on all of these conflicting factors, I've decided that it's best to leave this bird unaged and therefore unidentified - I apologise to Tom for his efforts in twitching it! I definitely think it would be rejected if I submitted it as a Baltic Gull, purely based on being too pale. To be honest, whatever it is, there is clearly something weird going on with this bird, both in terms of its moult and its colour - the whole plumage just looks kind of grey and washed out (in direct comparison with the 3cy graellsii, for example, it clearly lacks the brown tones found on the worn coverts), and I wonder whether this indicates some sort of underlying pigment abnormality.

So overall, a frustrating but educational bird, and illustrates the problem with fuscus identification when only assessing moult in the absence of other crucial features, such as the very dark mantle feathers. Recognising a Baltic Gull through accelerated moult requires accurate aging, as the whole point is that a 2cy fuscus is advanced enough to look like a 3cy graellsii/intermedius. I would welcome any thoughts on this bird and in the meantime I will continue looking.

Two nice finds

Northern Cross Valley Way

As soon as I've abandoned the patch it seems that I've begun finding birds again. Northern Cross Valley Way (a new site near Witney) has been attracting a run of good waders recently, including two Avocets on 26th April. I dipped these birds the next day and was pretty gripped by the potential of the site. It looks a lot like Appleford in its heyday, before it became overgrown - muddy scrapes just looking absolutely perfect for waders, all easily viewable from a raised bund overlooking the pit. This complex of lakes is also the first big area of water going west-east after the Cotswold Water Park, which no doubt helps to pull birds down as they migrate overland from the Severn to the North Sea.

On 4th May I arrived early in the morning and could see some small waders through bins on the far side of the pit. Setting up the scope, I began panning through them - Ringed Plover, Ringed Plover, Ringed Plover... Temminck's Stint. Nice. I'll admit I let out a little cheer. A self-found tick for me and only the second I've seen in the county after a frustratingly distant bird on Otmoor in 2017. It's a rather scarce bird in the county (last records being short-stayers in 2023 and 2018) and it's been a quiet spring, so unsurprisingly it generated a bit of a twitch - I think at least 30 people saw it throughout the day.

Initial distant views of the Temminck's Stint with a Ringed Plover, digiscoped at 60x magnification and heavily cropped!

First responders - Adam digiscoping at the front, courtesy of Justin Taylor

The bird then flew to the central spit and gradually moved closer, offering pretty good scope views as it crept around feeding. Still too distant for the camera even with the 2x extender, although I eventually managed to get this digiscoped image which I was happy with. After I left it apparently disappeared for a bit and was much harder to see amongst vegetation at the eastern end of the pit in worsening light as the day progressed.

Having been slightly rejuvenated by this find after a terrible effort/reward ratio over the past few months, I've been enjoying local birding a little bit more. I found a Black Tern at Appleford whilst doing a casual county day list by bike (not really a serious big day attempt - managed 104 birds and cycled ~150km) and enjoyed a couple of female Blue-headed Wagtails - sadly no males, nor rarer subspecies for me this year so far. I've also been making an effort to take more photos of common birds on my walks around Otmoor and Farmoor.

Black Tern, Appleford New Workings - the first in the county this year

Female Blue-headed Wagtail at Noke Farm, Otmoor

Note clear demarcation between olive mantle and blue-grey head, as well as pure white supercilium and throat

With a female flavissima

Barn Owl, Otmoor

Bittern, Otmoor

Marsh Harrier, Otmoor

Amazing light that evening but had to use high ISO levels!

Swift, Farmoor

Yellow Wagtail, Stanton St John

Gulls have been rather irregular at Stanton St John this year but given that it's only five minutes from my house I've been making sure to drive past and check it quite regularly. Early morning on 8th May I was pleased to see a large group of gulls swimming in the water and loafing around the edge of Shepherd's Pits - clearly having roosted the night before. Immediately I saw a nice 2cy Caspian Gull which is clearly different to the lingering Farmoor individual - a good bird in Oxfordshire in May.

The pictures below are the Farmoor bird for comparison - it's been around for quite a while and has now started its primary moult. A similar-looking hybrid has also been present and is causing some confusion!

Having checked most of the gulls on the water, I began searching through the more distant flock loafing around the field margins. There was a partially obscured Lesser Black-backed Gull that looked interesting, being obviously darker and smaller than the surrounding birds. It was sitting pretty tight but I decided to take a few record shots, assuming that it was an intermedius.

Initial views of the bird obscured in the grass

After a few minutes it began walking around and I was struck by its blackish mantle and structure - remarkably long-winged and attenuated at the rear end. I also realised that it was a subadult, with black markings at the end of the bill and only tiny white tips to the primaries. I began wondering if this could be a 3cy Baltic Gull candidate, although knew that I had to get a shot of the open wing to assess the primary moult. The bird looked unusually advanced (more like a 4cy) and I was struggling to age it with any certainty.

The flock was soon spooked and the whole lot took flight - luckily, I was still looking through the viewfinder and had the camera in position to take some photos, although with rather suboptimal settings for a flying bird (shutter speed far too slow and only the central focus point active)! Whilst the photos are not the sharpest, I was pleased to document the spread wing from both the top and underside. Examining them on the back of my camera, I could see that there was no active moult and two generations of primaries present - a clear indicator of 3cy fuscus!

Blurred photo but really shows the difference in size with this 2cy Lesser Black-backed Gull

I think the photo below shows the moult limit most clearly, with the outer two feathers (P9/10) appearing more brown and translucent against the background, as well the tips being slightly less rounded and lacking the small white tips present on P1-8. All secondaries are third-generation feathers. The contrast between the mantle and outer primaries is limited (in graellsii and most intermedius there is typically an obvious difference in colour), and there is a slight brownish cast to the greater and primary coverts which is a good feature for aging immature birds compared with full adults. Other than the dark eye, black bill tip and lack of P10 mirror the bird appears remarkably advanced - it has a fully white adult-type tail (3cy birds across all three Lesser Black-backed Gull subspecies usually show a mixture of white and black-tipped feathers), an unmarked white underwing without any marks on the underwing coverts, and deeply saturated yellow bare parts. Based on the size, thin bill lacking gonydeal angle and rounded head, I think this is an obvious female.

3cy fuscus is currently not acceptable as a vagrant by most rarities committees in Western Europe, including the BBRC. Both graellsii/intermedius can also moult primaries in winter and thus have two generations of primaries as a 3cy in spring. With fuscus this moult strategy seems to be the default, whilst in graellsii/intermedius it is uncommon but tolerably regular. The reasoning, then, is that since this moult strategy is not unique to fuscus, it is theoretically possible for an intermedius to show features which overlap completely with fuscus, including blackish mantle and small, attenuated structure. Having said this, I've yet to find any photographs of ringed 3cy intermedius which look exactly like this bird... if anyone has any please send them! Indeed, the examples of advanced intermedius used to illustrate these arguments do just look like advanced intermedius rather than birds that are easily confused with fuscus. This Norwegian bird (link) is the closest I can find. Typical 3cy graellsii/intermedius tend to combine grey scapulars with a worn, brown, unmoulted wing (sometimes with a row or two of coverts replaced) - they are instantly recognisable in the same way that 3cy fuscus is a striking bird.

I'd also argue that some 3cy fuscus must be identifiable if they have retained second-generation primaries with small P10 mirrors. Whilst this relates to only a small proportion of birds, as long as a candidate can be confidently aged as a 3cy this surely rules out graellsii/intermedius, which never (or almost never?) show mirrors on their second generation primaries. As far as I can tell, this hasn't been stated explicitly in any papers relating to fuscus identification, so the BBRC probably hasn't recognised it as a potential criterion for identifying 3cy fuscus.

Whether this latest Stanton St John individual is a textbook 3cy fuscus or an unusually advanced 3cy (or indeed retarded 4cy) intermedius - which also happens to share the full range of typically fuscus features - should be weighed up against the fact that confirmed 2cy fuscus are now known to be scarce but regular in Britain. I do think that the BBRC already applies a probabilistic approach to the assessment of certain rarities, where a combination of many "soft" features is sufficient to separate e.g. Zino's/Feas/Desertas Petrel, so long as these features are sufficiently documented through good photographs. I hope that this will soon be the case with Baltic Gull and I do plan on submitting this bird. I also think some clarity in the literature would help the BBRC with these complex assessments.

A bit of a deep dive now into Lesser Black-backed Gull moult, so an early point to bail from this blog post if one is so inclined!

One thing I found baffling after looking through lots of photos of 3cy fuscus was how many of them show large mirrors on P10 as well has having a moult limit in the outer primaries. Initially I thought that this was a sample of birds that happen to show mirrors in their second-generation primaries, now retained as worn outers. However, having a P10 mirror on second-generation primaries is a relatively uncommon occurrence - and some of these photos showed large mirrors (crossing over both feather webs) rather than the small and indistinct mirrors found on second-generation primaries.

The literature seems rather confused on this topic and I spent quite a while trudging through various papers attempting to explain why this pattern is observed. One school of thought seems to suggest that some birds which acquire P10 later on will show larger mirrors as the bird will be older and therefore produce a more adult-like feather. These may be individuals that arrest their first moult cycle in the spring of their 2cy and then replace the worn, juvenile P9/10 with second-generation feathers a few months after their peers. However, I'm struggling to find evidence that later moult leads to accelerated progression of feather pattern towards an adult-like pattern - indeed, I would have thought this would be correlated with earlier rather than later moult.

I found a more complete explanation from Chris Gibbins, who had obviously gone through the same thought process. His paper here (link) speculates that some of these birds have already moulted into third-generation primaries in their second winter, and those that show a moult limit have therefore replaced the inners with fourth-generation primaries. The most advanced fuscus may have already replaced half their primaries with third-generation feathers in the summer of their 2cy - so it is reasonable to imagine that the rest of these are replaced on arrival in their wintering grounds. Some Baltic Gulls would therefore remain a year ahead of other Lesser Black-backed Gulls in terms of their progression.

This makes some sense - it also seems possible that third-generation primaries are variable enough to show the full range from completely lacking a mirror to having a relatively large mirror. Indeed, the moult limit on the Stanton St John bird is very indistinct, with the retained P9/10 being barely browner than the inner primaries and not particularly worn/tatty. I find it hard to believe that these feathers are more than a year old, and have been used for three long migrations. Compare with the 3cy Lesser Black-backed Gull behind it - the second-generation feathers on that bird seem more worn than on the Baltic Gull, despite being grown a few months later. If you assume that the moult limit in the outer primaries in 3cy fuscus is between second- and third-generation primaries, the criteria for identifying 3cy fuscus suggested by Winters (2006) in his Dutch Birding paper seems to make more sense for intermedius! Perhaps this could be explained by the bird having replaced its P9/10 in the late summer of its 2cy (i.e. a slightly slower-moulting fuscus).

It is unfortunate that the third and fourth generation primaries (especially the inner ones) look very similar on Lesser Black-backed Gulls, which means that it is hard to reconstruct the moult progression based on feather pattern alone. I would imagine that 3cy fuscus shows a lot of variation in moult, with some birds having a mix of second- and third-generation primaries and particularly advanced birds having fourth-generation inner primaries. Indeed, this is the conclusion of this article on gull-research.org (link - a fantastic resource) - scroll down to see a list of observed moult strategies in 3cy fuscus returning to Finland. The moult pattern shown by my candidate bird is described as the "common strategy". I guess then, the problem is how to age an unringed bird accurately in the first place, given that bare-part colouration is more variable - and then how these differ from 3/4cy intermedius.

It's possible that I have misunderstood all this completely so if anyone has any comments that would be greatly appreciated. Clearly I need to visit some fuscus colonies this year and get some experience looking at birds of various ages within their core range - again, looking for any recommendations for good sites!