The unsound approach

Ok, so anyone following latest sightings knows the story, on the 24th September 2008, AL and Paul Brown had a brief view of a very orange juvenile Hen Harrier ssp on North Ronaldsay. AL was able to get a few seconds of poor-quality video of the bird as it passed about 300 m away. The bird was seen again, poorly, the next day but not subsequently. I (AL) flogged this like a dead horse for a few days whilst still on Ron, managed to cobble together some video-grabs, and forwarded them to anyone who would listen. Which, in the end was quite a few people. Most were luke-warm about it, unsurprising considering the crap images; vital features could not be resolved and the bird looked to be left in limbo. Weeks past, I was unable to sleep or eat and would wake up in the middle of the night shouting BOA! Actually thats not true, I slept soundly but was generally pissed off that I hadn't identified it and that the circumstances had not conspired to deliver me a proper bird like a Middendorf's Grasshopper Warbler. Anyhow, I thought I could walk away from it until Julian Hough kept appearing in my inbox, telling me I was as mad as box of weasels if I didn't pursue it. With this encouragement and the erection of a thread on Surfbirds, my interest was temporariy re-galvanised. John Bell - the 3rd observer forwarded on John Martin's BB paper, dealing with the acceptance of the 1982 Scilly bird. At the time of the sighting we only had Garner and  Wheeler and Clark which although useful did not provide enough information to seal the deal. Had I read them beforehand however, maybe I would have paid more attention to some of the key features - despite having seen hundreds of Northern Harriers (whilst living in the states in 2000/01 and subsequently extended visits in 2002 and 2004) I really hadn't paid much attention to them - other than to note that fresh juveniles were typically more orange than orange Tango and looked quite attractive. I have attempted to revisit the North Ron harrier below, using state-of-the-art print screen techniques using WMP on maximal size. If anyone thinks they can do better I'll forward the original file.    

Fig. 1 typical juvenile hudsonius, Connecticut, October (Julian Hough, http://www.naturescapeimages.net) note 6 barred 8th primary, contrasty soild-ish boa and limited flank streaking.

Figs 2 & 3. Hen Harrier ssp. (A. Lees) The best grabs of the underwing resolve 5 barred 8th primary (average for hudsonius, exceptional for cyaneus), apparently unstreaked orange breast (supposedly undocumented in cyaneus -  but see this rufous bird and almost this one - rare even for hudsonius although cf this one), orange-cinnamon underparts - compare with Fig. 15 which show a truer likeness of the colour [filthy scope, dull conditions] (average for hudsonius, exceptional for cyaneus), IMO the bird shows little white around the eye, comparable with many images of Northern. The main bone of contention is the nature of the boa - here apparently quite diffuse, but still prominent, apparently three-toned but this may be a photographic artefact.  

Figs 3 & 4. (A. Lees) 3 is perhaps the sharpest image of the head and neck and appears to show a sharply demarcated boa contrasting with an orangey upper-breast. This feature is also extremely prominent in Figure 4 - giving it a vested appearance,  note also the minimal white eye patches, cf this cyaneus with restricted streaking

Figs. 5 & 6, Both seem to show a well-demarcated apparently solid boa, the darker fore-breast may be caused by some streaking, as is the dingier area around the flanks? Upperwing appears very dark in Fig 5. Note pale inner-primaries! Some cyaneus can show a prominent thick boa

Figs. 6 & 7. Note dark-brown/blackish (in life) mantle with some orangey blotches. Hooded appearance is pronounced on right hand image, neither shows much white around the eye. The white rump patch was particularly striking large, a feature also noticed on the Scilly bird. Is there any evidence for any cyaneus showing a mantle colour comparable to this one - also cf Figs 5, 7 & 8.   

Figs 7 & 8 (A. Lees) If this was a mystery photo competition then I would probably plump for left: Snail Kite, right: Pallid Harrier. The upperparts of this bird were exceptionally dark.

Figs 9 & 10. Fig 9 is a frame taken at about the closest point the bird comes to the observers, it may be a video artfefact but there appears to be some streaking on the upper breast, almost bleeding from the boa. Head pattern is still extremely distinctive n both frames however. Compare the boa with this bird and this one.

The most Northern-esque Hen Harriers I could find are this one and this one  both from Estonia, the first has little streaking and a good orange(ish) underpart colouration, however, the boa is poorly defined and this bird only has 4 primary bars, the second seems to have 5 primary bars, virtually unstreaked yellowy-orange underparts but again a poorly marked boa. It would have been interesting to see the mantle on these critters.   

Figs 11 & 12 (Figs 196 & 197 in Martin 2008) (Barrie Widden) Although image quality of both the North Ron bird and the Scilly bird are not of the highest standard*, one can find more simlarities than differences. Chief among the latter are the apparently even more reduced white eye-patch in the '82 bird and the apparently more solid boa. The '82 bird was also supported by field descriptions from some of the UK finest. Our lack of any detailed description will probably leave this bird in limbo. However, for me at least I can not see any significant differences between the two birds that could separate them, considering the slightly lower resolution of the deinterlaced video-fields. 

*The slide originals are however of a much higher quality and therefore much more informative (JM pers com).   

Fig. 13.  juvenile Northern Harrier, Connecticut, December (Julian Hough, http://www.naturescapeimages.net) This is an advanced juvenile, note the boa and face pattern are congruent with the Ron bird.  

Fig. 14 juvenile Northern Harrier, Connecticut, December (Julian Hough, http://www.naturescapeimages.net) Streaking is most pronounced around the flanks, note the prominent pale leading edge to the wing cf Fig. 16.

Fig 15. Cropped and reorientated images of a Northern Harrier from CT (JH), the Scilly Northern (BW) and the North Ron harrier (AL). Allowing for image quality, the boa appears to be identical in extent (if not solidity) in all 3 individuals.  

Figure 16. montage of images of the North Ron harrier, these have been brightened and sharpened and the contrast altered, the plumage tones are a better reflection of the true colour of the bird in the field, with the underparts being orange and the upperparts a dark chocolate as Terry's intended.

Table 1 - Appendix 1 from Martin (2008), illustrating the features considered to be useful in separating juvenile Hen from Northern Harrier. The features supportive of Northern Harrier, and shown by the North Ron bird are marked with a tick mark; it seems to tick most of the boxes.    

Implications for the acceptance of extralimital hudsonius

I've spent more time recently than is healthy looking at harrier images but it doesn't take that long to find individual cyaneus away from the Atlantic fringe (and thus even more unlikely to be genuine vagrants) showing many characteristics of hudsonius. Cyaneus can be unstreaked on the breast, can show 5-barred primaries*, can show limited white around the eye, can show a quite solid broad boa  (this would appear solid at a distance) etc. Obviously hudsonius should only be identified with the full suite of featutes. However, many of these cyaneus share multiple hudsonius features, and all are at least quite orange. These traits may therefore be genetically linked and if this is the case there must be a chance of finding a cyaneus showing all these features at once. It is also possible to find individual juv. hudsonis without many of their own key features - like this one with a poorly-marked boa, this one with extensive streaking, this one with a poorly-marked face etc. So far at least, I have yet to find a cyaneus with six primary bars - a featue of some hudsonius that might be diagnostic. This feature is not shared by either the North Ron bird or the Scilly bird which would (hypothetically) be rendered unacceptable if this criterion were adopted.  

*this bird is very hudsonish

Circumstances

Had I found this bird on the Isles of Scilly I would have had no qualms about broadcasting it immediately as a putative hudsonius. North Ron is a different matter however, it lies close a large source population of cyaneus (albeit a declining one) which were recorded frequently during my stay - pertaining to various different individuals. Without better views, I didn't want to broadcast the news and drag everyone back to Ron for a bird much crapper than Cretzschmar's Bunting. I did speak to the diseminators of bird news about it and they concurred. As it happens the bird was never pinned, and it would probably have been a waste of money and carbon trying to twitch.

Raptors don't like oceanic crossings. However among broad-wings, harriers are much more apt to cross the sea than most others, if sanctijohannis Rough-legs can make it then so can Northern Harriers. Northern Harrier has even made it to Hawaii (Oahu and Midway). 2008 was like 1982, an exceptional year for Nearctic vagrants. On the same day as the harrier a certain heron was first seen in Galway. Winds weren't as suitable as '82 but a crossing aided by a low pressure south of Greenland would certainly have been feasible (Fig. 17). Southward migration of Northern Harriers occurs from August to November, with birds appearing in the Caribbean islands in October and in southern Central America in mid-October (MacWhirter and Bildstein 1996). The arrival of candidate transatlantic vagrants at high latitudes birds in mid-late September would thus be expected. What however is the chance of a transatlantic crossing versus a freak Hen Harrier? I don't know.  

Fig. 17 Pressure charts for 24th September 2008 and 22nd September 1982.

Should you (I) care?

Maybe. Wink et al. (1998) reported a genetic divergence of 1.7% between cyaneus and hudsonius, not huge for an intercontinental split but significant enough to be interesting. Ferguson-Lees et al. consider it to form a superspecies with Hen and Cinerous Harrier and favour a three-species treatment. Gene Flow must be low, at least across the Atlantic, but what happens across the Bering? I should probably let it go but sometimes, at least after a personally crap autumn one just has to riact to these things.   

Conclusion

I don't know. I appreciate why a question mark remains over the North Ron bird with respect to the nature of the boa but I struggle to see the Scilly bird as a significantly better candidate if this is the case. If the North Ron bird is a cyaneus then it is exceptional and perhaps inseparable in the field from hudsonius, see here for more information on apparent orangey cyaneus. My biggest problem is reconciling the features used to accept the Scilly bird with those used to 'dismiss' the North Ron bird, indeed even the features used to separate the two at all (hopefully that does not sound too bitter and twisted!). Maybe these images might inspire a bit more research and debate and an attempt to refine the character-index for the cyaneus super-sp. Really, I don't know, maybe we can trawl up more images of eastern nominate birds?

Acknowledgements

Thanks to John Martin, Martin Garner, Julian Hough, Killian Mullarney, Dick Forsman, James Gilroy, Stuart Piner, John Bell, Chris Batty and Brian Small for their comments on this beast.....