Monday, 21 October 2013

Apparent NORTHERN HARRIER in Cambridgeshire

Well after that big dark peregrine recently I wasn't expecting another heart attack raptor so soon - but yesterday a hen harrier at the Ouse Washes with a dark hood and unmarked bright orange underparts really put the wind up me. It had been reported a couple of days earlier at a nearby location and put out by the news services as a possible pallid but then dismissed by another observer as a warm hen harrier. Now it was passing right in front of the hide I was in at the Ouse Washes. This bird looked to me bright orange below and more importantly, completely unstreaked - like a potential American race hen harrier (hudsonius, known as marsh hawk or now more commonly as Northern harrier) a difficult bird to i/d with records generally needing good photo evidence to show the relevant features that separate it from the many orange juvenile cyaneus hen harriers encountered.

My camera was in the car.

It was a fair walk to retrieve the camera but eventually I returned and settled down to wait. Luckily the harrier came back several times and performed well allowing me to get a good selection of shots. When I checked them however I wasn't sure the amount of barring on the underside of the primaries (4-5 on the longest ones) was sufficient for a Northern as I remembered 5-7 was the key number with P10 usually showing 4 or 5 not 3 and a bit as in my bird. In the mean time I got as many shots as I could as the bird passed back and forth causing panic amongst the thousands of wigeon and teal present.

So, could this bird be a Northern harrier? Back home I checked some references, looked at my notes on the 2 'marsh hawks' I saw in 2010 (in Wexford and Norfolk) and perused the photos more closely. Pro-hudsonius features include the well demarcated dark ear-coverts and hood and bright orange underparts (lacking streaking even on the undertail and present only very faintly on the flanks), fairly uniform in intensity but perhaps a little paler around the thighs and also extending across the underwing coverts; the bright coppery orange upperwing covert fringing (though appears more buff in some photos); the narrow middle bar across the secondaries apparent on this bird is also seen as a good pro-hudsonius feature. There were 4 - 5 bars on the underside of each of P7-9 putting the bird at the very edge of the range of hudsonius but fairly typical of cyaneus (a classic huds would show perhaps one extra bar on each). Other concerning features include the rather broad rufous eye crescents just about joined at the rear though their width and degree of connectivity behind the eye seems to vary with the angle of observation (should be narrow, whitish and clearly broken in typical hudsonius). The upperparts though dark do not appear as dark as in many classic hudsonius though as with most features this seemed to vary with the light. The bird was probably a female (later confirmed, see third party comments below) with the appearance of a dull yellow-brown iris but I knew I didn't have a full grasp of the variability of features between the sexes in juveniles. A number of features of this taxon are known to be more reliable in one sex than the other (Hough, 2011).

It is clear there are no 'magic feathers' for this form with no single diagnostic feature able to clinch a positive i/d and the criteria needed seems very much a 'work in progress' with both forms showing a range of variability with much overlap. 'Orange' cyaneus birds seem rather frequent and clearly cause a bit of a headache (though unlike this bird they usually have some streaking obvious). A combination of features need to be present and I was initially worried about the head pattern around the eye whilst the amount of primary barring seemed to me more in favour of cyaneus. Time to get in the experts, so I sought further opinion on these pics from several individuals with considerably more relevant experience than I have!

My intial thoughts were that this bird falls within the range of variation of both forms. It is not the first such occurrence and certainly won't be the last! Identification of orange ringtail harriers has never been more of a challenge with the issue of hybrid hen x pallid an increasing problem in the mix (one of these was present at nearby Ouse Fen fairly recently). Apparent 'good' Northern harrier records are on the rise this side of the Atlantic and remarkably this taxon now appears annually at Tacumshin, Co Wexford (all juvs too so not even returning birds) though there are still only three accepted British records. I do think however there are too many unanswered questions regarding this form which does not appear as diagnosable in the field as once thought. It appears that advanced level identification of many birds across the board tends to lead to muddy waters and more unanswered questions just as we think we're getting to grips with them. Perhaps it's better just to kick back and enjoy the bird rather than get bogged down in the minutiae of exactly what form it is and where it's from!

Update: Or perhaps not? I've just received the following very positive comments from Killian Mullarney:

Having now had a little more time to examine them a number of points are clear.
It is a juvenile, and with such a dark eye, definitely a female.
It has just five bars on the longer outer primaries (which is common in both hudsonius and cyaneus).
It has a faint hint of a fourth  bar on the outermost primary, just outside the dark-tipped under primary coverts. This is not so exceptional in juv male cyaneus, but it may exceptional in juv female. It is typical for hudsonius to have four bars on p10, the innermost sometimes faint, as in your bird.
I can discern no streaking whatsoever on the breast, flanks or underparts. While it is possible that even closer-range images will reveal some very fine streaking, such a complete absence of visible streaking at the range of your observations appears to be outside the variation seen in cyaneus.
The neck (collar) is certainly darker and more heavily streaked than the vast majority of cyaneus but appears to be perfect for hudsonius.
The overall darkness of the head and neck, contrasting strongly with the bright rufous-buff and essentially unmarked underparts is typical of hudsonius. It does not look quite as dark as in many hudsonius, but I'm not aware of any evidence that cyaneus can appear as dark, and have as subdued streaking as in your bird.
The apparently rather dark upperwing coverts and very distinct and extensive pale cinnamon markings on the lesser and some median coverts appears to match what is much more usual in hudsonius than in cyaneus; I'm not sure if cyaneus are ever as well marked as this, but I don't think I have seen one with such prominent, extensive and bright patches on the upperwing.
The clear-cut and very dark centres to the axillaries and greater underwing coverts contrasting with the paler rufous-buff and unmarked median and lesser underwing coverts is a strong pointer to hudsonius.
One point that may not be of any significance, but in most hudsonius the bright apricot-buff colour of the underparts looks 'smooth' and even; in your bird there is a slight blotchiness that I cannot recall having seen in the any images of hudsonius I have studied.
There may be some other points that will occur to me later, but I think the combination of features shown by your bird is strongly indicative of it being another Northern Harrier.
Why are we suddenly recording this previously ultra-rare taxon so frequently over the past four years or so, in Ireland and Britain? I really don't know. Obviously there are concerns that the established identification criteria may underestimate the potential for some cyaneus to match the appearance of hudsonius much more than is currently known, but this seems unlikely. There are also concerns that the reason we are suddenly seeing so many hudsonius is that a number of birds that made the transatlantic crossing a few years ago are now breeding (with cyaneus?) on this side of the Atlantic. I guess it is possible, but so far there is no evidence for this, and one might expect there to be at least as many indeterminate or intermediate-looking birds as perfect-looking examples?

And from Julian Hough:
If the bird is dark-eyed as it looks in the image, and it is a female, it would dispel my concerns about the lack of streaks on the upperbreast and flanks and the middle secondary bar. Lack of streaks is good for female hudsonius. Female hudsonius also have a tendency to show more uniform vents - males look uniform at a distance but close up they have thinner, slightly darker shaft streaks. Females also tend to show fewer bars on the primaries than males in a small sample size i looked at from birds banded at Cape May. Also, they tended to show a thicker middle secondary bar compared to males, which again fits with your bird, as does the darker greater underwing coverts. 

One picture on the ground seems to show a rather saturated and warm bird!

These ones actually convince me a bit more than the close-ups! The blurry topside shot clearly shows what appears to be, if I saw it here, a typical Northern Harrier! 

I just saw Killian's email on the crops and his notes about the eye color which he believes shows a female too. In the field, a bulky bird with broad wings would back this up, as opposed to a smallish, narrow-winged male.

I feel weirdly cautious but can't point to anything that doesn't fit Northern based on the images which build a better composite of the bird. 


Nils Van Duivendijk: Advanced Bird ID Handbook (2011)
Hough, J: The 'Marsh Hawk' Conundrum
Martin, J.P: Northern Harrier on Scilly: New to Britain
Wallace, D.I.M: American Marsh Hawk in Norfolk


  1. Impressive photos James. What lens were you using?

  2. Seymour it.
    Canon 100-400mm IS zoom for the flight shots. Others digiscoped with Fujifilm Finepix JV through a Leica APO Televid 65.