Which accipiter is this?

Lately I've been enjoying blogging here about ID questions, like with the redpolls recently. However, I don't consider myself an ace birder who needs to show off his identification skills - God no. These ruminations are as instructive to myself as I hope they may be to some of my readers. I certainly learned stuff about redpoll identification that I didn't know before.

North American raptor identification is perhaps different, because I am supposed to be somewhat of an expert on that. After all, I have been hired to count migrating raptors for six consecutive migration seasons, all over North America. If I didn't know a thing or two about raptor identification by now, I'd better move on to something else. But after six field seasons I still feel I am learning about raptors, although I've probably mastered the basics.

And these days there are many people who know their hawks. It's never been easier to get proficient in raptor identification, for it has come a long way in North America. Books like Hawks in flight (Dunne, Sibley & Sutton 1988) and more recently Hawks from every angle (Liguori 2005) have helped hawk watchers across the continent tremendously. Of course, nothing substitutes for sheer field experience.

The classic ID challenge remains between that of separating female Sharp-shinned Hawks from male Cooper's Hawks. In the days of the first Peterson Guide, this was thought to be practically impossible in the field, and really the best way to do it was to shoot the bird. We now know how to go about separating those two species, even under difficult viewing conditions, without having to shoot them. We tend to think that separating Cooper's Hawk from Northern Goshawk is easier - and usually it is.

Let's move on to the problem at hand. It probably goes without saying that the bird pictured above is a juvenile accipiter. But which one? Photo taken on March 31, 2009 at Whitefish Point, northern Michigan.

It may be difficult to get a sense of the relative size of this bird, just on the basis of this picture alone. In the field, such things are much more apparent, for example by wing action. This bird was close, and it was large. Not a sharpshin then, if that was even a consideration.

So we've narrowed this down to immature Cooper's Hawk vs immature Northern Goshawk. That was relatively easy; from this point forward, things become considerably more difficult!

Looking at plumage, we notice that the streaking on the body is rather fine, and doesn't extend all the way down. The undertail coverts appear unmarked. Both these things are good for Cooper's Hawk, not for the typical goshawk.

What about the tail? Relatively thin, it seems - another point for Cooper's Hawk. Tip not very rounded and without a broad white tip. But remember: this is a spring individual, with a worn tail. Most accipiter tails look like this in spring. Tail shape and presence/absence of a terminal white tail band is a more useful field mark on fresh tails in fall. So this tail certainly could be a Cooper's tail, despite the fact that it's not round and doesn't have a white terminal band.

What about the 'wavy pattern' on tail bands? This is a well-known diagnostic trait of immature goshawks. It would have been easier to judge on an open tail, but click on the photo and see if you can spot some wavy patterning on the tail. I can. Let's chalk it up as a tentative plus for gos, while acknowledging that we can't see it well enough to be absolutely certain.

One field mark for separating gos from coop, mentioned in Dunne, Sibley & Sutton, appears to have lost some currency, for it is not found in any of the later references, like Wheeler (2003) or Liguori (2005): number of tail bands. In classic cocky Pete Dunne style, he says "if you are close enough to see a zigzag pattern, why not just count the tail bands? Goshawks show four, Cooper's only three." I went back and applied this formula to birds photographed for both the Wheeler and Liguori guides, and it works most of the time on uppertails, but practically never on undertails. I don't want to sidetrack too much here, but plate 190 in Wheeler (2003) for example shows an immature Cooper's Hawk upperside with four bands clearly visible. Undertails is a different matter probably, because undertail coverts only partially obscure the inner tail bands, so you're left wondering which ones to count.

Let's also consider head projection. The neck on this bird does not seem particularly long: a point against Cooper's Hawk and in favor of goshawk. Do we see a white supercilium? It looks like there's something there, but it's really hard to judge from this angle.

Below I have more photos of the same bird, but before we look at those, let's try to finish the ID of this bird on the basis of the top photo alone.

Liguori (2005) makes much of 'hand projection': the way the primaries seem to stick out on a partially folded wing. Based on the top photo, hand projection does not seem very pronounced from that angle. Slight plus for coop then.

I think I see a facial disk. Liguori (2005) says about this: "Northern Goshawks also have pale auriculars that form a fairly defined facial disk." About the pale supercilium he says that "although juvenile Sharp-shinned and Cooper's Hawks can exhibit a pale superciliary, the superciliary of juvenile goshawks is always prominent." The pale supercilium (or superciliary, same thing) appears to be there but is hard to see from this angle. The facial disk is better visible.

Summarizing what we have so far, and feel free to disagree with me:

• unstreaked undertail coverts: better for coop
  
• long, thin tail: better for coop
  
• wavy pattern on tail bands: better for gos (but difficult to judge in photo)
• head projection: better for gos (because limited)
• hand projection: better for coop
  
• supercilium: better for gos (but difficult to judge in photo)
• facial disk: better for gos

As you can see, my scale is ever so slightly tipping over towards gos, and that's what I called it in the field (and also on WPBO's hawk blog).

Now let's look at those other photos, and see if we can finish the identification. Note that those aren't half as good as the top photo, but they are all the same individual. Click on them for larger views.

Here's the bird coming up to the hawk platform. Chesty? Yes, I think so. Long projection of hand? In this photo: yes, actually. White supercilium? Yup. All three field marks are good for goshawk.

Positively pot-bellied, I'd say. White supercilium and facial disk also clearly visible in this shot.

This angle provides another opportunity for judging head projection, tail length and width, hand projection, and body width.

And finally, a shot of the bird flying away. Note the mottled upperparts, and the four tail bands.

New Jersey Audubon sells bumper stickers with the phrase "What Would Pete Dunne Do?" Well, in this case I'm sure he would count the tail bands.

Would he call it a Northern Goshawk? I don't know, but I do. What do you think?

References
Dunne, P., Sibley, D. & Sutton, C. (1988) Hawks in flight. Houghton Mifflin, Boston.
Liguori, J. (2005) Hawks from every angle. Princeton University Press. Princeton.
Wheeler, B.K. (2003) Raptors of Eastern North America. Princeton University Press, Princeton.