I think Bob Boekelheide gives good answer but it's probably best to consider counting the obviously GW (wingtips same medium grey as mantle) and obviously WG (dark grey mantle and black tips) and then count GWxW intergrades (with non-matching dark grey to blackish tips) as all the rest.

There is a paper that gave the percentage of GWxW intergrades to GW as 38% (from a sample of 54 birds!) in the Puget Sound. See table 1 in:

DOUGLAS A. BELL
The Condor 98: 527-546 (1996)
Genetic Differentiation, Geographic Variation and Hybridization in Gulls of the Larus Glaucescens-Occidentalis Complex

Bell finds wingtip color and mantle color to be (statistically) the best features to discriminate intergrades from the two "pure" species.

I note Sibley uses the same approach for differentiating hybrids from in 2nd edition on p222 "Identification of Hybrid Gulls". Compare his diagnostic callouts with p213 and p215 for the GW and WG to the hybrids on p222 (and I think he did this in the 1st edition too, IIRC).

I think that's what eBird (and the CBC too) have settled on too.

How well birders can do it is another matter, of course. In writing up the counts for CBC data I see people have different biases for this count e.g. I tend to count our local gulls with "too many hybrids" but other tend to count only GW gulls (and no hybrids).

It's easier to make this judgement for a standing bird. In flight it's a bit more tricky.

On Jan 7, 2016, at 3:14 PM, B Boekelheide wrote:

> Hi, Larry ? OK, I?ll take the bait.
>
> Unfortunately, the answer to your question is: You cannot possibly know if you are basing your identification on phenotypes. Here at the Sequim-Dungeness CBC, our 10-year average number of pure Glaucous-winged Gulls is 579 and the 10-year average number of Olympic Gulls is 2201. Does this mean that we should expect about 20.8% (579/2780) of these large, pink-legged gulls around here to be pure Glaucous-winged Gulls? Unfortunately, no...
>
> Since we live in the middle of the Olympic Gull hybrid swarm, in which obvious hybrids have backcrossed over many generations with supposedly pure individuals, even those birds that look like pure Glaucous-winged Gulls may not be pure Glaucous-wingeds. The phenotypes of hybrids are all over the map, through a continuum from relatively pure Glaucous-winged to relatively pure Western, so how do you know?
>
> Unfortunately it is likely a losing battle to look through a group of 500 large pink-legged gulls around here and come up with a firm number of pure Glaucous-wingeds, Westerns, or hybrids, but we valiant birders try to do it all the time. What field marks do you use? How gray should the wingtips be? How light should the mantles be? How much head streaking should there be? Sorry! Mix in winter populations that came from Alaska to California and it becomes even more complicated.
>
> As birders we want so much to pigeon-hole every bird into a nice species/hybrid, and the experts can pass judgement on which species hybridized with which species to create specific-looking gulls, but is it meaningful when you?re looking through hundreds of these gulls, all of which have variable phenotypes (and genotypes)?
>
> So when eBird expects us to choose between four choices for the great majority of our lovely large pink-legged gulls (i.e. Glaucous-winged Gull, Western Gull, Western X Glaucous-winged Gull, and Western/Glaucous-winged Gull), how do we choose? It is usually possible to separate out relatively pure Western Gulls. But when I look through several hundred large pink-legged gulls in Dungeness Bay, all standing in different postures and frequently blocking one another, to definitively designate each one as either a pure Glaucous-winged or a hybrid Olympic Gull seems to be a fools task, whether I use phenotypic field marks or not. I suggest there should be a different category that those of us around here use most of the time with large groups of these gulls, like "Glaucous-winged and/or Olympic Gull.? That is a different, more specific category than Western/Glaucous-winged Gull, but broader than Western X Glaucous-winged Gull. Calling them gull sp. or Larus sp. is not good, because that lumps in several other species they clearly are not.
>
> Or maybe we just call them large pink-legged gulls.
>
> Nevertheless, they are fascinating birds that deserve our respect, regardless of their parentage.
>
> Bob Boekelheide
> Dungeness
>
>
> From: Larry Schwitters <leschwitters at me.com>
> Subject: [Tweeters] Glaucous-w. X Western Gull (hybrid)
> Date: January 6, 2016 at 4:17:28 PM PST
> To: Tweeters <Tweeters at u.washington.edu>
>
>
> Anyone care to offer an educated guess as what the percentage of Glaucous-winged X Western Gull (Olympic Gull) would be expected in with 500 Glaucous-winged at the Marysville sewage lagoon this time of year?
>
> Larry Schwitters
> Issaquah
>
> _______________________________________________
> Tweeters mailing list
> Tweeters at u.washington.edu
> http://mailman1.u.washington.edu/mailman/listinfo/tweeters

--
Kevin Purcell (Capitol Hill, Seattle, WA)
kevinpurcell at pobox.com | at kevinpurcell
http://flic.kr/kevin_g_purcell