I'm currently wrestling with how to report Rufous/Allen's-type hummingbirds in my eBird lists. I realize that the large majority, and possibly all, of those hummingbirds I'm seeing in December in Carpinteria are Allen's, but I'd prefer something less probability-based if I'm going to report them as "Allen's".
Paul Lehman in Birds of Santa Barbara County refers to the difficulty of distinguishing female and immature birds in the field, while implying that adult males *can* be distinguished. My understanding is that a top-to-bottom rufous back can be used to reliably identify an adult male Rufous. But for the partially-green-backed adult males, I'm troubled by Sibley's note that among Rufous Hummingbird adult males, "up to 5 percent show mostly green back". If that's correct, then it would seem to mean that any given adult male "Allen's" hummingbird could in fact be a rare (doubly rare, if outside the Rufous migration dates) green-backed Rufous.
Maybe I'm overthinking this. But here are the possible approaches I'm trying to sort out:
Approach 1 (strict): Report all birds at all times as "Rufous/Allen's", *except* adult males with solid rufous backs, which can be reported as "Rufous". The downside to this is I never actually get to report a species identification of "Allen's".
Approach 2 (medium strictness): For times of the year when Rufous is rare, report non-adult-male birds as "Rufous/Allen's", solid rufous backed males as "Rufous", partially green-backed adult males as "Allen's". Upside: I get to report some "Allen's", at least outside the peak Rufous migration times. Downside: I need to do further analysis to decide what cutoff dates to use.
Approach 3 (least strictness): Same as Approach 2, except I go ahead and apply some sort of probabilistic weighting for the non-adult-males, too, such that at peak Rufous migration times (mid-March to mid-April, and again July to August, maybe?) I report females and immatures as "Rufous", while reporting them as "Allen's" the rest of the year.
I'm currently vacillating between Approaches 1 and 2. I'd be grateful for any insights others are willing to offer. Thanks!