Dispatches from the front lines of the GBBC
February 18, 2013
My Dearest Wife,
We crested the hill today to find the full strength of the Backyard Bird Counters allied against us. Their numbers were uncountable. Their energy boundless. The bird records came at us full bore and I watched many of my fellow reviewers – good men, family men – fall by the wayside. They weren’t ready. I wasn’t ready. The horrors were beyond count. I saw species never before recorded in North America reported from my county. I saw erroneous Red-headed Woodpecker records piled like so much cord wood. In the end it didn’t matter it if was a Purple Finch or a House Finch, they went into the checklist with seemingly no consideration as to which side of the line they were on, both species accounts sacrificed on the altar of citizen science. I shant know if I will ever be the same again after this bloody campaign, but I trust the powers that be know that this is all for the best. All I know is that too many good bird accounts have been soiled today. It seems as though this battle will not be over as quickly as we had hoped. The records continue to come in, less continually than before but we stand before them broken with scarcely the willpower to determine their validity before we, too, will lay down before the wave and in doing so be consumed by it.
Send my love to the children. I pray that they will recognize their father when I return.
I have long known about the “holes” in eBird. The species pairs that result in one, less common, species being reported erroneously in inflated numbers because any attempt to tighten the filters to catch it would result in a deluge of legitimate records getting hung up and turning into a pain in my Indian Wild Ass. I was always very aware of two pairs: Purple/House Finch and Sharp-shinned/Cooper’s Hawk. The former, in both cases, are uncommon but not rare and often mistaken for the latter. I assumed this was happening on a small scale, but I didn’t worry too much about it as I trust the heavy eBird users in the state to get things right. If people report a dozen or so Purple Finches when they were really House Finches the signal is going to be lost in the noise of legitimate data. It’s not going to matter in the long run.
But this year the Great Backyard Bird Count was run through eBird, and I keep a pretty close tab on my North Carolina eBird. I quickly realized, upon receiving a Needs Alert for Wake County that was page after page of Purple Finch records, that something was seriously up. A quick jaunt through the data later turned up checklist after checklist of GBBC participants that had included Purple Finch but neglected to record any House Finches. Not that this is completely unheard of, but it’s a giant red flag that something sketchy is going on.
I began sending emails, literally hundreds of them. And with every email, for which eBird helpfully provides a template that every user knows, I included a couple paragraphs on how to separate House and Purple Finch. And you know what? 60% of them, more or less, came back saying they’d made a mistake. That is a completely ridiculous percentage of Purple Finch records that had issues. And I didn’t even consider the ones that had both Purple and House Finches in questionable proportions, because I have a sneaking suspicion that people are counting the red ones as Purple Finches and the brown ones as House Finches.
So it got me thinking about other confusing species. Not just the Sharpie/Coop thing, but ones regular birders might not even think about. I began to look into Red-headed Woodpeckers over the weekend and found that 90% of them were on lists that neglected to include Red-bellied Woodpecker. Once again, emails were dispatched. Once again, an alarmingly high percentage of the responses admitted mistakes.
So I looked at Herring Gull records without Ring-billed Gulls. Loads of them. Common Ravens without American Crows. The same. House Wrens without Carolina Wrens. Great Cormorants without Double-cresteds. None of these are flagged by the filters, nor do they really need to be. And these are just the obvious ones. Who knows what else is out there, just waiting to be discovered? The prospect is terrifying.
Don’t get me wrong, I am one who is almost always willing and able to help out a rookie birder with an ID problem. Many times I actually enjoy it. But this is so overwhelming that I’m having trouble wrapping my head around where to even start here. It is a lot of emails, and they never stop arriving. And while I tried to keep up with it as the records came in they’re still coming in. And now all the new records are all tied up with the ones I’ve reviewed and validated. They may never be sorted out. How could they be?
Unless, of course, eBird allows the capability to search for checklists that contain a chosen species, but are absent another chosen species. That would be great to use year round rather than once a year when the tsunami of data comes in and buries us all.
In any case, I really think I’m starting to hate the GBBC.