A member of the Records Committee and Taxonomic Sub-Committee of the British Ornithologists' Union published a peer-reviewed paper today that calls into question a key piece of evidence regarding David Luneau's famous four-second video and supports David Sibley's critical interpretations published in March 2006.
"There appears to be no reason to question the anecdotal sight records of Ivory-billed Woodpecker presented in Fitzpatrick et al (or in many online sources), because some of them appear credible, albeit brief," writes J. Martin Collinson in the open-access journal BMC Biology. "However, to regard the Luneau video by itself as presenting anything other than an unidentified woodpecker falls below the standards of proof normally required for scientific publication: the images are not good enough."
Collinson, a senior lecturer and research group leader at the Institute of Medical Sciences at the University of Aberdeen, a co-author of the BOU's taxonomic recommendations, and a self-described "universal data skeptic," compared Luneau's images with new video that shows Pileated Woodpeckers making four escape flights similar to that made by the bird recorded in Arkansas. The new video was shot by David Nolin.
His conclusion: critical frames used for identification of the Luneau video woodpecker as an Ivory-billed Woodpecker are consistent with Pileated Woodpecker, and the wingbeat frequency of the bird in the Luneau video may also be consistent with Pileated Woodpecker, at least for short periods of flight.
"Analysis of the videos of Pileated Woodpecker has supported the hypothesised interpretations of key frames of the Luneau video by Sibley et al," Collinson writes. "Although the rebuttal of that comment in Fitzpatrick et al asserted that flexion and motion of wings of Pileated Woodpeckers could not produce the images seen in the Luneau video, it has been shown here that they can."
"The Ivory-billed Woodpecker may persist in continental North America, and there is enough anecdotal evidence to make this a possibility," he writes, "but the Luneau video does not support the case. The balance of evidence would suggest that the bird in the Luneau video is more likely to have been a Pileated Woodpecker, but the search for Ivory-billed Woodpecker should continue."
Asked about Collinson's contribution to the debate about the Luneau video, woodpecker expert and Ivory-bill skeptic Jerry Jackson said: "The Collinson analysis provides further insight and should put to rest claims of 'confirmation' of the continued existence of Ivory-bills. Of course, it will not. Science is a human endeavor that thrives and advances through such debate.... It is now time for an admission that proof is not in hand -- time for recognition of how much we don’t know about Pileated Woodpeckers."
I was in the audience at the 2006 AOU meeting at which Cornell Lab director John Fitzpatrick played a 1935 audio recording of an Ivory-bill beating its wings at a rate of 8.6 wingbeats per second. The bird in the Luneau video, he said, beat its wings 8.7 times a second. And both wingbeat rates, he said, were faster than the wingbeat rate to be expected from a Pileated.
Collinson disputes this. He reports that the mean frequency value for the first four wingbeats made by the Pileated Woodpeckers he studied were 7.1, 6.7, 8.6, and 8.0 wingbeats per second.
"The fact that in only four recorded escape flights of Pileated Woodpecker, two were recorded for which the initial escape flight wingbeat frequency (8.0 s–1 and 8.6 s–1) exceeded that previously recorded for this species shows that previous datasets were too limited to make this conclusion."
Asked about this, Fitzpatrick told Associate Editor Matt Mendenhall he didn't disagree that the birds in the Nolin video show approximately the same wingbeat rate as the bird in the Luneau video. "For the first few flaps upon launch, the wingbeat is very rapid, but the speed declines almost immediately. The wings beat deeply, sometimes described as a floppy flight, to the point where they almost meet below the bird’s body. You see that well in the Nolin videos. The Nolin videos are characteristic of other Pileated videos, and, frankly, of every Pileated Woodpecker I’ve ever seen."
"The Nolin video supports our point that the Luneau video is not consistent with Pileated," Fitzpatrick said. "The bird in the Luneau video maintains a high wingbeat rate that is quite different from a typical Pileated. So, not surprisingly, we’ve reached the opposite conclusion from Collinson."
Collinson also asserts that as Pileateds fly away from the camera, their plumage is hard to distinguish from that of the Ivory-bill. He suggests that the Pileated's distinctive black trailing wing edges can be spotted in the Luneau video as the wings stroke downward. Previous analysis suggested these were the black wingtips of the Ivory-bill.
"The three plumage features described in Sibley et al that are incompatible with Ivory-billed Woodpecker (black secondary feathers on upper surface of left wing, brighter white primary bases, and a black band curling round the wing tip) are seen consistently in the Luneau video and are recapitulated throughout the video of Pileated Woodpecker," Collinson writes.
Asked about plumage, Fitzpatrick said: "Our view is that he should have done his analysis by deinterlacing the videos.... We have deinterlaced the Nolin videos, and of course, the birds are very consistent with Pileated characters. You see the black trailing edge of the wings, black on the dorsal surfaces. This is not true of the Luneau video. In fact, the Nolin video strengthens our conviction that the Luneau video is inconsistent with Pileated Woodpecker. " -- C.H. and M.M.