The above still, from a video shot in the Pearl River Basin in Louisiana on February 20, 2006, is the first image of a perched Ivory-billed Woodpecker to be obtained in decades. While paddling upstream in a kayak at 7:25 that morning, I flushed an ivorybill that was perched less than 20 meters away on the side of a broken-off tree on the bank. The bird flew to the left, and high-pitched calls immediately started coming from that direction. I had heard the same calls during an encounter with two ivorybills in the same area two days earlier. I turned on the video camera, kept it aimed at the source of the calls, and captured the video sequences that are discussed below. This panorama shows the tree where the ivorybill was initially perched (to the right) and where it was briefly visible through a gap in the vegetation (to the left). This panorama shows the rest of the scene all the way to a fork in a tree deep in the woods (to the left), where the bird appeared more than 10 minutes after it flushed and 218 meters from where it was initially perched. A reference tree is marked in both panoramas in order to show where they meet.

Crest size and head shape. I noticed movement in a tree deep in the woods at the 10:09 mark in the video. I zoomed in with the camera using my right hand and tried to spot the bird through binoculars using my left hand. Despite never finding the bird, I managed to obtain some useful footage during the next minute and a half. The image below shows the bird perched on the right side of a branch at the 10:19 mark, just after I got the camera on it. A few seconds later, it rotated behind the branch and hid. Then it climbed a little higher and rotated back into view. The view is at an oblique angle in which the back of the bird is rotated away from the camera as illustrated in the artwork by Michael DiGiorgio. The volume of the crest and the rounded head shape are consistent with the ivorybill specimen but are not consistent with pileateds (many photos are available for comparison here).



Dorsal stripe and raised crest. Since the video was obtained early on an overcast morning, it's difficult to see field marks in raw images of the bird silhouetted against the sky. After adjusting the brightness, a feature consistent with the left dorsal stripe is visible in the top left image (this feature is difficult to see on some monitors). The similar feature to the left of the dorsal stripe is a broken branch as illustrated in the artwork by Michael DiGiorgio. The faint light streak across the head is a small sprig of vegetation. It's clear from the image on the lower left (from about one second earlier in the video) that there is no vegetation in the position where the dorsal stripe feature appears when the bird moves into view. In contrast to the image appearing in the "crest size and head shape" section above, the crest appears to be raised in the image in the lower right as illustrated in the artwork.



Posture and profile. The images below compare the postures and profiles of an ivorybill (photo by Tanner), a pileated, and the bird in the video at the 10:27 mark. Relative to the pileated, the ivorybill has a leaned-back posture and a long and thin neck. Roll the mouse over the image on the right to compare the bird in the video and the ivorybill, which have similar postures and profiles. Images from the video have been posted for more than two years, but nobody has come forward with a photo of a pileated that comes close to fitting the posture and profile of the bird in the video. Many pileated images can be accessed here for comparison. Out of a small set of ivorybill photos, there exists a good match. Out of a large set of pileated photos, none has been found that comes close to matching.



Comparison of bills. When I obtained the video, I was sitting in the kayak, which was backed into the opposite bank at a known position. Using a laser rangefinder that is accurate to within a half meter, I determined that the distance from the camera to the point where the bird was perched is 128 meters. From a known distance, I photographed these casts of ivorybill and pileated bills that were made from museum specimens. Scaling this photo to account for distance and overlaying, I obtained the bill comparison in the photo below. The pileated cast appears to be smaller than the bill of the bird in the video. The ivorybill cast appears to be a good fit.



Physiology and low-speed flight characteristics. Although the pileated and the ivorybill have a superficial resemblance, these woodpeckers actually have major physiological differences as illustrated in the idealized sketches below. The pileated (top sketch) is relatively light and has relatively broad wings. These characteristics are suitable for a bird that remains within a relatively small territory. The ivorybill (bottom sketch) is about twice as massive as the relatively small subspecies of pileated within its range and has relatively narrow wings. These characteristics are suitable for a bird that needs power to access beetle larva under tight bark and the ability to make rapid flights over long distances to seek out this food source, but the ivorybill pays a price for these capabilities. As illustrated in the sections that follow, this species requires deep and rapid flaps during takeoffs and short flights, which are easy for pileateds.



Short flight. The bird flaps once during a short flight across the fork at the 10:35 mark in this movie (which plays at half speed). Individual frames of the video appear below. The wings appear to be narrow, and the flap rate is about 7.5 Hz. The deep and rapid flap is consistent with the physiology of an ivorybill (a massive bird with narrow wings) but dramatically different from the flaps of the pileateds taking similar short flights in these videos:

example 1 (half speed)
example 2 (full speed)
example 3 (full speed)
example 4 (full speed)



Take off in level flight. At the 11:30 mark, the bird takes off in level flight with deep and rapid flaps in this movie (which plays at half speed). There are similar flaps in the takeoff from behind a tree in the 2007 Choctawhatchee video. It's interesting to compare with the pileated flights in this movie (which plays at half speed). I spent many months intensively observing pileateds in the Pearl and never noticed the deep and rapid flaps that I have observed several times in ivorybills. There is a need for an extensive study of pileated flap rates in level flight. I took several samples, and the highest flap rate that I measured was 4.4 Hz. Cornell took many samples and obtained similar results. If 7.5 Hz is ever achieved in level flight, it's hard to explain why events in the 5 to 6 Hz range are not readily observed. As an alternative to deep and rapid flaps, an ivorybill can take off by using gravity to accelerate. A bird in the 2007 Choctawhatchee video drops from the canopy and begins to level off after reaching a high speed.

Recording of a possible alarm call. I heard high-pitched calls during two encounters with ivorybills and recorded them during the second encounter. On February 18, I heard kents coming from an ivorybill behind a fallen tree on the bank. I quietly paddled up to the edge of the bank, but the ivorybill evidently didn't detect me and kept calling. Perhaps a minute later, a second ivorybill started calling from the opposite bank. Then it apparently noticed me near the first bird and gave several harsh scolding calls. The high-pitched calls then started coming from the direction of the second bird, and the first bird went silent. On February 20, the same calls started coming from the direction of an ivorybill right after it flushed. This audio recording is from the first two minutes of the video. The high-pitched calls sound similar to the Blue Jay bell call but don't match published sonograms of bell calls. In the sonograms of known and putative kent calls, all of the frequencies appear to turn on and off in synch. None of the bell calls published in the Auk by Kramer and Thompson in 1979 have this property, but the high-pitched calls do have this property (even though they sound very different from kents). During both encounters, the calls started at key moments and came from the direction of an ivorybill. On February 20, the source of the calls tracked an ivorybill as it moved through more than 200 meters of dense habitat. I wouldn't have obtained any video footage if the calls hadn't allowed me to keep track of the movements of the bird. Tanner described a high-pitched call that is given when ivorybills are disturbed, which is consistent with the observations. There were sixteen calls during the first two minutes after the camera was turned on. While drifting downstream in the kayak, I kept the camera aimed at the source of the calls. After the calls stopped, I backed the kayak into the opposite bank and kept watch. Minutes later, there were four more calls from deeper in the woods, and then the ivorybill appeared perched in that area.

Through the gap. The high-pitched calls were recorded at 1, 6, and 15 seconds into the video. The calls then stopped for 30 seconds as I drifted near the spot where the bird was hiding a short distance back in the woods. The next calls were given at 45, 52, 56, 67, 82, 85, 86, 90, 92, 100, 104, 110, and 120 seconds. The bird appears briefly through a gap in the vegetation at the 34 second mark (nearly in the middle of the 30 second gap between calls). This footage is shown below as a series of enlarged and cropped frames, which are separated by 1/60 of a second. The vegetation has been stripped away in the sketches below by Michael DiGiorgio, which illustrate my interpretation of frames 01, 08, and 11. Roll the mouse over the sketches to see comparisons with the video frames. In frame 01, the bird is perched in the gap facing away from the camera. The triangular white patch on the folded wings is a key field mark of the ivorybill. In frame 08, the underside of the right wing has a thin black strip up the middle. Although this feature is very different from the underwing pattern illustrated in the National Geographic field guide, it is similar to the underwing pattern appearing in a recently released Tanner photo. In frame 11, the white trailing edge of the top of the right wing appears to be visible near where the wing attaches to the body.





Wing aspect ratio. According to data that Cornell obtained in Arkansas, wing aspect ratio ranges between 1.4 and 1.9 for pileateds. This quantity is 2.4 for an ivorybill that was photographed in the Singer Tract in 1935. Bob Brown made an aluminum model wing with an aspect ratio of 2.4, which I bent into the shape of the right wing in frame 08 of the "through the gap" sequence. The red lines on the image on the right correspond to aspect ratios of 1.4 and 1.9. Roll the mouse over the image below for a comparison between the model wing and the right wing in frame 08 of the "through the gap" sequence. These comparisons suggest that the wing of the bird in the video has an aspect ratio of about 2.4. The right wing in frame 08 appears to be bowed. The ivorybill has several characteristics that favor wing bowing. Narrow wings are more flexible than broad wings. High frequency (flap rate) and high amplitude (deep flaps) correspond to high velocity and acceleration, which produce the forces that make wings bend. These forces also depend on the mass of the bird, and the ivorybill is a relatively massive bird.



Wariness. While perched in the fork, the bird can be seen hiding behind the branches and repeatedly peeking around in my direction. It appeared to attempt to get a better look in my direction by briefly rotating into view, hitching up to a higher vantage point, and flying across the fork. This extremely shy behavior is not consistent with pileateds, which are relatively tame in the Pearl. This bird was deep within the cover of the woods. The photo below shows the view from the base of the tree in which it was perched, looking in the direction where I was sitting in the kayak that morning.



Unusual movements. The bird rotates into view like a door on a hinge and then hitches up and behind the branch with a springing motion in this movie (which plays at full speed). In the image below, which is from near the end of the clip, the left foot appears to be visible as a light-colored spot against the relatively dark tree (this feature is difficult to see on some monitors). The perched ivorybill in this Tanner photo also appears to have a light-colored foot.



Best sightings. Michael DiGiorgio has painted ivorybills into photographs in order to bring to life my four best sightings of the 2006 search season. The February 2 sighting was about a half mile from the hot zone, where I had most of the other sightings and obtained the video. The bird flushed from near the bank and flew into the woods. I got my binoculars on it for a few seconds, clearly saw the white trailing edge of the right wing, and noticed the right dorsal stripe and all-dark head. The photo of the site was taken immediately after the sighting.



The February 16 sighting was the first one in the hot zone. I was quietly drifting downstream and came upon an ivorybill that flushed from close range on the bank. This was the most spectacular sighting. The light conditions were excellent, and I saw the brilliant white trailing edges that nearly met in the middle as the bird flew directly away and into the woods on rapidly beating wings. Late that afternoon, I staked out the area and heard three kents that provoked a stern territorial call from a pileated. Witnessing that interaction between the species was every bit as exciting as the sighting. The photos of this site and the other two sites that follow were not taken on the same days as the sightings.



The second of the three sightings on February 17 was nearly as spectacular as the sighting of the previous day. The light conditions were excellent, and the bird flew low across the water directly ahead as I drifted downstream. I wonder how many times an ivorybill went unnoticed as it flew behind me. The dorsal side of the bird was slightly tilted in my direction, which gave me a wonderful view of both wings.



While paddling upstream on February 20, I came upon an ivorybill perched on the side of a snag on the bank. In the poor light on that overcast morning, I was initially unable to make out field marks, but I instantly knew it was an ivorybill based on the posture and movements. I saw the white trailing edges as the bird flew to the left.