According to Tanner, scaling bark was the Ivory-billed Woodpecker’s primary foraging strategy during breeding season in Louisiana. Tanner wrote that the ivorybill is “capable of easily scaling away heavy bark that other woodpeckers could not loosen.” (Tanner 1942). All woodpeckers in genus Campephilus have specific anatomical characteristics that enable them to forage in this specialized way (Bock and Miller 1959). Following Tanner, most post-Singer Tract search efforts have looked for feeding sign as an indicator of presence. Because Tanner’s descriptions are somewhat vague and many of the photographs showing feeding sign are poor, these efforts have tended to focus on decay state and bark adhesion without taking bark characteristics and tree species sufficiently into account. I posit that tree species and bark and wood characteristics are key factors that should be considered. I further posit that extensive bark scaling on live and recently dead hickories (genus Carya) may be beyond the physical capacity of the Pileated Woodpecker.
As all regular readers know, I’ve been somewhat obsessively focused on bark and bark scaling since my earliest years of ivorybill searching. The reason for this interest is simple: it’s how Tanner found ivorybills or inferred their presence when he couldn’t find them (Tanner 1942). Unfortunately, as discussed in a number of posts, Tanner’s descriptions are somewhat opaque, and most of the published images of feeding sign, including those in the monograph, are not very illuminating. Indeed, some of them are consistent with pileated work that we’ve documented. Plate 8, shown below, is a prime example. The caption reads, “Ivory-bill feeding sign on a slender limb”.
Early on in my study of this subject, I hypothesized that certain kinds of bark scaling on hardwoods might be beyond the physical capacity of the Pileated Woodpecker. I still believe this to be true, a view that is supported by what we’ve documented for pileated and by numerous examples of pileated scaling from online sources. At the same time, the details of what types of work might belong in this category have shifted somewhat, especially as it has become clear that Pileated scaling can look like what’s shown in Plate 8 and that pileateds will scale bark from recently dead sweet gums.
This is not to suggest that ivorybills never scale small and medium-sized branches in a similar manner. According to Tanner they did so frequently; however, I have been focused on what may be diagnostic for ivorybill. It seems likely that there is considerable overlap between ivorybill and pileated work when smaller branches are involved (at least on sweet gums).
The sequences we obtained showing pileateds scaling a sweet gum limb have inspired me to look more deeply at the characteristics of hardwood bark and pursue some research avenues that I hadn’t considered previously. I’ve linked to some of the sources in recent posts, but I’ve had some additional insights that seem important enough to share. Every time I think I’ve run out of things to say on the subject, something new crops up.
Like virtually everyone else, I’ve followed Tanner and focused on two bark characteristics, “tightness” and thickness, but it recently struck me that other features might be important as well. And the literature, mostly from the lumber industry, supports this idea.
Tanner suspected that the Singer Tract ivorybills preferred sweet gums and Nuttall’s Oaks because the bark is thinner, and the thinner barked limbs had “more borers” than thick barked ones. While abundance of food was likely a factor, I suspect that, at least with respect to sweet gums and possibly Nuttall’s oaks, ease of scaling and access to food played a role.
It’s important to point out that in live trees, hardwood bark adhesion varies seasonally, with bark becoming tighter during dormant stages and looser (with considerably less variation from species to species) during the growing season. Bark is often if not always tighter on recently dead trees than on live ones (Stokland et al. 2012).
In addition, “The structural and chemical traits of dead wood, inherited from the traits of living trees, are also major drivers of wood decomposition and these traits vary greatly among tree species.” (Cornellisen et al. 2012). The authors of the linked paper point out that other factors, including size and site, can also contribute to the way that bark loosens post-mortem, but specific traits seem to be paramount, especially since the scaling we deem to be suggestive, whether on standing or downed wood, is on trees that are alive or are recently dead. Because the scaling has a very distinctive appearance, we also deem as suggestive hickory snags and stubs that appear to have been scaled some years ago, even if they are in a considerably later stage of decay overall. Bark attached to hard wood on these longer dead stubs and snags often remains tight for 3 or more years after death.
A 1978 report, entitled Bark and Wood Properties of Pulpwood Species as Related to Separation and Segregation of Chip/Bar Mixtures examined bark morphology and strength properties in 42 different pulpwood species and identified factors that impede the mechanical removal of bark from logs. These include: cellular structure, bark adhesion, bark strength, bark toughness, wood toughness, specific gravity/density, and moisture content. (Institute of Paper Chemistry 1978) One caveat about this report: a subsequent paper gives the sample size for each species, and in many cases (including sweet gums) it was only 2 (Einspahr et al. 1982)
It may be counterintuitive, but the authors found that shagbark hickory was far and away the most difficult bark to remove. (The tightly adhering layer is thin, beneath the dead bark that gives the species its shaggy appearance.) One key finding was that:
“Morphologically, the presence of fibers increases inner bark strength and, when sclereids (a type of cell) are present, bark strength is decreased. Inner bark strength, in turn, has a major influence on hardwood wood/bark adhesion. The multiple regression equation employing wood toughness and inner bark strength accounts for 72% of the wood/bark adhesion variation encountered.”
Sclereids are virtually absent in hickories (Nanko 1980) and a few other species that don’t approach the hickories in bark strength and bark and wood toughness (Eastern cottonwoods, yellow poplars, white ashes, and black willows). These tables are particularly illustrative:
Shagbark hickories are the extreme outlier in this study, in terms of adhesion, as well as in terms of inner and outer bark toughness and strength; there are very few shagbarks in our search area, and we have never found scaling on one. I have been unable to find specific information about bitternut hickory bark strength or toughness, but the industry’s debarking problem applies to all species in the genus Carya due to the near absence of sclereids in conjunction with the other factors. Moreover, the industry does not differentiate among hickory species (Timber Mart South 2016). This 1996 paper is worth quoting at length in this regard (full text is not readily accessible):
The amount of published literature dealing with hickory debarking is very limited. Often it is only mentioned as an example of one of the hardest tree species to debark. One study quantified this by measuring the strength of the bark-to-wood bond of 42 hardwood species, including hickory. According to Einspahr et al., the dormant season bark-to-wood adhesion for hickory is greater than 3000 kPa, which is a tenfold difference from the growing season and nearly three times as great as the dormant season wood/bark adhesion for quaking aspen (Populous tremuloides, L.), a species considered to be extremely difficult to debark in the northern United States.
Einspahr et al. also microscopically examined the failure zone in an attempt to correlate morphological differences with bark-to-wood adhesion. For hardwoods in general, they found that during the growing season, failure occurred in the cambium or in the xylem just inside the cambial zone. Conversely, dormant-season failure occurred in the inner bark. They also found that fibers in the bark increased the inner bark strength while sclereids decreased inner bark strength. Hickory bark can contain between 15 to 20 percent fiber and contains less than 0.05 percent sclereids.
While these studies have confirmed that hickory is difficult to debark, they have not addressed possible solutions to the problem. As a result, hickory is often left behind during harvesting, reducing the total usable fiber from a given stand and, over time, increasing the percentage of the species in the hardwood resource, compounding the problem of future harvests.
When a tree dies, the bark eventually loosens and detaches naturally as the cambium decays. After felling, the cambium remains alive until it has consumed all available food or dries out. Moisture loss, while causing cambial death, initially greatly increases the strength of bark attachment because additional bonding between fibers occurs as the secondary valence bonds with water are broken (Belli 1996).
Thus, even though hickory bark adheres less tightly than sweet gum bark during the growing season, it seems likely that it’s harder to scale year round, given its much greater wood and bark strength and toughness. It is also clear from my observations that sweet gum bark loosens far more rapidly than hickory bark post mortem. Note that we have found fresh scaling on both live and recently dead hickories.
Based on specific gravity of the bark – averaging 0.72 for shagbark and 0.60 for bitternut – and bark moisture content – averaging 34% of dry weight for shagbark, and 60% for bitternut – it seems likely that bitternuts are somewhat easier to debark than shagbarks but considerably harder to debark than virtually any other tree species in our search area.
Comparing bitternut hickories to other species, most oaks have a considerably higher average moisture content in their bark (Chestnut and Southern red, including Nuttall’s oaks, are exceptions) and similar specific gravities. Sweet gum bark has an average specific gravity of 0.37 and an average moisture content of 91% of oven dry weight. (Schlaegel and Willson 1983, Miles and Smith 2009). But oaks and sweet gums have sclereids, and sweet gums and all tested oak species score far lower on bark toughness and strength than shagbark and, by inference, bitternut hickories. Sweet gums and the tested oak species are fairly similar in these regards, but I suspect that the higher density and lower moisture content in oak bark makes it harder to scale and may mean that oak bark adheres more tightly than sweet gum bark for a longer period after death.
I posit that when it comes to woodpecker scaling, dormant season bark adhesion, inner and outer bark strength, and inner and outer bark toughness are all relevant factors. We know that Pileated Woodpeckers remove sweet gum bark with some difficulty and that even on medium-sized limbs, they are not consistently able to remove bark cleanly down to the sapwood. It’s also clear that bitternut hickory bark is very difficult to remove, second only to shagbark hickory in our search area. This further reinforces my view that the work on hickories is not the work of Pileated Woodpeckers.
Click here and here for examples of the hickories that are scaled in a manner we hypothesize is diagnostic for Ivory-billed Woodpecker. Also be sure to watch this YouTube video of a Crimson-crested Woodpecker (Campephilus melanoleucus) foraging. (Thanks to Phil Vanbergen for finding the clip and the scaled hickory at the second link.) I’m reposting the link to the video here because I think it very clearly illustrates many of the characteristics we associate with Ivory-billed Woodpecker work on hickories, although the species of tree being fed on is unknown. Note the striking similarity in appearance and also that the work of the substantially smaller billed Crimson-crested is not as clean around the edges as the work we’re ascribing to ivorybills.
There were no bitternut hickories in the Singer Tract, but there were congeners – pecans and water hickories. Tanner observed ivorybills scaling on these species twice and digging once. For pileateds, there were 4 instances of digging and none of scaling, as opposed to 5 scaling and 9 digging on sweet gums. The relative abundance of water hickory and pecan at Singer was 2.7%; approximately 10% of the trees in our search area are hickories, and hickories are second only to sweet gums in terms of the number of scaled trees we’ve found. While Tanner’s is obviously a minuscule data set, it may support the hypothesis that live and recently dead Carya bark is too tough for pileateds to scale extensively, if at all.
There are a number of hardwood species found in potential ivorybill habitat that are somewhere between sweet gums and hickories in terms of how easily scaled they may be and how soon after death bark decay and loosening set in – eastern cottonwoods, black willows, water tupelos, some oak species, red maples, green ashes, honey locusts, persimmons, and elms – in these species, it seems likely that close examination of the scaling and bark chips can provide some clues.
Previous Ivory-billed Woodpecker searches have focused on bark adhesion and state of decay when considering scaling as possible foraging sign. Bark morphology, dormant season adhesion, inner and bark outer strength, and inner and outer bark toughness, and wood toughness are all relevant to the ease with which bark can be scaled from live and recently dead hardwoods. Specific gravity and moisture content are also factors. Bark from trees in the genus Carya is difficult to remove industrially, and members of this genus are likely the most difficult trees to scale throughout the historic range of the ivorybill. Since Pileated Woodpeckers scale sweet gum branches with some difficulty and do not consistently remove bark down to the sapwood, it may be beyond the physical capacity of Pileated Woodpeckers to scale hickories extensively and cleanly, while leaving large pieces of bark behind. Extensive work on hickories that has a distinctive appearance may be diagnostic for ivorybills; this distinctive appearance of this scaling may also be the key to recognizing Ivory-billed Woodpecker foraging sign on other species.
This may be no more than an aside, but it may be a relevant data point. I recently observed a Pileated scaling briefly on a live 14″ DBH Norway maple in my yard near New York City. The photos show that the sap is flowing. The appearance of the scaling is exactly what I’d expect for Pileated, with strips about half an inch across. Norway maple may be a decent stand-in for sweet gum; while its bark has a higher specific gravity, 53 as opposed to 37, the moisture content of the bark is almost identical, 91% as opposed to 90%.
Bock, Walter J. and Waldron Dewitt Miller, The Scansorial Foot of the Woodpeckers, with Comments on the Evolution of Perching and Climbing Feet in Birds, American Museum Novitates, #1931, 1959
Belli, Monique L., Wet storage of hickory pulpwood in the southern United States and its impact on bark removal efficiency, Forest Products Journal. Madison 46.3 (Mar 1996): 75.
Cornelissen, Johannes H.C., Ute Sass-Klaassen, Lourens Poorter, Koert van Geffen, Richard S. P. van Logtestijn,Jurgen van Hal, Leo Goudzwaard, Frank J. Sterck, René K. W. M. Klaassen, Grégoire T. Freschet, Annemieke van der Wal, Henk Eshuis, Juan Zuo, Wietse de Boer, Teun Lamers, Monique Weemstra, Vincent Cretin, Rozan Martin, Jan den Ouden, Matty P. Berg, Rien Aerts, Godefridus M. J. Mohren, and Mariet M. Hefting, Controls on Coarse Wood Decay in Temperate Tree Species: Birth of the LOGLIFE Experiment, Ambio. 2012 Jul; 41(Suppl 3): 231–245.
Einspahr, D.W, R.H VanEperen, M.L. Harder et al. Morphological and bark strength characteristics important to wood/bark adhesion in hardwoods, The Institute of Paper Chemistry, 1982: 339-348.
Institute of Paper Chemistry, Project 3212, Bark and wood properties of pulpwood species as related to separation and segregation of chip/bark mixtures, Report 11, 1978.
Miles, Patrick D. and W. Brad Smith, Specific Gravity and Other Properties of Wood and Bark for 156 Tree Species Found in North America, United States Department of Agriculture, Forest Service. Northern Research Station, Research Note NRS-38, 2009.
Nanko, Hiroki, Bark Structure of Hardwoods Grown on Southern Pine Sites (Renewable Materials Institute series), Syracuse University Press, 1980.
Schlaegel, Bryce E. S and Regan B. Willson, Nuttall Oak Volume and Weight Tables, United States Department of Agriculture, Forest Service. Southern Research Station, Research Paper SO-l 86, 1983
Siry, Jacek, ed., Species Detail Report, Timber Mart-South, 2016
Stokland, Jogeir N., Juha Siitonen, and Bengt Gunnar Jonsson, Biodiversity in Dead Wood, Cambridge University Press, 2012
Tanner, J.T. The Ivory-billed Woodpecker,National Audubon Society, 1942.
Thanks to Fredrik Bryntesson, Steve Pagans, Chris Carlisle, and Bob Ford for their help with this post.
I’m still in mourning and adjusting to the loss of my friend. Thanks to all who have expressed appreciation for our work and a desire for it to continue. I’m sure Frank would have felt the same, and with that in mind, this will be the first of two or three installments discussing Pileated Woodpecker work on sweet gums that we’ve recently documented.
After my December trip, Phil Vanbergen and John Williams retrieved the trail cam we had deployed on December 21. They took the camera to Frank’s house, reviewed the card, and found that two Pileateds had visited the downed tree on the 22nd and had scaled some bark near the base of a medium to large limb. Phil, who has spent time with me in the field and who has paid close attention to my approach to analyzing feeding sign, immediately suspected Pileated for this work, based on the appearance of the scaling and the characteristics of the bark chips.
Rather than extract the images at that time, Frank and Phil opted to redeploy the camera. Although I had not yet seen the frames, many of my last communications with Frank, both on the phone and via email, touched on this subject. He was tickled by the fact that we’d anticipated and documented scaling activity on an untouched limb and was eager to get back out and see for himself. Sadly, that was not to be.
Phil and I retrieved the trail camera on January 28. I had visited the site on the 26th and had noted some additional scaling consistent with what I’d expect for Pileated Woodpecker, although with some bark chips on the larger side. As it happened, the second round of scaling had taken place approximately three hours earlier, five weeks to the day after the first.
In both instances, it appears that almost all the scaling was done by a female, although the image quality is too poor for me to be 100% certain. In both cases, the bird spent approximately 15 minutes on the trunk. It seems that squirrels (seen briefly at the beginning of the January series) are responsible for the modest quantity of scaling on the upper, less vertically oriented, part of the limb; this was my instinct at the time, and the idea is supported by the footage. The full time lapse sequences are at the bottom of the page. Phil extracted both sequences, and Steve Pagans created a slower version of the January 26th clip. The first four photos in the tiled mosaic series below were taken by Phil Vanbergen.
I’ll go into more detail in subsequent posts, but for now, I have a few observations.
- This work has a distinctive appearance, what I’ve called a layered look to the edges, that is consistent with what I’ve previously hypothesized for Pileated.
- While some of the bark chips are on the large side for what I have ascribed to Pileated, none are anywhere near as large as the larger ones that that we’ve ascribed to ivorybill. In addition, the chips found at this location and at another where I suspect the source is PIWO, seem to be less uniformly large in size and sometimes show signs of being taken off in layers, which matches what’s visible on the limbs.
- The tree in question was no more than six months dead, and the bark at the edges of the scaled area remained tight; however, dormant sweet gum bark is in the midrange of tightness relative to other hardwood species.
- This is a decay class that Tanner associated with ivorybills not PIWOs, but it’s clear that Pileateds can and do scale very recently dead sweet gum limbs, at least in mature bottomland forests.
- Tanner’s photographs provide little guidance in terms of differentiating between Pileated and Ivory-billed Woodpecker work on high branches. I suspect he thought of his monograph as more epitaph than guide to identifying feeding sign. Nevertheless, his descriptions offer some clues. “Scaling, the Ivory-bill works steadily, removing all the bark for quite an area; one may work at a spot for an hour or more.” And for Pileateds, “What scaling Pileateds were observed to do was mostly on loose bark and was never as extensive or as cleanly done as the work of the ivorybills.
To conclude this installment, we already suspected that Pileateds can and do scale freshly dead sweet gums before the bark has loosened; these images show them doing it in a way that is inefficient and neither ‘extensive’ nor ‘clean’. The total surface area scaled over approximately 30 minutes is modest compared to scaling we suspect to have been done by ivorybills. In addition, PIWO work has some characteristics that may be recognizable upon close examination of the affected limbs and bark chips. The fact that these characteristics can be seen on medium-sized sweet gum limbs, with their relatively thin and only moderately tight bark, suggests that it should be even more evident on larger limbs, boles, and other tighter barked species. More on this and on bark chips in subsequent posts.
I’m looking forward to returning to Louisiana at the end of the month and have high hopes that our trail cams will reveal just what’s removing bark from the trees in our search area.
In the meantime, I thought I’d compile a new group of links to images showing bark scaling (without accompanying excavation) done or suspected to have been done by Pileated Woodpeckers. Obviously long dead snags are excluded. Compare these images with the work we suspect to have been done by ivorybills, all on live or freshly dead wood. As discussed in this and other posts, it’s our hypothesis that the differences are anatomically determined.
And another (scroll down).
Apparently a long dead snag, suspected PIWO, note small chips on the ground and the way they appear to have been flaked rather than pried off.
I suspect the work from Congaree shown on this Cornell Mobile Search Team page is Pileated. As the commentary indicates, it’s somewhat consistent with what would be expected for IBWO, but it’s patchy and not extensive:
This appears to be a softwood. (Our diagnostic criteria include only hardwoods). Even so, note the layered appearance:
Another softwood, also showing the tendency to flake bark off in layers rather than knock off large chunks:
Hard to tell the age of this snag, but note the layered appearance on the right. The apparently stripped limb in the foreground appears to be long dead:
On the morning of the last day of Mark’s last visit, April 26, 2015, the forecast (and more importantly the doppler confirmed it) was terrible, and Mark had a terrible head cold and or allergy issue going on. He was feeling terrible! However, earlier that week he had brought in an interesting photo of a clearly active, freshly constructed cavity that was approximately 65 meters from the road he spotted it from. It didn’t take a lot of persuading to get him to agree to go with me the morning of the 26th. While we both expected the cavity (for a number of reasons, in spite of its apparent shape) to house a Pileated Woodpecker or PIWO, the shape was intriguing (from the road, anyway) enough to make it worthwhile to see what popped out the next morning. We arrived at daylight on the 26th. I strolled quietly out to about ten yards from the cavity tree. Mark, with his bigger lens stayed with the truck. I played a Pileated Drum, followed by a “scolding” “Kuk” series. Immediately, the male PIWO pictured below popped his head out. This is Mark’s best result – the light, and angle was bad and the ISO makes the pic quite grainy.
Then, he perched in a nearby tree, scolding.
I’ve never noticed a PIWO doing that with its crest before. Interesting… After scolding me for a bit, he circled around and flew off – likely to check on a nearby mate. Having bothered him enough for the day, I eased back to the truck and Mark and I headed home. These poor photos are the best we could do – with a bird we fully expected to be there and good open shooting. The point being made is – birds are difficult to photograph under the best of conditions. The reason I wanted to do a post on it? I wanted to share a cool moment in our efforts with our readers. Hope y’all enjoy… Frank