Central place foraging by beavers (Castor canadensis): a test of foraging predictions and the impact of selective feeding on the growth form of cottonwoods (Populus fremontii)

by Mark A. McGinley and Thomas G. Whitham

Oecologia 1985 66:558-562

This study sets out to see if a general model for animal foraging applies to beavers. Like so many studies, this one examined behavior where beavers are not troubled with the choices presented by a mixed hardwood forest. The San Juan River in southern Utah affords beavers little more than a choice between which size cottonwood they want to cut down. The model posits that "animals feed in a way that maximizes net rate of energy or nutrient intake per unit time." In addition, "animals should forage more selectively farther from the central place," and "the number of items chosen should decrease with the distance from the central place." This model has been shown to apply to harvester ants, eastern woodrats, and whinchats.

Honing this model still further, there are alternative strategies suggested. When time is not a factor, animals will take larger items. When large items are a drag, so to speak, larger food items will be dropped when they are farther away from the central place. Three studies have supported the second alternative:

...in Massachusetts, Jenkins (1980) found that proportionally fewer large trees were felled as distance from the pond increased. Similarly, the average diameter of trees cut by beavers decreased with distance from the pond in North Dakota (Pinkowki 1983), and the size of the largest branches cut by beavers decreased with distance on Isle Royale (Belovsky 1984).

McGinley and Whitham found the perfect environment to test the first alternative. Cottonwood branches are rather small so there should be no time penalty if the beaver took the largest branch with a diameter of 30mm. So, since the large branches offer more food, as the beaver gets further from the pond it should maximize its efforts and take the large branches. In addition, the further from the river the larger were the branches. "Consequently, small branches should drop out of the diet as the distance from the river increases due to both decreasing profitability and the increasing abundance of more profitable branches."

So they found a sandy shore without much vegetation save for a stand of cottonwoods all about the same age and they established sample groups of ten at various distances up to 73 meters from the river, and meaured and counted the branches on those groups. First they found that beavers cut more branches off each tree closer to the river: 71% at 3.5m and 20% at 73 meters. They also found that at all distances the beavers preferred large branches. Indeed "with increasing distance the smallest branches were almost completely excluded from the diet." This was the opposite of what Jenkins found, but again, there was no time lost in taking the largest cottonwood branches. Indeed, Jenkins found that with one item foraged in Massachusetts, witchazel, which gets none too big, the beavers took larger branches when futher from the river.

Then the authors turn to the effect this behavior has on cottonwoods, finding that trees close to the stream are forced to rely on "asexual or vegetative" reproduction, while those further away mature enough to have seeds.

To me, this study simply proves that under perfect conditions beavers can behave in a predictable and, shall we say, logical manner. In the ponds I watch a model to explain what trees the beavers will take, and the manner in which they might take them is hard to come by. One factor in play here that may not be a factor in San Juan river is the process of discovery. For example, I find that beavers periodically harvest stands of large trees that are quite far from the pond, that it seems, they must have happened upon after several years of finding enough forage closer to the pond. Indeed these were poplars just a few meters beyond a group of oak they took out five years ago. The impetus for the discovery may have been a summer drought and a rather large colony. Getting to this grove of poplars required climbing about 30 feets and going over 50 meters from the pond which they could not see from the grove. I got the impression that once they discovered this they lost any sense of distance from the pond. In fact an ice storm three years before had quite limited the number of branches in the scraggily crowns of the trees. That said, they took branches that were by no means easy to carry to the pond. In addition beavers spend a great deal of time eating the bark on the trunk of the downed trees. I would assume the larger trees provide a larger meal when they do this.

I also notice a pattern in the way I notice beaver work. First I'll see small trees taken out, then medium sized trees cut and segmented and carried out and then large trees with branches stripped and trunk debarked. The beavers seem, as a colony, to decide to invest more time in certain areas, seemingly abandoning a favored spot upon discovery of a new area. Finally, on the issue of losing interest in small branches away from the pond, I was amused one winter to see a beaver ranging quite far from the pond under the bright winter sun. It failed in attempt to cut down a mid-size oak, then moved further into the woods, finally finding a small branch and dragging it all the way to the pond, passing along the way, quite a few branches of the same size.