top of page

Habitat Associations

of Fruit-Producing Species

As you might expect, species that produce fruit consumed by grizzly bears in the northern US Rocky Mountains are not distributed randomly relative to different habitats, especially as defined by temperature, moisture, and forest succession. This non-random distribution creates a broadscale mosaic that results in seasonal and longer-term annual changes in both where and in what amounts grizzlies eat fruit in this part of the world. As a consequence, it is worth examining these ecological distributions of fruit-producing species--both geospatial and temporal--as a foundation for understanding grizzly bear habitat relations, as well as historical and prospective future prospects for grizzlies in the northern Rocky Mountains.

Habitat Distributions

The species that produce most of the fruit consumed by Rocky Mountain grizzly bears differ substantially in their distributions relative to elevation (as a proxy for temperature) and site moisture. There are very few comprehensive surveys of where such species live, and in what abundance. The closest approximation is data collected by Bob Pfister and his colleagues in developing a forest habitat type classification for western Montana. They sampled species presence and abundance at numerous plots, which they used as a basis for defining habitat types with a characteristic elevation range and moisture signature, which they helpfully provided in a pioneering publication entitled "Forest Habitat Types of Western Montana."


I exploited their data to generate the graphs shown at left. Each panel shows the distribution of four different species, differentiated top to bottom, relative to site moisture, left, and elevation, right. The solid colored lines represent median percent canopy covers of each species, and the gray bounding bands, upper and lower quartiles.

There are several takeaways. Modal elevational (temperature) distributions increase from top (kinnikinnick, Arctostaphylos) to bottom (huckleberry, Vaccinium), with kinnikinnick at lowest elevations, and huckleberry at highest. Aside from this, kinnikinnick occupies the driest sites of these four, and huckleberry the wettest. Serviceberry (Amelanchier) and buffaloberry (Shepherida) occupy a wider range of site moistures, but differ by serviceberry being a somewhat lower-elevation species, and buffaloberry, higher.

The maps above provide a more literal geospatial view of how different fruit-producing species are distributed in two landscapes emblematic of the continental side of the Rocky Mountains straddling the US-Canada borderlands: The East Front of the Rockies in northwestern Montana at left, and the foothills and adjacent mountains of Alberta at right.

The map of the East Front shows the distribution of various habitats that support species producing fruit consumed by bears in green. Most of this map extends eastward into the grasslands abutting mountains to the west. The very narrow green tendrils extending out onto the plains correspond to riparian vegetation dominated by shrubs, cottonwood, and aspen. Chokecherry (Prunus) tends to be concentrated in these habitats, although this species is also comparatively abundant in foothills aspen forests as well. Aspen-dominated vegetation is shown in brighter green, and supports the most extensive stands of serviceberry (Amelanchier) in this region. Farthest to the west and deepest into the mountains, buffaloberry (Shepherdia) and huckleberry (Vaccinium) become increasingly abundant, coincident with conifer-dominated forests denoted by the darkest green. Each of these vegetation types and associated fruit-producing species is illustrated by the inset photos. A takeaway point is that, in contrast to serviceberry and huckleberry, chokecherry and serviceberry tend to be concentrated in ecotonal habitats abutting human-dominated landscapes.

The Alberta foothills maps shows modeled distributions of four important fruit-producing species. These maps and the underlying models were generated by Quinn Barber of the University of Alberta. Kinnikinnick (Arctostaphylos, in pink) is widely and contiguously distributed at lower elevations. By contrast, at least in this region, serviceberry (in purple) is patchily distributed as narrow ribbons that track attenuated favorable habitats. Buffaloberry (in dark red) shows up in three broad swaths at mid-elevations, north, central, and south. Huckleberry (in blue) also occurs in extensive contiguous stands, but at higher-elevations almost exclusively confined to the comparatively wetter southern portion of the this otherwise relatively dry region. Again, the takeaway point is that serviceberry and huckleberry tend to occur in more remote areas farther from human-exposed ecotones, which results in a safer foraging environment for the involved bears.

Wildfires are an important feature of grizzly bear habitat in northwestern Montana for a very simple reason. Both the abundance of fruit -producing species and the fruit production on them are stimulated by openings created by forest fires. This has been demonstrated conclusively for huckleberry (Vaccinium) and buffaloberry (Shepherdia), the two most important species exploited by grizzlies.

The graphs at right show fruit production (in this case, berries) for huckleberry, top, and serviceberry, bottom, as a function of percent forest canopy cover, left, and time since the last fire (i.e., stand age), right. Perhaps by chance, fruit production on both species peaks rather dramatically around 23 years since the last fire, on sites where tree canopies account for only 3-8% cover. 

The Importance of Wildfires

Fruit-Bearing Species: East Front

Keith Aune undertook one of the most intensive investigations to date of habitat use by grizzly bears in the Northern Continental Divide ecosystem. His study area reached along the north-south axis of the East Front, west into the main range of the Rocky Mountains. In addition to documenting the nature and location of grizzly bear foraging, he also characterized the vegetation in his study area.

I drew on his plot data to develop the figure at left which shows the relative abundance of huckleberry, buffaloberry, and serviceberry relative to an array of Pfister's habitat types ordered by elevation from lowest at bottom to highest at top. I also included whitebark pine (Pinus albicaulis), which is/was another high-quality but high elevation food in Aune's study area (see this entry for more on whitebark pine). Finally, I drew on multiple sources to identify zones within which lodgepole pine (Pinus contorta) and Douglas-fir (Pseudotsuga menziesii) are characteristic early-successional species that establish after wildfires.

Takeaways? Huckleberry is the most abundant fruit-producing species in mountainous portions of this region, concentrated in habitats that straddle lower and upper subalpine environments in which lodgepole pine tends to be the dominant seral tree species. By contrast, buffaloberry and serviceberry are much less abundant and distributed at lower elevations where Douglas-fir tends to be the dominant seral tree species. As a consequence, for reasons that I elaborate immediately above, fruit abundance in mountainous portions of the East Front is largely driven by fire frequency, extent, and duration at higher elevations where lodgepole pine is common.

The State of Fruit Production

So, given all this, what is the probable state of fruit production in grizzly bear habitat in areas where fruit is important to bears? As a preface, average intervals between fires in habitats that support the bulk of huckleberry are around 130 years. In areas supporting the bulk of buffloberry the interval is around 100 years (see this web site for more). Given these intervals and a peak in fruit production on both species between around 10 and 50 years post-fire (see above), one would expect roughly 31% of huckleberry habitat and 41% of serviceberry habitat to be at peak productivity, representing a burn extent of 0.8% per year in huckleberry habitats and 1% per year in serviceberry habitats.


So, what is the actual history? The map at right shows the extent of wildfires between 1980 and 2015 (red) in areas occupied by grizzly bears in the Northern Continental Divide ecosystem around 1980. The yellow dashed lines are isopleths of meat consumption, the inverse of fruit production (see this entry). The total area burned was around 25%, and, this, within a 27 year period, which amounts to a little less than 1% per annum, which is within the range of what one might expect given historical fire intervals in moist montane and lower and upper subalpine habitats of this region.

Interestingly enough, fire were more frequent between 1980 and the present compared to between 1940 and 1980, but comparable to what occurred during the late 1800s and earlier 1900s. 

bottom of page