Kenai Lowlands provide marginal habitat for expanding marten

Posted: Friday, May 23, 2008

For the last two years I have been studying the behavior and distribution of American marten on the Kenai Peninsula. As this project draws to a close, I would like to sum up what I have learned about these interesting creatures.

The capture of a marten in 2002 by refuge staff in the Swanson River oilfield after a nearly century-long absence of marten prompted this study to examine the current and historic distribution of marten across the entire Kenai National Wildlife Refuge. To accomplish this, I employed a variety of detection methods including harvest reports, carcasses from trappers, pelt sealing records, museum records, anecdotal observations and an aerial videography track survey flown in 2006. I summarized these surveys in a database spanning the last century with over 150 marten detections on the Kenai Peninsula, 70 of which occurred west of the Kenai Mountains in the last 30 years. This database provides strong evidence that marten have re-colonized the western Kenai Lowlands after a 100-year absence.

The reasons for the conspicuous absence of marten were the subject of the behavioral component of my research. I examined how marten use snow and habitat conditions, which could be key environmental factors that limit marten populations on the western Kenai. Snowcover at least 8 inches deep is important for insulating resting marten and their prey during periods of below-freezing temperatures. Mature, closed canopy forest with ample structure and debris are also important habitat components which provide marten with protection from predators, access to prey, and dry resting sites in the winter.

During the past two winters, I radio-collared and back-tracked marten in three study areas in the Kenai Mountains and Lowlands in order to understand marten use of snow and forest cover while resting and foraging. Marten in the lowlands study area foraged equally in white spruce/birch and black spruce forests, but rest sites were found only in white spruce/birch forests. White spruce forests generally contained more squirrel middens and deadfall, both important rest sites. I also found that while marten did not show any significant preference for specific snow depths while resting or foraging, average snow depths used by marten in the mountains were four times deeper than sites used in the lowlands, highlighting a fundamental difference in environmental conditions available to marten between the two regions.

Snow depths at rest sites in the lowlands averaged just 7 inches, in contrast to 50 inches in the mountains. The lowlands' shallow snow suggests that marten are exposed to higher levels of thermal stress in the lowlands. Using historic temperature and snow depth averages since 2000, I calculated the number of days at the Kenai Moose Research Center and at Cooper Lake during which snow cover was not deep enough to be insulating during below freezing periods ("stress days"). The Moose Research Center averaged over four times the number of stress days than Cooper Lake in the mountains. The majority of stress days occurred between November and January, a period frequently characterized by a combination of shallow snow and cold temperatures. I believe that, despite the availability of suitable forest cover in many areas, the Kenai Lowlands may represent sub-optimal habitat for marten because of its cold winters with thin snow cover.

Marten are, however, highly adaptable, and in spite of these marginal conditions, they have managed to expand their distribution from the mountains westward across the Kenai Lowlands in the past decades. Instead of relying on snowcover to insulate them while resting, they are using alternative rest sites such as squirrel middens. Squirrel middens are warm and dry, and depths of 12 inches are sufficient to insulate marten from low temperatures even in the absence of snowcover. In contrast, in the mountains, where snow depths were more than ample, marten rested beneath deep snow in sites under fallen trees and alder thickets. These differences in rest site selection are a good example of behavioral adaptation to different environmental conditions.

One might expect that suboptimal habitat and higher thermal stress in the lowlands would cause poorer population health. I found that body weights and fat reserves of necropsied marten from the lowlands were marginally lower than those from mountain populations. This suggests that lowlands marten are largely compensating for their more stressful environment by using alternative rest sites and finding sufficient prey to maintain thermodynamic requirements. Additionally, half of adult females collected from the lowlands showed evidence of past pregnancies, indicating the presence of breeding populations in the lowlands.

I am concerned, however, about the age and sex ratios of harvested marten. Lowland harvests showed equal numbers of males and females, in contrast with a 2-to-1 male-to-female ratio in the mountains. Also, lowland harvests were comprised of older individuals, half of which were older than two years, in contrast to mountain populations with a median age of one year. These data suggest that harvest pressure is cutting deeper into the breeding population in the lowlands, whereas mountain harvests are removing primarily younger, surplus individuals. While marten have re-colonized many areas of the western Kenai, lowland populations remain sparse and potentially vulnerable to over-harvest and climatic variation.

Although marten distribution on the Kenai Peninsula has expanded after a 100-year absence, will this trend continue in the face of a warming climate? Marten populations thrive in mature forests, with closed canopies and a consistent, deep snowpack. The interplay of several forces related to climate change will certainly alter all three of these habitat components and ultimately shape the future of marten on the Kenai.

Warmer temperatures will likely reduce thermal stress on marten especially in the lowlands, assuming they can continue to find suitable resting sites. Average maximum snow depths in the mountains have been increasing at rates of 0.44 inches/year, and 0.06 inches/year in the lowlands over the past 38 years, but it will be interesting to see whether these trends continue as winter temperatures continue to rise.

As the Kenai becomes drier and fires become more frequent, lowland spruce forests may be converted to early successional hardwood stands less suitable for marten. More frequent spruce-bark beetle outbreaks could increase the amount of woody-debris available as potential rest-sites, but they may at the same time decrease overhead cover and the prey base. Predicting population trends given the balance between opposing environmental trends is difficult, but something that will intrigue Kenai biologists as they examine furbearer ecology in the future.

I would like to thank everyone at the Kenai National Wildlife Refuge and all area trappers who contributed to making my research a success.

Andy Baltensperger is a graduate student at Colorado State University, finishing work on his M.S. thesis on the Kenai National Wildlife Refuge.

More details of Andy's marten study were reported in the Refuge Notebook of Nov. 9, 2007, available in the Refuge Notebook archive at

* n n

You can check on new bird arrivals or report your bird sighting on the Kenai National Wildlife Refuge Birding Hotline 907-262-2300.

Subscribe to Peninsula Clarion

Trending this week:


© 2018. All Rights Reserved. | Contact Us