seasonal rest-site preferences of pine martens in the northern sierra nevada

Seasonal Rest-Site Preferences of Pine Martens in the Northern Sierra NevadaAuthor(s): Wayne D. SpencerSource: The Journal of Wildlife Management, Vol. 51, No. 3 (Jul., 1987), pp. 616-621Published by: Wiley on behalf of the Wildlife SocietyStable URL: http://www.jstor.org/stable/3801279 .

Accessed: 04/07/2014 15:21

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Wiley and Wildlife Society are collaborating with JSTOR to digitize, preserve and extend access to The Journalof Wildlife Management.

http://www.jstor.org

This content downloaded from 82.46.97.105 on Fri, 4 Jul 2014 15:21:33 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=black

http://www.jstor.org/action/showPublisher?publisherCode=wildlife

http://www.jstor.org/stable/3801279?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


616 FURBEARER ECOLOGY IN MAINE * Major and Sherburne J. Wildl. Manage. 51(3):1987

eastern Maine. M.S. Thesis, Univ. Maine, Orono. 36pp.

MESSIER, F., AND C. BARRETTE. 1982. The social system of the coyote (Canis latrans) in a forested habitat. Can. J. Zool. 60:1743-1753.

MOHR, C. O. 1947. Table of equivalent populations of North American small mammals. Am. Midl. Nat. 37:223-249.

MONTHEY, R. W. 1978. Relative abundance of mammals in commercially harvested forests in Maine. Ph.D. Thesis, Univ. Maine, Orono. 64pp.

MOORE, T. D., L. E. SPENCE, AND C. E. DUGNOLLE. 1974. Identification of the dorsal guard hairs of some mammals of Wyoming. Wyo. Game and Fish Dep. Bull. 14. 177pp.

PALMAN, D. S. 1977. Ecological impact of Interstate 95 on small and medium-sized mammals in northern Maine. M.S. Thesis, Univ. Maine, Orono. 34pp.

, AND V. B. RICHENS. 1979. Effects of In- terstate 95 on small mammals in northern Maine.

Univ. Maine at Orono, Life Sci. and Agric. Exp. Stn. Res. Life Sci. 26. 12pp.

PARKER, G. R. 1981. Winter habitat use and hunting activities of lynx on Cape Breton Island, Nova Scotia. Pages 221-248 in J. A. Chapman and D. Pursley, eds. Proc. Worldwide Furbearer Conf., Worldwide Furbearer Conf., Inc., Frostburg, Md.

RICHENS, V. B., AND R. D. HUGIE. 1974. Distri- bution, taxonomic status, and characteristics of coyotes in Maine. J. Wildl. Manage. 38:447-454.

ROSENZWEIG, M. L. 1966. Community structure in sympatric Carnivora. J. Mammal. 47:602-612.

SHIELD, J. 1972. Acclimation and energy metabo- lism of the dingo, Canis dingo and the coyote, Canis latrans. J. Zool. (London). 168:483-501.

TOPPAN, F. 1935. Physiography of Maine. J. Geol. 43:76-87.

Received 18 March 1985. Accepted 7 January 1987.

SEASONAL REST-SITE PREFERENCES OF PINE MARTENS IN THE NORTHERN SIERRA NEVADA

WAYNE D. SPENCER,' Department of Forestry and Resource Management, University of California, Berkeley, CA 94720

Abstract: I quantified selection of various types of resting sites by American pine martens (Martes amer- icana) in the northern Sierra Nevada and attempted to identify important factors in rest-site selection. Seven martens were observed resting 180 times in 95 sites from July 1979 through September 1980. The relative use and availability of trees, snags, stumps, and logs were compared. During periods of <100% snow cover, live trees were used most frequently as resting sites although they were used less than expected by availability; relative to availability snags were used most. Snags used by martens were almost exclusively large firs (Shasta red [Abies magnifica] and white [A. concolor]) that retained most of their bark. Live trees used by martens were primarily large lodgepole pines (Pinus contorta). During periods of complete snow cover, all resting sites were beneath snow; live trees were not used. Although logs provided 52% of subnivean resting sites, snags and stumps were used more relative to their availabilities. Cavities in decayed wood beneath snow, especially those used by Douglas' squirrels (Tamiasciurus douglasii) as cone caches, seemed preferred for winter resting sites. Martens reused subnivean resting sites more often than non-subnivean sites and sometimes reused the same site for several consecutive days.

J. WILDL. MANAGE. 51(3):616-621

Pine martens use a miscellany of resting sites throughout the year. Although these have been variously described (Marshall 1942, Mech and Rogers 1977, Campbell 1979, Masters 1980, Si- mon 1980, Hargis 1981, Steventon and Major 1982, Buskirk 1984), no reports have addressed the availability of refugia relative to their se-

lection as resting sites by martens. Additionally, only Buskirk (1984), working in Alaska, has ex- amined the attributes that affect seasonal rest- site selection. The objectives of this study were to quantify marten preferences for different types of resting sites in seasons differing in snow cover and to explain rest-site selection via attributes of the refugia.

Funding was provided by Mclntire-Stennis Proj. 3501 of the Calif. Agric. Exp. Stn., Region 5 of the U.S. For. Serv., the Wollenberg Found.,

Present address: Department of Ecology and Evo- lutionary Biology, University of Arizona, Tucson, AZ 85721.



J. Wildl. Manage. 51(3):1987 MARTEN REST-SITE PREFERENCES * Spencer 617

and the Univ. Found. Wildl. Fund. R. H. Barrett

provided advice throughout, and W. J. Zielinski, M. F. Dedon, C. Kellner, and A. P. Clevenger assisted in the field.

METHODS The study area was the 40-km2 Sagehen Creek

Drainage Basin, between 1,880 and 2,620 m elevation, on the east side of the Sierra Nevada, California. The vegetation and climate have been detailed previously (Spencer et al. 1983).

Six martens (2 M and 4 F) were fitted with 26-g radio collars. They were located using hand- held telemetry receiving gear 179 times in 95 different sites from July 1979 through Septem- ber 1980. Ninety-eight percent of these observations were on 4 martens (1 M and 3 F) with 21-62 resting observations each. One additional observation was on an unmarked marten. Al-

though generalization from data on so few individuals is risky, tentative conclusions are war- ranted because these martens inhabited a wide range of habitat types (Spencer et al. 1983), and they all exhibited similar patterns of rest-site selection.

Variability in radio-signal strength was mon- itored prior to and during each location attempt to determine whether the marten was moving or stationary, so that refugia into which the animals were chased by the observer were not counted as resting sites. All resting sites were flagged for subsequent relocation, and marten behavior, weather, and habitat data were recorded. I classified each resting site to a general type (e.g., snag, live tree, or log) and recorded measurements appropriate to that type follow- ing snowmelt in 1980. For subnivean sites, I was often able to pinpoint the marten's resting site

using evidence collected after snowmelt. Within the home range of each marten, I

calculated the density, and hence relative availability, of the 4 predominant types of resting sites-trees, snags, stumps, and logs-using 170 0.04-ha habitat plots distributed randomly within marten home ranges (Spencer 1981). Snags were defined as standing, dead trees >1.5 m tall; stumps were <1.5 m tall. For usage-availability comparisons between the 4 rest-site types, relative availability had to be adjusted for differences in the minimum diameter that provides a potential resting site for a marten. Thus, I assumed that the smallest tree, snag, stump, and log used by a resting marten represented the minimum size of each that provided a suitable

refuge. Although this ignores other potential selection factors (e.g., tree species), it provides a reasonable basis for determining the relative availabilities of the 4 major rest-site types within marten home ranges.

I evaluated rest-site selection for 2 periods: 100% snow cover and <100% snow cover. All

resting sites were beneath snow as long as it covered the ground completely; subnivean sites were not used when snow was patchy.

Ideally, an unbiased statistical test of usage- availability data is desired for preference analysis, such as the mean rank-difference method proposed by Johnson (1980). Unfortunately, this method requires a larger sample of individual animals than I had. Instead, I pooled observations over all individuals, so that sample size is the number of observations rather than individuals, and used Strauss' (1979) linear preference index to compare use and availability. Pooling was justified because comparisons among the 4 martens with

>_21 observations revealed no no-

table individualistic differences in the relative rankings of rest-site types selected. In pooling observations, I weighted the availability of each type within each home range in proportion to the number of observations for the correspond- ing marten.

The linear preference index is subject to bias because the investigator's decision to include or omit any component in the analysis can affect index values for other components (Johnson 1980). To avoid this bias I analyzed preferences both with and without rest-site types that had this effect.

RESULTS AND DISCUSSION Forty-six of the 180 observations of resting

martens were in man-made structures. Lumber piles and buildings were used by martens as shelter regularly throughout the year. The fol- lowing seasonal analyses are restricted to observations of martens using natural resting sites.

Although most observations were made during daylight, all rest-site types were used both day and night. Marten often remained in a sin- gle resting site for >20 hours during winter (Zielinski et al. 1983), and 26 nighttime checks throughout the year found them using specific sites that they also had used diurnally.

Non-Subnivean Rest-Site Selection

Trees, snags, stumps, and logs provided 86% of the non-subnivean resting observations (Table



618 MARTEN REST-SITE PREFERENCES * Spencer J. Wildl. Manage. 51(3):1987

+0.4 -

> +0.2 w 0 z_ w z ___

z 0 w uJ LL. w

m - w

a. -0.2

-0.4

0.8

7USE

SAVAILABILITY 0.6- z 0

o 0.4 a. 0 a.

0.2

0 w a.

oz

I- RESTING SITE RESTING SITE

Fig. 1. Relative use and availability of rest-site types within pine marten home ranges and a rest-site preference index (PI) during periods of <100% snow cover. (PI = relative use minus relative availability with 95% Cl's [Strauss 1979].)

1). Although live trees were most frequently used, snags were used more, relative to their availability, than any of the other 3 most-used types (Fig. 1).

Although martens sometimes rested in exposed sites, such as atop tree limbs, the majority of observations (68%) were in enclosed structures, such as cavities or burrows. Exposed sites offer little protection from adverse weather and avian predators and are used chiefly for sunning

Table 1. Number of individual resting sites and number of collective observations of 7 pine martens using various types of resting sites at Sagehen Creek, California, July 1979-Sep- tember 1980.

Non-subnivean Subnivean

Observa- Observa- Type Sites tions Sites tions

Live tree 19 24 0 0 Snag 18 20 7 12 Stump 5 6 7 15 Log 3 6 13 28 Log pile 0 0 6 8 Rock pile 3 4 2 5 Willow shrubs 3 4 1 1 Sedge nesta 1 1 0 0

Total 52 65 36 69 a Sedge "form" or nest on the ground at base of lodgepole pine.

(Simon 1980; W. D. Spencer, unpubl. data). Martens resting in exposed sites usually ap- peared nervous upon my approach and often fled. Those resting in cavities were less prone to flee, and they never fled from subterranean sites. Mech and Rogers (1977), Campbell (1979), and Masters (1980) also found marten to typically use cavities and burrows.

Martens were found resting in hollows inside standing dead trees at least 7 times. Three observations were of martens in woodpecker nest holes in snags. Snags used as resting sites were almost exclusively large-diameter (: = 102-cm diam at breast height [dbh], range = 58-147) fir

snags. Comparing snags used by martens with the availability of snags of various sizes reveals that martens selected resting sites from among only the largest 15% of snags. Snags used by martens also generally retained most of their bark (2 proportion of trunk covered by bark =

68%, range = 15-100). Fir snags tend to retain their bark longer than other species, which has- tens decay and cavity formation in the humid sub-bark environs (Raphael 1980).

Similar to my findings, Campbell (1979) reported that 56% of rest-site observations during snow-free months in Wyoming were in large Engelmann spruce (Picea engelmannii) and subalpine fir (Abies lasiocarpa) snags. In con- trast, martens apparently seldom use snags in eastern North America, possibly reflecting re- duced availability of large snags or different selection criteria imposed by the environment. Of 8 rest-site observations in Minnesota (Mech and Rogers 1977), and 5 in New York (Masters 1980), during April and May, none were in snags. Also, citing 27 summer observations, Steventon




and Major (1982) reported that martens in Maine

typically rested in conifer crowns. Most live trees used by martens were large

( = 62-cm dbh, range = 23-147), deformed lodgepole pines. Lodgepole pine was used dis-

proportionately to its relative abundance and seemed to be strongly favored over other species (X2 = 34.5, P < 0.001, 1 df) even though large lodgepole pines (>23-cm dbh) were only 57.6% as abundant as fir trees of comparable size. Live- tree resting sites were typically on limbs or in forks (11 observations), or in dense clumps of branches (witches' brooms) created by pine dwarf mistletoe (Archeuthobium campylopo- dum) (8 observations). Lodgepole pine may have been preferred for live-tree resting sites in part because it has a high incidence of mistletoe in- festation (W. D. Spencer, unpubl. data). In addition, 58% of observations of live-tree use were in riparian lodgepole associations, which are primary marten foraging areas (Spencer et al. 1983) but which lack fir snags.

Twelve observations were of martens resting in cavities in or beneath stumps or logs, includ-

ing 1 subterranean hollow created by the decay of a large root. Martens selected from among the largest 23 and 4% of available logs and

stumps, respectively.

Subnivean Rest-Site Selection When snow cover was continuous, sites be-

neath the snow were used exclusively. Ninety- one percent of subnivean observations were associated with logs, stumps, and snags; live trees were not used (Table 1). Cavities in decayed wood accounted for 51% of the subnivean observations; another 41% were beneath snow-covered logs or in crevices between piled logs. Al-

though logs provided 52% of subnivean resting structures, snags and stumps seemed to be used more than logs, relative to their availability. This result depends upon whether trees, which were not used as resting sites during periods of com-

plete snow cover, are included in the analysis as available sites. When the preference index is calculated including live trees, the index values for logs, stumps, and snags are all significantly positive, with little difference among them (Fig. 2A). However, if trees are deemed "unavail- able" and omitted, the index value for logs be- comes significantly negative, whereas values for

snags and stumps remain significantly positive (Fig. 2B).

For subnivean resting sites, the height of the

resting structure seems inconsequential, except perhaps as related to snow depth. Consequently, there appears to be no difference between preferences for snags and stumps during the period of 100% snow cover (whether or not trees are included in analysis) (Fig. 2). In fact, subnivean

snag and stump sites should probably be con- sidered 1 type. Snags used during the period of 100% snow cover were significantly shorter than those used during other periods (T = 4.0 ? 4.6 [SD] m vs. I = 13.1 ? 5.5 [SD], P < 0.001). All of the subnivean stump and snag resting sites were highly decayed, whereas non-subnivean

resting sites included tall, hard snags. Thus, in

general, it appears that the soft bases of large, dead fir trees that provide cavities near or beneath the ground offer ideal winter resting sites.

All previous investigations of winter rest-site use by martens also have indicated that most or all sites are beneath snow (Marshall 1942, Simon 1980, Hargis 1981, Steventon and Major 1982, Buskirk 1984). Pulliainen (1980) suggested a lat- itudinal gradient in the use of subnivean resting sites by European pine martens (M. martes), noting that martens farther north used more sub-snow sites, perhaps due to more severe win- ters there. There is, however, no evidence for such a gradient in American pine martens.

Seventy-six percent of all subnivean observations, and 88% of those associated with dead- wood refugia, involved martens in cavities used as cone caches by Douglas' squirrels. Although the proportion of available refugia that were used by squirrels was not recorded, snow-track-

ing observations and the low proportion of habitat control plots having squirrel cone caches (23%, Spencer 1981) indicate that squirrel cone caches are selected preferentially. Similarly, Buskirk (1984) found 70% of marten winter

resting sites in red squirrel (T. hudsonicus) middens. Middens may add to the comfort of the

resting site because piled cone scales trap air and can act as an insulative bed. In addition, squirrel midden cavities provide martens with

ready access to the subnivean zone and may provide them with food. Whereas Murie (1961) and Buskirk (1984) believed that martens evict- ed or cohabited with red squirrels in their dens, de Vos (1951), Quick (1955), and Simon (1980) believed that martens killed squirrels in their shelters. I did not break open any marten resting sites to look for squirrel remains but found cir- cumstantial evidence for the latter hypothesis (Spencer 1981, Zielinski et al. 1983).



620 MARTEN REST-SITE PREFERENCES * Spencer J. Wildl. Manage. 51(3):1987

+0.4 A. B. +0.4

w 1 o +0.2 z

0. .

Lz 0 w

w U-

w -0.2 a- +"

-0.4 L

0.8

z 0.6 Q USE AVAILABILITY

0

a.. O 0.4- O

a. 0.2-

0 w

wR 03 S 0

CD ESTING S EZ C/)CO CO CCO CO

RESTING SITE Fig. 2. Relative use and availability of rest-site types within pine marten home ranges and a rest-site preference index (PI) during periods of 100% snow cover, calculated with trees included (A) and excluded (B) as potential resting sites. (PI = relative use minus relative availability with 95% Cl's [Strauss 1979].)

A comparison of sites where martens were observed resting 1, 2, or

_3 times revealed a great-

er incidence of revisitation for subnivean than non-subnivean sites (x2 = 14.22, P < 0.001, 2 df). (Only 12% of non-subnivean sites were known to be revisited vs. 42% of subnivean sites.) Perhaps the more benign climate during snow- free seasons places fewer constraints on the types of resting sites used by martens so that a greater range of potential sites is available. However, on average, martens also captured larger prey (which may be used for multiple meals) in winter than in summer (Zielinski et al. 1983). Dur-

ing winter, they cached and either stayed with or returned to the carcasses of 3 snowshoe hares

(Lepus americanus), and 1 each of Douglas' squirrel, northern flying squirrel (Glaucomys sabrinus), and blue grouse (Dendragapus ob- scurus).

As noted by Buskirk (1984), a miscellany of

resting sites scattered throughout the home

range, each convenient to primary foraging patches, allows a marten to choose a resting site suitable to current conditions with a minimum of travel. Thus, a variety of large snags, stumps, and logs dispersed throughout the home range




seem requisite for high quality marten habitat. Of particular importance are large, dead trees, such as firs, that are prone to the formation of cavities near ground level, since these provide ideal shelter during winter.

LITERATURE CITED BUSKIRK, S. W. 1984. Seasonal use of resting sites

by marten in south-central Alaska. J. Wildl. Man- age. 48:950-953.

CAMPBELL, T. M. 1979. Short-term effects of timber harvests on pine marten ecology. M.S. Thesis, Colorado State Univ., Fort Collins. 71pp.

DE Vos, A. 1981. Tracking of fisher and marten. Sylva 7:15-19.

HARGIS, C. D. 1981. Winter habitat utilization and food habits of pine martens in Yosemite National Park. M.S. Thesis, Univ. California, Berkeley. 57pp.

JOHNSON, D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61:65-71.

MARSHALL, W. H. 1942. The biology and management of the pine marten in Idaho. Ph.D. Thesis, Univ. Michigan, Ann Arbor. 107pp.

MASTERS, R. D. 1980. Daytime resting sites of two Adirondack pine martens. J. Mammal. 61:157.

MECH, L. D., AND L. L. ROGERS. 1977. Status, dis- tribution, and movements of martens in north- eastern Minnesota. U.S. For. Serv. Res. Pap. NC- 143. 7pp.

MURIE, A. 1961. Some food habits of the marten. J. Mammal. 42:516-521.

PULLIAINEN, E. 1980. Winter habitat selection, home

range, and movements of the pine marten (Martes martes) in a Finnish Lapland forest. Pages 1068- 1087 in J. A. Chapman and D. Pursley, eds. Proc. Worldwide Furbearer Conf. Worldwide Fur- bearer Conf., Inc., Frostburg, Md.

QUICK, H. F. 1955. Effects of exploitation on a marten population. J. Wildl. Manage. 20:267- 274.

RAPHAEL, M. G. 1980. Utilization of standing dead trees by breeding birds at Sagehen Creek, Cali- fornia. Ph.D. Thesis, Univ. California, Berkeley. 195pp.

SIMON, T. L. 1980. An ecological study of the marten in the Tahoe National Forest, California. M.S. Thesis, California State Univ., Sacramento. 187pp.

SPENCER, W. D. 1981. Pine marten habitat preferences at Sagehen Creek, California. M.S. The- sis, Univ. California, Berkeley. 121pp.

,R. H. BARRETT, AND W. J. ZIELINSKI. 1983.

Marten habitat preferences in the northern Sier- ra Nevada. J. Wildl. Manage. 47:1181-1186.

STEVENTON, J. D., AND J. T. MAJOR. 1982. Marten use of habitat in a commercially clear-cut forest. J. Wildl. Manage. 46:175-182.

STRAUSS, R. E. 1979. Reliability estimates for Ivlev's electivity index, the forage ratio, and a proposed linear index of food selection. Trans. Am. Fish. Soc. 108:344-352.

ZIELINSKI, W. J., W. D. SPENCER, AND R. H. BARRETT. 1983. Relationship between food habits and ac- tivity patterns of pine martens. J. Mammal. 64: 387-396.

Received 30 July 1985. Accepted 7 February 1987.



seasonal rest-site preferences of pine martens in the northern sierra nevada

Documents