Forest Management Guidelinesfor the Provision ofMarten Habitat

VERSION 1.0

May 1996

W. Robert WattJames A. BakerDavid M. HoggJohn G. McNicolBrian J. Naylor

Cette publication technique n’est disponible qu’en anglais .

© 1996, Queen’s Printer for OntarioPrinted in Ontario, Canada

Single copies of this publication are available at no charge from the address listed below:

Ontario Ministry of Natural ResourcesForest Management BranchForest Program Development SectionSuite 400, 70 Foster DriveSault Ste. Marie, OntarioP6A 6V5

Attention: Guidelines

This paper contains recycled materials.

This is a summary of the guidelines thathave been prepared to help planning teamsand managers address the needs of mar-ten in the preparation and implementationof forest management plans.

They are designed to maintain habitat ofsufficient quality and quantity to supporthealthy populations of marten.

Marten habitat needs are addressed at 2spatial scales - the landscape level, and theforest stand level.

The attached document consists of (i) anoverview of marten ecology and habitatrequirements, (ii) the implications of forestmanagement on marten habitat, (iii) addi-tional detail about the recommendations tomanage habitat, (iv) an example applicationof the recommendations, and (v) a discus-sion of the anticipated evolution of theguidelines.

Recommended Guidelines

To meet current and future habitat needs ofmarten throughout the boreal forest ofOntario, the following practices are recom-mended:

Landscape Level

1. Maintain 10 - 20 percent of the forest,which has the capability to producemarten, in suitable conditions.

Summary of Forest ManagementGuidelines for the Provision ofMarten Habitat

Suitable conditions are achieved wherestands have the following characteristics:

• the composition of the overstorey is greater than 40 percent in spruce, fir, or cedar. Mixed composition is preferred over pure stands.

• the canopy closure of conifer is greater than 50 percent

• are found within preferred FEC site and vegetation types (those rated as “good” or “fair” in Appendix 1).

• are at least 15 m in height (approxi- mately 80 years old). Uneven canopies are preferred.

2. Suitable marten habitat should bearranged in “core habitat areas” be-tween 30 and 50 km2 in size. A mini-mum of 75 percent of core habitatshould be comprised of suitablestands.

3. Partial harvesting can occur in asmuch as 30 percent of the core habitat,provided it retains 50 percent of theoriginal conifer basal area and crownclosure of at least 50 percent.

4. Core areas should be connected byriparian reserves and other unhar-vested forest. While connections neednot be continuous, gaps of open habitatmore than 1 - 2 km across should beavoided.

i

Stand Le vel

5. Harvesting methods should be modi-fied within the bounds of the Occupa-tional Health and Safety Act, to retainat least 6 dead or declining trees perhectare, with at least 2 of those ex-ceeding 30 cm dbh.

6. Silvicultural practices that retain logs,stumps, and other coarse woody debrison site should be encouraged.

Application of the Guidelines

These guidelines must be considered inthe preparation and implementation offorest management plans in the BorealForest Region of Ontario.

In the area of transition between the borealand the Great Lakes St. Lawrence forest,managers must make a choice to applyeither the marten, or the pileated wood-pecker guidelines. It is not required thatboth be applied. The choice of whichguideline to apply should be based on anassessment of habitat capability and suit-ability for the management unit and inconsideration of any ecoregional direction.

The complexity of natural systems pre-cludes a rigid set of rules. Rather, theguidelines identify key principles and rec-ommendations that must be adapted to fitlocal situations based on the professionaljudgement of experienced practitioners.However, any deviations from the guide-lines must be recorded and rationalized inthe forest management plan on the basis ofcompelling biological or socio-economicconcerns.

ii

Forest Management Guidelinesfor the Provision ofMarten Habitat

Preface

These guidelines have been prepared to help planning teams develop and implementsound forest management practices that contribute to ensuring the long-term health ofOntario’s forests. They comply with the Crown Forest Sustainability Act (CFSA; RSO1994), as well as the requirements of the April 1994 Decision of the EnvironmentalAssessment Board.

Using the Guidelines

The Forest Operations and Silviculture Manual (CFSA, Section 68), lists theseguidelines as ones which must be considered during the preparation andimplementation of forest management plans.

The considerable ecological variation associated with natural systems precludes a rigidset of rules to cover all situations facing forest planners and managers. Rather, theguidelines identify key principles and recommendations that must then be adapted to fitlocal situations based on the professional judgement of experienced practitioners. Anydeviations from the guidelines must be recorded and rationalized in the forestmanagement plan on the basis of compelling biological or socio-economic concerns.

Development of the Guidelines

These guidelines have been developed by combining current scientific evidence andexpert opinion. A companion document is being prepared to provide additional detail on theexisting scientific evidence for the recommendations contained in this manual. Also, spatialand non-spatial habitat supply models, and associated technical manuals are in develop-ment to assist in applying the guidelines. When improved tools are available, and as ourunderstanding of the interactions between marten and their environment improves, revi-sions to the guidelines will occur. At a minimum, the guidelines will be revisited every 5years.

iii

TABLE OF CONTENTS

Page

Summary ....................................................................................................................... i

Preface ..........................................................................................................................iii

1.0 Introduction ...................................................................................................... 11.1 Evolving Management Philosophy ...........................................................11.2 Evolving Management Methods and Technology .................................... 1

2.0 Marten Ecology and Habitat Use .................................................................... 22.1 General Description and Distribution .......................................................22.2 Life History and Development...................................................................32.3 Food Habits and Energetics..................................................................... 32.4 Population Density and Home Range ......................................................42.5 Habitat Relationships ...............................................................................42.6 Forest Ecosystem Classification and Marten .......................................... 72.7 Habitat Suitability Index Models and Marten ........................................... 7

3.0 Implications of Forest Management Activities ............................................... 83.1 Clearcutting ............................................................................................. 93.2 Partial Harvesting .................................................................................... 93.3 Silvicultural Interventions .......................................................................103.4 Landscape Composition ........................................................................103.5 Landscape Configuration .......................................................................10

4.0 Recommended Forest Management Guidelines ......................................... 114.1 Landscape Composition.........................................................................114.2 Landscape Structure ..............................................................................124.3 Stand Level ............................................................................................12

5.0 Application of the Guidelines ........................................................................13

6.0 Future Directions ............................................................................................146.1 Habitat Supply Forecasting ....................................................................146.2 Ecological Land Use Planning................................................................15

Acknowledgements ...................................................................................................16

Literature Cited ..........................................................................................................17

Appendix 1 - Forest Ecosystem Classification Interpretations ............................ 21

Appendix 2 - Applying the Guidelines: An Example Forest .................................22

v

1.0 Introduction

The marten (Martes americana) waselevated to the status of a “provinciallyfeatured species” by the EnvironmentalAssessment Board’s ruling on timber man-agement on Crown lands (April 1994). TheBoard directed that guidelines be preparedto manage for the habitat of marten in theboreal forest, and for the pileated wood-pecker (Dryocopus pileatus) in the GreatLakes-St. Lawrence forest of central On-tario.

The provision of marten habitat will supplyat least some of the habitat required byvarious other species that are also associ-ated with mature and overmature forests,cavity trees, and coarse woody debris.

1.1 Evolving Mana gement Philosoph y

Historically, the Ministry of NaturalResources (MNR) attempted to managehabitat for some game species, and thosespecies whose long-term survival was ofconcern (i.e. vulnerable, threatened andendangered species). In recent years, theappreciation of the connections among thecomponents of natural systems, and therecognition of the intrinsic value of allspecies has grown. At the same time, it hasbecome increasingly difficult to manage forthe specific, often conflicting, habitat needsof an ever growing list of species. Thus,MNR’s approach to resource managementhas been shifting to the maintenance of

Forest Management Guidelines forthe Provision ofMarten Habitat

entire ecological systems and all theirassociated biological diversity. This per-spective does not preclude management ofhabitat for individual species (such asmarten), as long as it does not threaten thelong-term well being of other species, orthe functioning of the overall biologicalsystem.

This evolution in thinking about resourcemanagement has been reflected in, andencouraged by, a number of governmentpolicy initiatives. For example, Ontario’sPolicy Framework for Sustainable Forestsand the 1994 Crown Forest SustainabilityAct promote the long-term health of forestecosystems. At the national level, Ontariohas indicated support for the provisions ofthe 1995 Canadian Biodiversity Strategy.

These guidelines are expected to be ap-plied in a way that contributes to the main-tenance of ecological systems andbiodiversity.

1.2 Evolving Mana gement Methodsand Technology

A key consideration in the maintenance ofecological systems is the need to manageat a number of spatial scales. Not only is itnecessary to manage for particular habitatfeatures at the forest stand level, but prop-erties of landscapes (of which theforest stand is part) must also be managed.

1

Maintenance or creation of particularlandscape characteristics (percent of foresttypes and age classes, forest patch sizeand distribution, etc.) will increase thelikelihood that all the biological diversityassociated with the landscape will beperpetuated. This multi-scaled perspectiveof habitat management requires that plan-ning at the forest Management Unit (MU)level be closely linked to broader land useplanning.

Thd properties of both landscapes and foreststands that together are considered neces-sary to ensure perpetuation of martenpopulations.

In the past guidelines have been criticizedfor being too prescriptive, and for not beingdirected at achieving specific goals. Inresponse to these criticisms, and to movetowards more adaptive management, themarten guidelines reflect current under-standing of the key habitat relationshipsthat are being used and tested in varioushabitat suitability index (HSI) models. Aswell, one or more habitat models andassociated technical manuals will ulti-mately support the application of the guide-lines. Those managers who wish to usehabitat supply modelling will find someassistance within these guidelines.

These guidelines consist of 5 major parts:(i) a brief overview of marten ecology andhabitat requirements; (ii) a brief descriptionof the implications of forest managementfor marten habitat; (iii) recommendationsfor providing marten habitat during forestmanagement planning at the landscapelevel and within forest stands; (iv) an ex-ample of how to apply the guidelines inforest management planning; and (v) adiscussion of anticipated improvements tothe guidelines, and future directions forforest management.

A more detailed review of the scientificevidence supporting these guidelines is inpreparation.

2.0 Marten Ecology and HabitatUse

2.1 General Description andDistribution

Marten are a medium sized member of theweasel family. They are between 500 and680 mm in length and weigh between 500and 1400 grams at maturity. Males aregenerally 20 - 40 percent larger than fe-males but otherwise are similar in appear-ance. They range in colour from tan to darkchocolate brown and have an irregularcream to amber coloured throat patch.

(GRAPHIC - CAJ008 (same as cover)

Marten

Marten are well distributed across NorthAmerican coniferous forests, ranging fromAlaska to Newfoundland, and from the treeline to the southern extent of coniferousforest in most regions (Strickland andDouglas 1987). Habitat change associatedwith human settlement and agriculturaldevelopment along with over trapping insome areas has resulted in a reduction ofthe marten’s original range along its south-ern borders (Thompson 1991, Buskirk andRuggiero 1994).

2

2.2 Lif e Histor y and De velopment

Marten are a long-lived species with arelatively low reproductive rate and slowphysical development. Marten mate in lateJune through early August and young areborn in March or April (Strickland andDouglas 1987). Female marten typicallymate for the first time at 15 months of ageand have their first litter at 2 years. Onlyone litter is produced each year. Litter sizeranges from 1-5 with a mean of 2.85(Strickland and Douglas 1987). Litter sizemay be age dependent, peaking at 6 yearsof age and declining to near zero by 12years (Mead 1994). Young marten developslowly by mammalian standards and areweaned at 42 days (Mead 1994). Both theage at which reproductive maturity isreached, and the annual reproductiveoutput for marten are lower than expectedfor an animal of its body size (Buskirk andRuggiero 1994).

Although marten are relatively long-lived,with some marten reaching 14+ years inthe wild, the mean age of 6,448 trappedmarten from central Ontario was less than1 year (Strickland and Douglas 1987). Inan uncut boreal forest, mortality of martenin a trapped population was estimated at34.5 percent, of which the majority ofdeaths occurred by trapping (28.1 percentof total mortality). However, in cutoverboreal forest, trapping accounted for thedeaths of all marten, for that portion of thepopulation within the cutover (Thompson1994). This rate of mortality in trappedpopulations is similar to that found in anaccessible logged area in Maine wheretrapping accounted for 90 percent of alldocumented mortalities (Hodgman et al.1994). Males are two to three times moresusceptible to trapping than females(Strickland and Douglas 1987), likely due totheir larger foraging areas where theyencounter more traps (Yeager 1950).

Other sources of mortality include starva-tion, and predation by other forest carni-vores. Females and immature animals aremore susceptible to starvation because ofthe increased energy demands of repro-duction, and for younger animals, thepoorer quality of habitat to which they haveaccess (Hawley and Newby 1957).

2.3 Food Habits and Energetics

Marten are opportunistic feeders and theirdiet includes a wide variety of animal andplant materials (Strickland and Douglas1987). Red-backed voles (Clethrionomysgapperi) and meadow voles (Microtuspennsylvanicus) are the principle items inthe diet along with snowshoe hare (Lepusamericanus) (Buskirk and MacDonald1984, Thompson 1986, Martin 1994) Red-backed voles, although a staple in the dietof marten, are usually only taken in propor-tion to their availability (Weckwerth andHawley 1962, Buskirk & MacDonald 1984).The larger prey items (meadow voles andhare) are preferred and are especiallyimportant in the marten’s winter diet (Raine1987, Thompson and Colgan 1994). Hareincreases in importance in the diet of mar-ten as the winter progresses (Thompsonand Colgan 1990).

Marten remain active throughout the winter.They do not store large energy reserves asfat prior to winter, nor is their long slenderbody shape conducive to energy conserva-tion. Marten are estimated to requireapproximately 80 Kcal/day at rest duringwinter, roughly equivalent to the energyobtained from 3 voles (Buskirk 1983). InOntario, Thompson (1986) reported that 85percent of the caloric intake of marten inwinter was comprised of hare when harewere abundant. Raine (1981) in Manitobareported that hare made up 53 percent ofmarten diets in winter. Access to largerprey may allow marten to remain inactive in

3

- 15o C were associated with reducedmarten activity in northern Ontario(Thompson and Colgan 1994).

2.4 Population Density and HomeRange

Marten occur at low densities and occupylarge home ranges. Marten densitiesestimated in Algonquin Park ranged be-tween 1.2 - 1.9 / km2 with densities of

sheltered resting sites for longer periods oftime, conserving energy and avoidingextremely low temperatures (Thompson1986, Buskirk 1983).

The ability to capture sufficient prey duringthe winter months is critical to martensurvival and reproduction. Marten embryos(blastocysts) remain in a suspended statethroughout the winter. Active gestationdoes not begin until spring (Jonkel andWeckwerth 1963), when those females thathave come through the winter in goodshape begin active gestation. Reducedreproductive rates were observed in martenduring prey declines in northern Ontario(Thompson and Colgan 1987). Martendensities were observed to fluctuate by asmuch as 85 percent during that same preydecline (Thompson and Colgan 1987).Similar reductions in population densitywere associated with prey reductions inMontana (Weckwerth and Hawley 1962).

During summer months, marten are mostactive at night when their principle prey isactive, and therefore most easily detected(Buskirk 1983). However, during winter,marten switch to being most active duringthe warmest part of the day, even thoughthis does not coincide with the activitypatterns of their prey (Buskirk 1983,Thompson and Colgan 1994). Martenspend the coldest periods in resting sites,often beneath the snow (Steventon 1979,Raine 1981). Temperatures of less than

resident adults estimated at 0.6 marten /km2 (Francis and Stephenson 1972).Thompson and Colgan (1987) estimatedmarten densities (adults and juvenilescombined) as high as 2.4 / km2 in uncutforest during the fall when prey was abun-dant, and as low as 0.4 / km2 in cutoversduring a spring when prey were not abun-dant. In Maine, adult marten densitieswere estimated to be 1.2 / km2 in unhar-vested forest, and 0.4 / km2 in clearcutareas (Soutiere 1979). Archibald andJessup (1984) estimated fall densities inthe Yukon to be 0.6 adults / km2. Expecteddensities for a carnivore of equal body sizeto that of marten are approximately tentimes higher than those reported for marten(Buskirk and Ruggiero 1994).

Home ranges of male marten are 2 - 3times larger than those of females andusually overlap several female homeranges. Home ranges estimated throughmark-recapture studies in Algonquin Parkwere 3.6 km2 for males and 1.1 km2 forfemales (Francis and Stephenson 1972).A number of radiotelemetry studies esti-mated home range sizes between 2.0 and15.0 km2 for males and 0.8 and 8.4 km2 forfemales (Strickland and Douglas 1987).Thompson and Colgan (1987) reportedtelemetry-based home ranges of 3.4 and1.0 km2 for males and females respectivelyin uncut boreal forest in Ontario. Homeranges for marten are 3-4 times larger thanwhat would be expected based on bodysize (Buskirk and Ruggiero 1994).

2.5 Habitat Relationships

In general terms, marten in the borealforest are inhabitants of conifer dominatedforests and associated vegetation typesfound nearby. In most studies, martenhave been found to prefer mesic (moist)coniferous forest in the latter stages ofsuccession, especially those with complexphysical structure near the ground. Marten

4

require a number of key habitat compo-nents including maternal den sites wherethe young are raised, winter dens andresting sites, abundant prey, and protectionfrom the elements and predators.

2.5.1Maternal Dens

Maternal den sites are usually located incavities in dead and living trees, or fallenlogs (Spencer 1981, Martin and Barrett1983, Hargis and McCullough 1984, Wynneand Sherburne 1984). Large structures arerequired to provide sufficient space forraising young. Hollow cedar logs andstanding stems larger than 40 cm diameterat breast height (dbh) were favoured inMaine (Wynne and Sherburne 1984).

2.5.2Resting Sites and Dens

Marten use of resting sites differs betweenseasons. In summer, marten rest in thecanopies of large conifer trees, usually inmature conifer forest (Steventon and Major1982, Wynne and Sherburne 1984). Largediameter cavity trees are important summerresting sites in western North America, butare less important in eastern NorthAmerica, perhaps due to their reducedavailability. Although marten forage widely,they return to familiar resting sites withintheir territories (Simon 1980).

In winter, marten use resting sites and densthat are well insulated beneath the snow.Thirty of 31 winter den sites in Maine werebeneath the snow in natural cavities formedaround large decayed stumps (Steventonand Major 1982). All dens encountered byThompson (1986) in northern Ontario werebeneath the snow. In Manitoba, Raine(1981) noted winter den sites associatedwith the seed caches (middens) of redsquirrels beneath the snow. Althoughmarten may forage in a wide variety ofconditions during winter, requirements for

resting sites are more specific. When snowcover is continuous the vast majority ofresting sites were associated with largelogs, stumps and dead or declining trees(Lofroth and Steventon 1990).

(GRAPHIC - downed log stump.Same as cover)

Winter den sites are often associated with largestumps, decayed logs, and squirrel middens.

2.5.3Foraging Habitat and CoarseWoody Debris

Thompson and Colgan (1987) suggest thatthe rate at which marten capture prey is adominant factor influencing habitat choice.During winter months mice and voles con-tinue to be active beneath the snow andprovide a major component of the marten’sdiet. In order to hunt in these “subnivean”spaces, marten search out the naturalhollows and air spaces in the snowpackwhich form around leaning trees, stumps,logs and other low physical structures com-monly referred to as coarse woody debris(CWD). Steventon and Major (1982) andSherburne and Bissonette (1994) suggestthat the abundance of CWD may influencemarten hunting success and thereforehabitat suitability. Marten use of subniveanaccess sites in and around coarse woodydebris has been described in most winter

5

habitat studies. However, Thompson andColgan (1994) believe that the importantrole of larger prey items such as haredecreases the importance of CWD innortheastern boreal forests. Snowshoehare are usually caught on the surface or inslight depressions in the snow.

(GRAPHIC - abundant coarse woodydebris. A3 - scenic forest floor)

Abundant coarse woody debris provides ac-cess to small mammal prey during thewinter

In summer, coarse woody debris continuesto play an important role as habitat for miceand voles. Large logs in intermediatestages of decay and rotten stumps arecritical components of the habitat for red-backed voles and other small mammals.The spaces beneath and within the logs areused as travel routes by small mammals.Hummocks, stumps and logs are primelocations for nesting. (Gunderson 1959,Raphael 1988).

2.5.4Canopy Closure

Overhead canopy cover by conifers is acrucial component of marten habitat in theboreal forest. Snow depths beneath conif-erous canopies are lower than in eitheropen areas or deciduous stands, andtemperature regimes are also moderatedby the overhead canopy.

Intermediate levels of canopy closurebetween 40 and 70 percent seem to bepreferred. Canopy closure less than 30percent is avoided during winter (Koehleret al. 1975, Spencer et al. 1983). Snyderand Bissonette (1979) reported better livetrapping success where the canopy closureof spruce and fir was greater than 50 per-cent. Marten in Alaska prefer sprucestands where overhead cover exceeds 70percent (Buskirk 1984).

Intermediate canopy closure is often asso-ciated with late successional conditions,where individual tree mortality has createdcanopy gaps as well as significant coarsewoody debris. Small canopy gaps mayalso provide more diverse understoreyconditions that are favoured by small mam-mal populations (Hunter 1990), and snow-shoe hare (Koehler 1990).

2.5.5Forest Vegetation Types

Overwhelmingly, studies of marten habitatuse have reported a preference for mesicor moist conifer and conifer-dominatedmixedwood forest types (Buskirk andPowell 1994). Francis and Stephenson(1972) and Taylor and Abrey (1982) foundmarten to prefer mixed forests over pureconifer conditions in central Ontario, similarto the findings of Soutiere (1979) in Maine.de Vos (1952) noted heavy use of cedarswamps, and a preference for white spruceover black spruce or jack pine-sprucemixes in northeastern Ontario. Balsam firand black spruce forests are used in New-foundland (Snyder and Bissonette 1987)and black and white spruce dominatedforests are the main habitat types used bymarten in Alaska (Buskirk and MacDonald1984).

6

(GRAPHIC - mature and overmaturestands. same as cover)

Mature and overmature stands of fir and spruceare preferred habitat for marten.

Use of late successional conifer forests ishigher in winter than summer (Campbell1979, Soutiere 1979, Steventon and Major1982, Buskirk et al. 1989). Avoidance ofdeep snow (Koehler and Hornocker 1977),cold winter temperatures (Buskirk et al.1989) and predators are apparent reasonsfor heavier winter use of coniferous stands(Buskirk and Powell 1994).

Juveniles are found in a wider range ofconditions than adults (Burnett 1981,Buskirk et al. 1989). Young animals arelikely excluded from the best habitats byadults, and suffer higher mortality rates asa result.

2.6 Forest Ecosystem Classificiationand Marten

The Forest Ecosystem Classification (FEC)systems used in boreal Ontario (Sims etal.1990, McCarthy et al. 1995) integrate anumber of factors that may influence mar-ten habitat choice, and therefore providesome opportunity to characterize habitats.FEC classifications integrate bothoverstorey and understorey composition aswell as soil features including texture,

moisture and drainage regimes, and theform and amount of humus. FEC does nothowever, explicitly characterize structuralcomponents of forest stands.

D’Eon and Watt (1994) hypothesized mar-ten use of FEC types in northeasternOntario, and those predictions were subse-quently field tested (Bowman et al. inpress). Racey et al. (1989) also used FECas a basis for predicting marten use offorest types in northwestern Ontario.Appendix 1 summarizes habitat relation-ships of marten based on FEC types usedin the boreal forest of Ontario.

2.7 Habitat Suitability Index Modelsand Marten

Habitat suitability index (HSI) models havebeen developed in a number of jurisdictionsto combine the various habitat require-ments of marten into a single habitat rating.Allen (1982) developed the first suitabilitymodel for marten. This model was used torate winter habitat, which was assumed tobe most limiting. Allen (1982) used 4variables to define suitability:

V1: Percent tree canopy closure wherestands with less than 25 percentcanopy closure are assumed to haveno habitat value in winter, and standswith 50 percent or greater canopyclosure were optimal.

V2: Percent of overstorey comprised ofspruce or fir where stands with 40percent or greater spruce or fir in theoverstorey were considered optimaland stands with no spruce or fir contin-ued to have very low value.

V3: Successional stage of the stand -seedling/sapling development stages(A) were assumed to have no winterhabitat value; polewood / sapling

7

8

stands (B) had low value; young stands(C) had moderate value; and matureand overmature stands (D) had optimalvalue as winter habitat.

V4: Percent cover of downfall (CWD) where25-50 percent cover by woody materiallarger than 7.6 cm in diameter wasconsidered optimal.

Allen (1982) combined the four variables toobtain a final result, as follows:

HSI = (V1 * V2 * V3 * V4) 1/2

(GRAPHIC - see hard copy)

Allen’s (1982) four variables used to rate mar-ten habitat suitability.

Most other HSI models for marten areadaptations of Allen’s original. In CentralOntario, Naylor et al. (1994) modifiedAllen’s (1982) model to accommodate theuse of Ontario’s Forest Resource Inventory(FRI). Stand height was substituted forsuccessional stage, percent spruce or firwas expanded to provide for weightedrankings of 8 different conifer species, andCWD was ranked by FRI working groups toprovide an estimate of the availability ofcoarse woody debris. The rating for standheight was doubly weighted to acknowl-edge the perceived importance of thisvariable to marten habitat.

Preliminary testing of the Central Ontariomodel with trapline harvest data has beenundertaken with encouraging results(Naylor et al. 1994).

Bowman et al. (in press) investigated winterhabitat use near Timmins, Ontario andfound that the best predictors of martentrack occurrence and abundance were, inorder of importance: percent composition ofspruce and fir, stand height, the abundanceof coarse woody debris and canopy clo-sure. These studies confirm the basicstructure of Allen’s (1982) model.

3.0 Implications of ForestManagement Activities

The available evidence suggests thatmarten are dependent on mature andovermature conifer dominated forests inOntario. Thus there is a potential competi-tion for stands - to either maintain them asmarten habitat, or to harvest them forindustrial purposes.

Fifty years of fire suppression in Ontariohas apparently contributed to the develop-ment of a forest dominated by older agedstands (Ward and Tithecott 1993). Martenpopulations have increased over that timedespite increased trapping (Strickland andDouglas 1987). As this older forest contin-ues to age and succumbs to disease,blowdown, fire and insects, marten popula-tions are expected to decline. The “normal-ization” of forest age class structures tomeet industrial demands for wood fibre inboreal Ontario will necessarily add to theanticipated decline.

Forest industry requirements for woodpromote rotation ages that fall below theage at which most stands begin to exhibitthe structural characteristics required bymarten. Intensive silviculture practices may

further reduce the availability of structuralfeatures used by marten, and further sim-plify the species composition of coniferdominated mixedwood stands preferred bymarten (Thompson 1991).

3.1 Clearcutting

The effects of clearcutting on marten habi-tat suitability and use are both immediateand long lasting. Densities of marten innew clearcuts are very low and stay low forup to forty years (Thompson and Harestad1994, Soutiere 1979, Steventon and Major1982, Snyder and Bissonette 1987). Homeranges are 3-4 times larger in areas thathave been clearcut, and mortality inclearcuts is markedly higher than in uncutforests (Thompson 1994).

Where abundant coarse woody debrisremains following a stand replacing distur-bance such as fire or clearcutting, thenegative impacts on marten may be par-tially mediated (Buskirk and Ruggiero1994). The amount of CWD remaining maydepend on the logging system being used,where full tree systems may leave lessCWD than either tree length or shortwoodsystems.

Clearcutting in spruce dominated borealforests results in marked changes in thesmall mammal community. Red-backedvoles are replaced as the dominant smallmammal by less preferred prey (e.g. deermice and shrews) in upland habitats (Simsand Buckner 1973, Martell and Radvanyi1977, Martell 1983) and by meadow volesin lowland spruce where grasses andsedges colonize recent cutovers. Snow-shoe hare usually do not colonize cutoverhabitats for up to seven years, and peakuse normally does not occur until after 20years (Litvaitis et al. 1985).

Although in some cutovers, meadow voles

can become quite common, they remainlargely inaccessible to marten if subniveanaccess sites are not abundant. In manyclearcuts, coarse woody debris is uncom-mon, having been trampled, windrowed,removed from the site, or burned in pre-scribed fires. Meadow voles and hares arealso noted for their population fluctuationsbetween years, making their availability asprey unpredictable from year to year. In asevere winter in northern Ontario, meadowvoles were absent from a recent lowlandspruce cutover until June (Courtin unpubl.).

Marten that do forage in cutovers in winterexperience higher mortality rates due topredation and trapping (Thompson 1994).

3.2 Partial Harvesting

Partial harvesting of forest stands providessome opportunity to maintain marten popu-lations in harvested forests. Diameter limitcuts in Maine that left 40 percent of theoriginal spruce-fir basal area did not reducemarten densities (Soutiere 1979). Campbell(1979) reported that harvesting up to 57percent of the harvestable trees had littleimpact on habitat quality. Partial harveststhat leave 40 - 50 percent of the canopy or50 percent of the basal area of foreststands should continue to provide suitablemarten habitat (Spencer et al. 1983, Lofrothand Steventon 1990).

Partially harvested stands usually developmore varied structural conditions whichprovide habitat for a diverse prey base formarten. The replacement of red-backedvoles as the dominant species in the smallmammal community that was observed inclearcut forests has not been observed inpartial harvests in mixedwoods (Martell1983), nor in diameter limit cutting in low-land black spruce forests (Courtin andBeckerton 1996).

9

maintain a residual population of martenwithin a commercially clearcut forest, 25percent of that forest should remain inmature and overmature conditions.

3.5 Landscape Configuration

Understanding of the impacts of landscapeconfiguration and pattern on marten islimited. Nevertheless, it is believed thatcontinued fragmentation of the matureconifer forest will have negative impacts onmarten populations.

As forests become more fragmented, thepopulations of many species fluctuate moremarkedly (Wiens 1985). The combinationof delayed sexual maturity, low reproduc-tive output, and marked reductions inmarten pregnancy rates during years of lowprey availability are indicative of the sus-ceptibility of marten populations to similarfluctuations.

(GRAPHIC - marten require large tracts.see photo in current draft. Slide available)

Marten require large tracts of mature coniferdominated forests.

The large home ranges, and correspondinglow densities that marten exhibit providefurther evidence that large areas of suitablehabitat are required to sustain their popula-tions.

Marten appear to avoid travelling throughlarge expanses of open habitat and rarely

3.3 Silvicultural Inter ventions

Clearcut and intensive renewal treatmentsdesigned to maximize wood production, doso at the expense of diverse structuralconditions within a stand (Thompson 1991).In turn, these treatments might contribute tolower prey densities and lower densities ofpredators including marten. Mechanical sitepreparation, for example, has been impli-cated in the elimination of red-backed volesfrom cutover sites (Martell 1983).

3.4 Landscape Composition

Forest landscape composition will continueto change as a result of the combinedforces of natural succession and forestmanagement activities. The area in olderage classes preferred by marten will de-crease. The availability of spruce/fir com-munities on the landscape is also expectedto change.

Current age-class distributions acrossmuch of Ontario’s boreal forest reflect thehistory of fire suppression. The percentageof older age stands will decrease markedlyover the next 40 years.

As forest stands are harvested, burned orotherwise disturbed, there is a naturaltendency for aspen and birch to colonizethese open sites, especially in mixedwoodforests. A reduced proportion of spruce onregenerating sites has been observedacross boreal Ontario (Hearnden et al.1992). This reduction in potential martenhabitat may be partially alleviated by theincreasing incidence of balsam fir.

The combined impact of increased youngeraged forests and reduced spruce composi-tion will have serious implications for thequantity and quality of marten habitat.

Soutiere (1979) believed that in order to

10

venture more than 500 - 1000 m from uncutforest areas (Steventon 1979, Soutiere1979, Steventon and Major 1982, Snyderand Bissonette 1987). Small populations ofmarten that are isolated from each other bylarge clearcut areas are at increased risk oflocal extinction. Should a random naturalevent deplete one of these isolated martenpopulations there would be a lower prob-ability of recolonization.

4.0 Recommended ForestManagement Guidelines

The recommendations that follow aredesigned to maintain healthy populations ofmarten. They recognize that an increasingamount of forest area will be maintained inyounger age-classed stands, but ensurethat networks of core habitat are main-tained across forest landscapes. Further, ifcore habitat areas are to serve their in-tended purpose, marten harvest should bemonitored and quotas adjusted wherenecessary to ensure that populations incore habitat areas are not undermined.

4.1 Landscape Composition

The key to maintaining healthy martenpopulations in managed forests is to ensurethat large areas of quality marten habitatare maintained across the landscape.

Determining just how much older forest isrequired to ensure healthy martenpopulations, requires sophisticated analyti-cal modelling and knowledge of detailedpopulation parameters, both of which arenot presently available. In the absence ofanalytical results, a decision must be basedon current understanding of natural distur-bance regimes, their variation, and theforest conditions they produce. VanWagner (1978) hypothesized that under

11

natural fire regimes approximately 1/3 ofthe forest area, on average, would be inmature and overmature conditions. How-ever, the extreme variation experienced insome natural systems might result in theloss of all suitable marten habitat duringsome time periods. In a managed forestwhere a continuous supply of many ben-efits is desired, and where natural eventssuch as fire and insects are also disturbingthe forest, these extreme levels of naturalvariation should be avoided. A conserva-tive level of suitable habitat must be estab-lished that is large enough to avoid the riskof losing marten habitat over large areas ofthe boreal forest.

Maintain 10 - 20 percent of the forest,which has the capability to producemarten, in suitable conditions .

Capable forest types are those that, atmaturity, support a spruce, fir, or cedardominated or mixed forest condition. Thesestands can be identified from current FRIstand descriptions, or based on FEC siteand vegetation types as rated in Appendix1.

Suitable conditions are achieved wherestands have the following characteris-tics:

• the species composition of the overstorey is greater than 40 percent spruce, fir, or cedar with mixed composition being preferred over pure stands.

• the canopy closure of conifers is greater than 50 percent.

• are found within preferred FEC site types and vegetation types for north- eastern and northwestern Ontario. They are those rated as “Good” or “Fair” in Appendix 1.

than 1 - 2 km in width should beavoided.

Core areas form the basis for recolonizingthe surrounding cutover forest and shouldbe well distributed across the landscape.Wherever possible, they should beinterconnected to promote continued ge-netic interchange and minimize the effectsof natural population declines within indi-vidual core areas.

Younger stands are unlikely to provide thestructurally diverse conditions, snags, andcoarse woody debris preferred by marten.Stands must have reached a stage wheremortality of larger trees has commenced.

4.3 Stand Level

It is recommended that harvesting prac-tices be modified within the bounds ofthe Occupational Health and Safety Act,to promote the retention of at least 6dead or declining trees per hectare, withat least 2 of these exceeding 30 cm dbh.

Dead and declining trees play an importantrole in marten reproduction and in thehabitat requirements of their prey. Contin-ued efforts to conserve dead and decliningtrees in managed forests are required inthe face of shorter rotations, mechanizedharvesting, and rigorous utilization stan-dards.

Further details on the management ofcavity trees, and otherwise dead or declin-ing trees, can be found in the guidelines forpileatedwoodpeckers (Naylor et al. 1996)and in a separate technical note for borealforests (Watt 1996).

Silvicultural practices that retain logs,stumps and other coarse woody debrison site are encouraged.

• are at least 15 m in height (approxi- mately 80 years old). Uneven cano- pies are preferred.

4.2 Landscape Structure

Suitable marten habitat should be ar-ranged in “core habitat areas” between30 and 50 km 2 in size. A minimum of 75percent of core habitat areas should becomprised of suitable stands.

Stands which are suitable for marten habi-tat should be maintained in aggregations ofsufficient size to support 25 or more adultmarten. Sub-populations of this size shouldbe able to persist for 40 - 50 years whenthe surrounding habitat should once againbe capable of supporting adult reproduc-tion. Core area requirements have beendeveloped based on conservative esti-mates of marten density between 0.6 - 0.8adult marten / km2 in good to moderatehabitat conditions.

A variety of other habitat types, includinghardwood dominated forests, wetlands,meadows, and small patch cuts add to thediversity of local landscapes and increasethe diversity and abundance of prey withinthe core habitat area.

Partial harvesting can occur in as muchas 30 percent of the core habitat area,provided it retains 50 percent of theoriginal conifer basal area and canopyclosure of at least 50 percent.

Partial harvesting promotes structurallydiverse stands and provides some opportu-nity to harvest wood from core areas.

Core areas should be connected byriparian reserves and other unharvestedforest. While connections need not becontinuous, gaps of open habitat more

12

The role of coarse woody debris in almostall aspects of marten ecology warrantsspecial consideration of this element inmanagement practices. Harvesting sys-tems that leave slash at the stump, ratherthan at roadside, and site preparationtechniques that minimize slash removaland alignment are preferred options.

5.0 Application of the Guidelines

It is mandatory that the marten guidelinesbe applied in boreal forest conditions inOntario. In the area of transition betweenthe boreal and Great Lakes St. Lawrenceforest, managers must make a choicebetween applying the marten or thepileated woodpecker guidelines. In thissituation, although managers may chooseto apply both guidelines, it is only manda-tory to apply one or the other, not both.The choice of which guideline to applyshould be based on an assessment of thehabitat capability and suitability for themanagement unit and in consideration ofany ecoregional direction.

The following illustrates the necessarysteps that a management team needs tofollow to ensure reasonable considerationof the guidelines and to perpetuate healthymarten populations.

Step 1 . The first step in applying theguidelines is to set preliminary objectivesfor marten populations and habitat withinthe planning unit. The context for settingthese objectives is found within existingland use plans, regional and provincialpolicy initiatives, and more specifically theresults of an eco-regional analysis of habi-tat supply for marten.

Step 2 . The planning team should thendetermine the total area within the planningunit that is capable of providing suitable

habitat for marten at some time during theplanning horizon. This determination ofproduction capability for marten is a criticalstep in formulating management alterna-tives, and may also provide rationale torevisit and alter the objectives decidedupon in Step 1.

Step 3 . At this point, the team can beginthe iterative process of developing andtesting forest management alternatives todetermine their effects on sustainability.The planning team should conduct analy-ses to determine the impacts of each alter-native on the availability of suitable martenhabitat, the sustainable flow of fibre andother forest products, and the sustainabilityof other forest values.

If the Strategic Forest Management Model(SFMM) is being used the model can beconstrained to find solutions that maintainat least 10 - 20 percent of the capableforest types in suitable conditions for mar-ten (i.e. >80 years). A number of modelruns should be conducted to determine theimpacts on all objectives of maintainingvarying proportions between 10 and 20percent of those forest types in suitable ageclasses for marten habitat.

Step 4 . Once the planning team hasdecided on the habitat supply scenario,they must determine the location andspatial arrangement of core habitat areasfor marten for each 20 year period within a100 year planning horizon. Special consid-eration should be given to the connectionsbetween core areas when they are beingallocated. The use of riparian reserves,younger stands, and areas scheduled forpartial harvesting should be considered aspotential corridors.

Step 5 . Any core area where operationsare proposed in the next 5 year periodmust be evaluated and planned in detail.

13

The planning team should ensure thatstands within the core habitat are indeedsuitable, and that in total, they compriseapproximately 75 percent of the core area.This detailed analysis provides the opportu-nity to adjust core area boundaries toaccommodate as many objectives aspossible.

Any proposed harvest treatments sched-uled for the core area must be laid out andidentified as part of an Area of Concern(AOC), with an appropriate prescription.

Step 6 . Silvicultural ground rules shouldbe modified within the bounds of the Occu-pational Health and Safety Act (RSO 1990),to reflect the requirements to retain stand-ing dead trees, live residual trees, andcoarse woody debris.

Appendix 2 contains an example of how toapply the guidelines.

6.0 Future Directions

This first version of the marten guidelineswill be revised when improved scientificunderstanding and/or new methods aredeveloped to more accurately forecastchanges in habitat quantity and quality.Future versions of the guidelines will alsobe consistent with, and complimentary toMNR’s proposed ecological land use plans.

6.1 Habitat Suppl y Forecasting

The need to consider long time frames andlarge geographic areas in applying theseguidelines highlights the need for habitatsupply modelling.

Eco-regional analyses of habitat supply willprovide a context for forest managementunit level decisions. Forecasts of habitatsupply will be used to guide decisions for

minimizing potential adverse effects on thefuture quantity and quality of habitat.

Preliminary estimates of habitat suitability(D’Eon and Watt 1994) have been linked tothe Strategic Forest Management Model(SFMM) and non-spatial estimates ofhabitat availability can be forecasted at theeco-regional and forest management unitlevel.

However, information about the spatialarrangement and distribution of varioushabitats, which non-spatial models cannotprovide, is required to understand theimplications for wildlife populations.

Naylor et al. (1994) have developed an HSImodel for marten in central Ontario, andhave linked that model to a forest simulatorto project marten habitat suitability underdiffering management strategies. Thismodel provides spatial estimates of habitatavailability at the coarse resolution ofindividual traplines or mapsheets.

Prototype models have been developed toprovide stand level spatial resolution.McCallum (1993) linked an HSI model to aspatial forest simulator and generatedspatially explicit forecasts of marten habitatquality. The model is still under develop-ment.

Ultimately it will be necessary to translateforecasts of habitat change into forecasts ofpopulation change. Reliable forecasts ofpopulation change are difficult becausethere is not necessarily a one-to-one corre-spondence between habitat change andpopulation change.

A population viability assessment (PVA)model is being investigated for use inOntario. In combination with a spatial forestsimulator this model would allow managersto spatially assess the viability (likelihood of

14

persistence) and the abundance of martenpopulations across managed forest land-scapes.

A number of knowledge gaps limit ourability to use these more sophisticatedmodelling techniques. Emergent propertiesof landscapes which influence martenhabitat use are not completely understood.PVA models that require estimates ofpopulation parameters such as age specificmortality rates, reproductive rates, anddispersal success have not been testedacross Ontario. Finally, the impacts ofindividual silvicultural treatments on compo-nents of marten habitat are only qualita-tively understood.

Improved scientific understanding of habitatand population interactions, and improvedmethods for applying that knowledgethrough spatial models will be incorporatedinto future versions of these guidelines.These improvements will provide a morerealistic portrayal of the expected effects ofhabitat change on the status of martenpopulations.

6.2 Ecological Land Use Planning

The future vision of land use planning inOntario consists of ecological land useplans for large, ecologically based planningareas, accompanied by operational plan-ning for smaller areas and managementunits.

Ecological land use planning will:(i) integrate direction from relevant provin-cial policies, (ii) establish broad objectivesand management standards for the keynatural resources within the planning area,and (iii) allocate land and natural resourcesamong competing uses. These ecologicalplans will provide a clear basis for opera-tional planning at the local forest manage-ment unit level.

Ecological land use plans will provide acontext to ensure that local operationalplanning decisions will contribute to sus-tainable resource use and the conservationof biodiversity.

An important component of ecological landuse plans will be to establish objectives forthe desired future forest condition. Theseobjectives will be based, at least in part onthe landscape and stand conditions ex-pected under natural disturbance regimes.Future versions of these guidelines willreflect this change.

15

Acknowledgements

It is a great pleasure to thank the followingindividuals for their contributions and sup-port in the development of these guidelines.

Ian Thompson reviewed these guidelinesand provided important insights into martenecology and management.

Ian McCallum and Geoff Bowman wereinstrumental in compiling and synthesizingthe body of scientific literature upon whichthese guidelines are based.

John Boos and Doug Riordan prepared theexample application and map in Appendix2.

Juan Laguna and Dave Baldwin providedinvaluable assistance in preparing thegraphics for this document.

Al Barauskas, in addition to his key role inediting the document, provided significanthelp in understanding the role of theseguidelines in future land use planning.

Brian Blomme contributed editorial com-ments and Sherry Kozak was instrumentalin the design and production of the guide-lines.

16

Literature Cited

Allen, A.W. 1982. Habitat suitability indexmodels: marten. US Fish Wildl. Serv. FWS/OBS-82/10.11. 9 p.

Archibald, W.R. and R.H. Jessup. 1984.Population dynamics of the pine marten(Martes americana) in the Yukon Territory.p. 81-97. in Olsen, R., R. Hastings, F.Geddes. comps., eds. Northern ecologyand resource management. Edmonton,Alberta. University of Alberta Press.

Bowman, J.C., J.F. Robitaille and W.R.Watt. In press. Forest Ecosystem Classifi-cation: a tool for managing marten habitat?(accepted For Chron.)

Burnett, G.W. 1981. Movements andhabitat use of American marten in GlacierNational Park, Montana. University ofMontana. MSc Thesis. 130 p.

Buskirk, S.W. 1983. The ecology of mar-ten in south-central Alaska. FairbanksAlaska. Univ. Alaska. PhD Thesis. 131 p.

Buskirk, S.W. 1984. Seasonal use ofresting sites by marten in south centralAlaska. J. Wildl. Manage. 48: 950-953.

Buskirk, S.W., S.C. Forrest, M.G. Raphaeland H.J. Harlow. 1989. Winter resting siteecology of marten in the Central RockyMountains. J. Wildl. Manage. 53: 191-196.

Buskirk, S.W. and S.O. MacDonald. 1984.Seasonal food habits of marten in south-central Alaska. Can. J. Zool. 62: 944-950.

Buskirk, S.W. and L.F. Ruggiero. 1994.American marten. p. 7-37. in L.F.Ruggiero, K.B. Aubry, S.W. Buskirk, L.J.Lyon, W.J. Zielinski Tech. eds. The scien-tific basis for conserving forest carnivores;American marten, fisher, lynx, and wolver-ine in the western United States. USDAFor. Serv. Gen Tech. Rep. RM-254.

Buskirk, S.W. and Powell, R.A. 1994.Habitat ecology of fishers and Americanmartens. p. 283-296. in Buskirk, S.W.,A.S. Harestad and M.G. Raphael. comps.,eds. Martens, sables and fishers: biologyand conservation. Ithica, N.Y. CornellUniversity Press.

Campbell, T.M. 1979. Short-term effects oftimber harvests on pine marten ecology.Fort Collins, CO. Colorado State Univer-sity. MSc Thesis. 71 p.

Courtin, G.M. and Beckerton, J. 1996.Does the practice of careful cutting main-tain biodiversity? presented at the NationalModel Forest Network Meeting. Cochrane,Ontario. Feb. 1996.

D’Eon, R.G. and W.R. Watt. 1994. A foresthabitat suitability matrix for northeasternOntario. OMNR Northeast Science &Technology. Tech. Manual TM004. 83 p.

de Vos, A. 1952. Ecology and manage-ment of fisher and marten in Ontario.Toronto, Ont. Ontario Dept Lands. For.90 p.

Francis, G.R. and A.B. Stephenson. 1972.Marten ranges and food habits inAlgonquin Park, Ontario. Ont. Min. Nat.Res. Rpt. No. 91.

Gunderson, H.L. 1959. Red-backed volehabitat studies in central Minnesota. J.Mammal. 40: 405-412.

17

Hargis. C.D. and D.R. McCullough. 1984Winter diet and habitat selection of martenin Yosemite National Park. J. Wildl. Man-age. 48: 140-146.

Hawley, V.D. and F.E. Newby. 1957. Mar-ten home ranges and population fluctua-tions in Montana. J. Mammal. 38: 174-184.

Hearnden, K.W., S.V. Millson, and W.C.Wilson. 1992. A report on the status offorest regeneration. The Ontario Indepen-dent Forest Audit Committee. Queen’sPrinter, Toronto. 117 p.

Hodgman, T.P , D.J. Harrison, D.D. Katnik,K.D. Elowe. 1994. Survival in an inten-sively trapped marten population in Maine.J. Wildl. Manage. 58: 593-600.

Hunter, M.L., Jr. 1990. Wildlife, forestsand forestry. Englewood Cliffs, N.J.Prentice Hall. 370 p.

Jonkel, C.J. and R.P. Weckwerth. 1963.Sexual maturity and implantation of blasto-cysts in the wild pine marten. J. Wildl.Manage. 27: 93-98.

Koehler, G.M. 1990. Demographic andhabitat characteristics of lynx and snow-shoe hares in north-central Washington.Can. J. Zool. 68: 845-851.

Koehler, G.M. and M.G. Hornocker. 1977.Fire effects on marten habitat in the Selway- Bitterroot Wilderness. J. Wildl. Manage.41: 500-505.

Koehler, G.M., W.R. Moore and A.R. Taylor.1975. Preserving the pine marten: man-agement guidelines for western forests.Western Wildlands 2: 31-36.

Litvaitis, J.A., J.A. Sherburne and J.A.Bissonette. 1985. Influence of understoreycharacteristics on snowshoe hare habitatuse and density. J. Wildl. Manage. 49:866-873.

Lofroth, E.C. and J.D. Steventon. 1990.Managing for marten winter habitat ininterior forests of British Columbia. p. 67-75. in Chambers, A. ed. Symposiumproceedings, Wildlife forestry symposium: aworkshop on resource integration for wild-life and forest managers. Prince George,B.C. Forestry Canada.

Martell, A.M. 1983. Changes in smallmammal communities after logging innorth-central Ontario. Can. J. Zool. 61:970-980.

Martell, A.M. and A. Radvanyi. 1977.Changes in small mammal communitiesafter clearcutting of northern black spruceforest. Can. Field Nat. 91: 41-46.

Martin, S.K. 1994. Feeding ecology ofAmerican martens and fishers . p. 297-315in Buskirk, S.W., A.S. Harestad and M.G.Raphael. comps., eds. Martens, sables andfishers: biology and conservation. Ithica,N.Y. Cornell University Press.

Martin, S.K. and R.H. Barrett. 1983. Theimportance of snags to pine marten habitatin the northern Sierra Nevada. p. 114-116.in Davis, J.W., G.A. Goodwin, R.A.Ockenfels. Tech. coords. Proceedings ofthe symposium on snag habitat manage-ment. USDA For. Serv. Gen. Tech. Rep.RM-99.

McCallum, R.I. 1993. Long term effects oftimber management on marten (Martesamericana). Habitat Potential in an OntarioBoreal Forest, MSc. Thesis. LakeheadUniversity. 190 p.

18

McCarthy, T.G., R.W. Arnup, J. Nieppola,B.G. Merchant, K. Taylor, and W.J. Parton.1995. Field guide to forest ecosystems ofnortheastern Ontario. OMNR, NortheastScience & Technology. Field Guide FG001.

Mead, R.A. 1994. Reproduction in mar-tens and fishers. p. 404-422. in Buskirk,S.W., A.S. Harestad and M.G. Raphael.comps., eds. Martens, sables and fishers:biology and conservation. Ithica, N.Y.Cornell University Press.

Naylor, B., J.Baker, D.Hogg, J.McNicol, andW.R. Watt. 1996. Forest managementguidelines for the provision of pileatedwoodpecker habitat. OMNR. (in press)

Naylor, B., T. Bellhouse, S. Christilaw andP. Weilandt. 1994. Development andvalidation of a habitat suitability indexmodel for the American marten in centralOntario. OMNR Central Region Scienceand Technology. Tech. Rep No. 29. 13 p.

Racey, G.D., T.S. Whitfield, and R.A. Sims.1989. Northwestern Ontario forest ecosys-tem interpretations. Thunder Bay, Ont.OMNR Tech. Rep. No. 46. 990 p.

Raine, R.M. 1981. Winter food habits,responses to snow cover, and movementsof fisher (Martes pennanti) and marten(Martes americana) in southeasternManitoba. Winnipeg, Manitoba. Universityof Manitoba. MSc Thesis. 144 p.

Raine, R.M. 1987. Winter food habits andforaging behaviour of fishers (Martespennanti) and martens (Martes americana)in southeastern Manitoba. Can. J. Zool.65: 745-747.

Raphael, M.G. 1988. Habitat associationsof small mammals in a subalpine forest,southeastern Wyoming. p. 359-367. inSzaro, R.C., K.E. Severson, and D.R.Patton. eds. Symposium Proceedings:management of amphibians, reptiles andsmall mammals in North America. USDAFor. Serv. Gen. Tech. Rep. RM-166.

Sherburne, S.S. and J.A. Bissonette. 1994.Marten subnivean access point use: re-sponse to subnivean prey levels. J. Wildl.Manage. 58: 400-405.

Simon, T.L. 1980. An ecological study ofthe marten in the Tahoe National Forest,California Sacremento, CA. CaliforniaState University. MSc Thesis. 143 p.

Sims, R.A., W.D. Towill, K.A. Baldwin, andG.M. Wickware. 1989. Field guide toecosystem classification for northwesternOntario. Toronto, Ont. OMNR. 191 p.

Sims, H.P. and C.H. Buckner. 1973. Theeffect of clearcutting and burning of Pinusbanksiana forests on the populations ofsmall mammals in southeastern Manitoba.Am. Midl. Nat. 90: 228-231.

Snyder, J.E. and J.A. Bissonette. 1987.Marten use of clearcuttings and residualforest stands in western Newfoundland.Can. J. Zool. 65: 169-174.

Soutiere, E.C. 1979. Effects of timberharvesting on marten in Maine. J. Wildl.Manage. 43: 850-860.

Spencer, W.D. 1981. Pine marten habitatpreferences at Sagehen Creek, California.MS Thesis. Univ. California, Berkely. C.A.

Spencer, W.D., R.H. Barrett, and W,J.Zielinski. 1983. Marten habitat prefer-ences in the northern Sierra Nevada. J.Wildl. Manage. 47: 1181-1186.

19

Steventon, J.D. 1979. Influence of timberharvesting upon winter habitat use bymarten. Orono ME. School of WildlifeManagement, University of Maine at Orono,M.A. Thesis. 24 p.

Steventon, J.D. and J.T. Major. 1982.Marten use of habitat in commercially clear-cut forest. J. Wildl. Manage. 46:175-182.

Strickland, M.A. and C.W. Douglas. 1987.Marten. p. 531-546 in Novak, M., J.A.Baker, M.E. Obbard, and B. Malloch, eds. Wild furbearermanagement and conservation in NorthAmerica. Ont. Trappers. Assoc. North BayOnt.

Taylor, M.E. and N. Abrey. 1982. Marten(Martes americana) movements and habitatuse in Algonquin Provincial Park, Ontario.Can. Field Nat. 96: 439-447.

Thompson, I.D. 1986. Diet choice,hunting behaviour, activity patterns, andecological energetics of marten in naturaland logged areas. Kingston, Ont. QueensUniversity PhD Diss. 179 p.

Thompson, I.D. 1991. Could martenbecome the spotted owl of easternCanada? For. Chron. 67: 136-140.

Thompson, I.D. 1994. Marten populationsin uncut and logged boreal forests inOntario. J. Wildl. Manage. 58: 272-280.

Thompson, I.D. and P.W. Colgan. 1987.Numerical responses of martens to a foodshortage in northcentral Ontario. J. Wildl.Manage. 51: 824-835.

Thompson, I.D. and P.W. Colgan. 1990.Prey choice by marten during a decline inprey abundance. Oecologia 83: 443-451.

Thompson, I.D. and P.W. Colgan. 1994.Marten activity in uncut and logged borealforests in Ontario. J. Wildl. Manage. 58:280-288.

Thompson, I.D. and A.S. Harestad. 1994.Effects of logging on American martenswith models for habitat management. p.3355-367. in Buskirk, S.W., A.S. Harestadand M.G. Raphael. comps., eds. Martens,sables and fishers: biology and conserva-tion. Ithica, N.Y. Cornell University Press.

Van Wagner, C.E. 1978. Age-class distri-butions and the forest fire cycle. Can. J.For. Res. 8: 220-227.

Ward, P.C. and A.G. Tithecott. 1993.The impact of fire management on theboreal landscape of Ontario. OMNR.AFFMB Publication No. 305. 12 p.

Watt, W.R. 1996. Managing for snags inboreal forests of northeastern Ontario.Northeast Science and Technology Techni-cal Manual (in press)

Weckwerth, R.P. and V.D. Hawley. 1962.Marten food habits and population fluctua-tions in Montana. J. Wildl. Manage. 26: 55-74.

Wiens, J.A. 1985. Vertebrate responses toenvironmental patchiness in arid and semi-arid ecosystems. p. 169-193 in Pickett,S.T.A. and P.S. White. eds. The ecology ofnatural disturbance and patch dynamics.San Diego, CA. Academic Press

Wynne, K.M. and J.A. Sherburne. 1984.Summer home range use by adult martenin northwestern Maine. Can. J. Zool. 62:941-943.

Yeager, L.E. 1950. Implications of someharvest and habitat factors on pine martenmanagement. Trans. N.A. Wildl. Conf. 15:319-334.

20

Appendix 1. Forest ecosystem classification interpretations formarten habitat suitability for northeastern and

northwestern Ontario.

Suitability NW Ontario NE Ontario NE OntarioVegetation Types Site Types Vegetation Types

Good 6, 7, 14, 15, 16, 19, 5A, 5B, 6A, 6B, 8, 6, 7, 8, 9, 19, 2020, 21, 24, 25, 31 9, 13 21

Fair 1, 4, 8, 9, 10, 11, 3A, 3B, 4, 6C, 11, 12, 13, 14, 18, 2222, 32, 33, 34 12

Poor 2, 3, 5, 12, 13, 17, 1,2A, 2B, 7A, 7B, 1, 2, 3, 4, 5, 10,18, 23, 26, 27, 28, 10, 14 11, 15, 16, 17,29, 30, 35, 36, 37, 23, 24, 25, 2638

21

The example forest is situated in the transi-tional forest between the clay belt and theChapleau Plains area in northeasternOntario. This area has historically producedhigh marten densities and harvests, andcontributes to the livelihood of 10 registeredtrappers.

The forest Management Unit (MU) has ahistory of logging dating back at least 25years and supports a number of forestindustries, most notably a lumber mill andan oriented strand board mill. Mill demandsfor wood exceed local supply, and woodsupply for the Region has been describedas tight but manageable.

The total forested area of the example FMUis 400,000 ha. However, only 285,000 ha isconsidered to have the “capability” to de-velop the mesic conifer dominated forestconditions used by marten. The remaining115,000 ha of forested land is comprised ofjack pine sand flats, upland dominated byrelatively pure aspen communities, andscattered black spruce bogs. None of theseconditions are likely to develop into amesic conifer-dominated condition withouta high level of management intervention.

The Strategic Forest Management Model(SFMM) was used to evaluate the implica-tions, and determine the sustainability, ofmaintaining suitable marten habitat on thelandscape throughout the planning horizonat three levels: (i) a minimum of 28,500 haof suitable habitat (10 percent of capablehabitat), (ii) 42,750 ha of suitable habitat(15 percent level), and (iii) 57,000 ha ofsuitable habitat (20 percent level).

The results of SFMM analyses indicatedthat maintaining an even flow of more than28,500 ha of suitable marten habitat, had

Appendix 2. Applying the Guidelines: An Example Forest

serious wood supply implications, and thatwood supply deficits in the 40 - 60 yeartime period were unacceptable. A follow upanalysis was conducted to determine whatmix of habitat levels might be sustainedgiven the socio-economic concerns identi-fied.

The planning team was able to identify thefollowing option which provided more thanminimal amounts of habitat in most timeperiods, and met the other plan objectives,including the wood fibre demands:

Years 0 - 20 57,000 ha (20 percent)Years 21 - 40 42,750 ha (15 percent)Years 41 - 60 28,500 ha (10 percent)Years 61 - 80 42,750 ha (15 percent)Years 81 + 57,000 ha (20 percent)

Having identified the time period specificobjectives for suitable marten habitat, theplanning team is faced with the difficult taskof identifying where and how the areatargets will be apportioned to core habitatareas.

A thematic map of the management unitshould be created, depicting the location ofall “capable habitat” conditions, and identi-fying the current age of the stands in 20year age classes. The planning team canuse this map to identify specific geographicareas where potential exists to establishcore habitat areas in each 20 year planningperiod. The selection of specific core areasshould include consideration for multiplevalues wherever possible. For example,core habitat areas for marten could alsosatisfy some of the late winter habitatrequirements of moose. Selection of areasin proximity to high quality moose summer/fall habitat, known aquatic feeding areasand calving sites should be considered.

22

Riparian reserves could also contribute tocore habitat areas in many locations.

In the case of the 41 - 60 year time period,the planning team would look for between 6and 10 potential core area sites of 3,000 to5,000 ha each. Stands that were currentlybetween 40 and 80 years of age wouldform the bulk of these potential core areas.In 40 years, those stands would be be-tween 80 and 120 years of age and suit-able for marten.

Once the planning team has identified thelocations of present and core habitat areas,the planning team concentrates on devel-oping detailed area prescriptions for anyproposed operations within core habitatareas that might be affected in the next 5year planning period. The accompanyingmap illustrates one such core habitat areawhere operations are proposed.

The core habitat area covers 3,437 ha inthe southeast corner of the base map. Thearea falls along the slope of an esker wheremixed coniferous -deciduous forest prolifer-ates. FEC types (5b, 8,9,12) are abundanton the lower slopes and make up the bulkof the core habitat areas. Many of the FRIworking group designations are to speciesother than spruce, fir, or cedar, but FRIattributes confirm that 40 percent of thespecies composition within those stands iseither spruce, fir or cedar.

The age of most stands within the corehabitat area fall between 80 and 90 years,and the planning team has identified thatthe area will contribute to marten habitatrequirements for the next two 20-yearplanning periods. Within the core habitatarea, 24 percent is comprised of maturejack pine and aspen communities that areunsuitable. Those areas have been pro-posed for harvest within the immediate 5year operating plan. In both cases, group

seed tree harvests are proposed to facili-tate regeneration and provide future cavitytrees. In addition, partial harvesting of 120ha of spruce-pine and spruce-aspen standsis proposed adjacent to the pine and aspenclear cuts. The partial harvesting will bedesigned to remove 30 percent of the basalarea, concentrating on merchantable jackpine and aspen stems.

Regeneration of the clear cuts will followthe existing silvicultural ground rules, andthe partial harvest blocks will receive nofurther treatment until the spruce compo-nent is harvested, sometime after 40 years.

23

(MAP - Bob Watt to provide a revised version)

24