maine wildlands network visionforeman, john gallo, steve gatewood, harvey locke, chris mcgrory...

MAINE WILDLANDS NETWORKVISION

A Scientific Approach to

Conservation Planning in Maine

WILDLANDS PROJECT

WE ARE AMBITIOUS. We live for the day

when grizzlies in Chihuahua have an unbroken

connection to grizzlies in Alaska; when wolf

populations are restored from Mexico to the

Yukon to Maine; when vast forests and flowing

prairies again thrive and support their full range

of native plants and animals; when humans dwell

on the land with respect, humility, and affection.

Toward this end, the Wildlands Project is working

to restore and protect the natural heritage of

North America. Through advocacy, education,

scientific consultation, and cooperation with

many partners, we are designing and helping

create systems of interconnected wilderness

areas that can sustain the diversity of life.

June 2002

Robert Long

Paula MacKay

Conrad Reining

Barbara Dugelby

Kathy Daly

For additiona l information contact

Wildlands ProjectP. O . Box 455Richmond, VT 05477802/434-4077info@w ildlandsproject.orgwww.w ildlandsproject.org

© 2002 by the Wildlands Project. Cover illustration © Dennis Logsdon.

MAINE WILDLAN DS NETWORKVISION

A Scientific Approach to Conservation Planning in Maine

reconnect restore rew i ldWILDLANDS PROJECT

ACKNOWLEDGMENTS

A great many people made the Maine Wildlands Network Vision possible. We

would first like to thank those who wrote sections of the document or otherw ise

made significant contributions to it*: Julia Ambagis, Judy Bond, Karen Beazley,

Jonathan Carter, Todd Cumm ings, Kristin DeBoer, Kathleen Fitzgerald, M ichael

Kellett, Tom Lee, Peter Quinby, Jym St. Pierre, Cheryl Veary, Peter Williams, and

George Wuerthner. We would also like to thank the many people who reviewed

drafts of this plan: Tom Butler, John Davis, Bart DeWolf, M ike D iNunzio, Dave

Foreman, John Gallo, Steve Gatewood, Harvey Locke, Chris McGrory Klyza,

Leanne Klyza Linck, Reed Noss, David Publicover, Peter Quinby, Justina Ray,

M ichael Soulé, Steve Trombulak, Ken Vance-Borland, and George Wuerthner.

This project was also greatly influenced by the products and work of many

other organizations. We would especially like to thank Sky Island Alliance,

Conservation Science, Inc., Conservation Biology Institute, BC Wild, and the

Southern Rockies Ecosystem Project for their valuable contributions. We would

also like to thank the Wildlands Project’s board members, Greater Laurentian

Wildlands Project Advisory Comm ittee members, and the participants in a

Maine Wildlands Network design workshop held at the University of Maine in

March, 1998, all of whom contributed directly or indirectly to this proposal.

Finally we w ish to thank the many organizations that provided monetary

support for this work including Ben & Jerry’s Foundation, Boston Foundation,

EnTrust Fund, Environmental Systems Research Institute, Inc., Fine Fam ily

Foundation, Foundation For Deep Ecology, Geraldine R. Dodge Foundation,

Kendall Foundation, Norcross Wildlife Foundation, Orchard Foundation,

Patagonia, Rockwood Fund, and Sweet Water Trust.

* Contributors do not necessarily endorse the views or opinions expressed throughout the document. Many contributors were solicited to write specific sections of the document, but were not involved in other sections.

WILDLANDS PROJECT MAINE WILDLANDS NETWORK VISION

Table of Contents

EXECUTIVE SUMMARY MAINE WILDLANDS NETWORK VISION.................................................................................1I. INTRODUCTION AND BACKGROUND .............................................................................................................................3

Wildlands Network Design: From Science to On-The-Ground Protection..................................................................................3The Natural Heritage of Maine ....................................................................................................................................................4Forest Management History .........................................................................................................................................................7The “Mega-easement” Phenomenon ................................................................................................. ............................................8The Status of Conservation in Maine ............................................................................................................................................8

II. ECOLOGICAL WOUNDS ....................................................................................................................................................13III. MISSION, GOALS, AND OBJECTIVES.............................................................................................................................24

Mission of the Maine Wildlands Network..................................................................................................................................25Goals of the Maine Wildlands Network .....................................................................................................................................25Objectives of the Maine Wildlands Network .............................................................................................................................25

IV. WILDLANDS NETWORK DESIGN APPROACH ...........................................................................................................27Rewilding...................................................................................................................... .............................................................27Core, Linkage, Compatible-Use Area Model...............................................................................................................................28Core Wild Areas .........................................................................................................................................................................28Wildlife Linkages .......................................................................................................................................................................29Study Areas.................................................................................................................................................................................30Three-track Wildlands Network Design Methodology ..............................................................................................................30Assumptions and Limitations .....................................................................................................................................................31

V. FOCAL SPECIES PLANNING...............................................................................................................................................34Focal Species Categories for the Maine Wildlands Network.......................................................................................................34Focal Species Selection................................................................................................................................................................35Summary of Profiled Species.......................................................................................................................................................37

VI. CONSERVATION AREA SELECTION PROCESS ...........................................................................................................38The SITES Selection Algorithm .................................................................................................................................................38Model Output.............................................................................................................................................................................39Data Layers .................................................................................................................................................................................40Element Target Levels for SITES Analysis..................................................................................................................................48SITES Selection Results..............................................................................................................................................................48

VII. WILDLANDS NETWORK DESIGN ..................................................................................................................................53Core Wild Area Delineation.......................................................................................................................................................54Wildlife Linkage Delineation .....................................................................................................................................................54

VIII. NETWORK DESIGN SUMMARY AND EVALUATION ..............................................................................................56Network Summary and Spatial Statistics....................................................................................................................................56Focal Species and Special Elements.............................................................................................................................................62Vegetation and Land Cover.........................................................................................................................................................63Biophysical Regions ...................................................................................................................................................................65Major Watersheds.......................................................................................................................................................................67Elevations ..................................................................................................................... ..............................................................69Conclusion..................................................................................................................... .............................................................71

IX. CONSERVATION ACTION.................................................................................................................................................72Overview ....................................................................................................................... .............................................................72Conservation Action Strategy .....................................................................................................................................................73Conclusion..................................................................................................................... .............................................................82

X. CONCLUSION .......................................................................................................................................................................83REFERENCES ..................................................................................................................... ......................................................85APPENDIX 1 – Participants, Recommendations, and Key Points from March 1998 Maine Wildlands Network Workshop Participants ............................................................................................98APPENDIX 2 – Compatible Conservation Efforts.....................................................................................................................99APPENDIX 3 – Focal Species Profiles .....................................................................................................................................105


List of TablesTable I.1 Maine physical statistics.................................................................................................................................................... 5Table I.2 Comparison of known number of species in Maine with numbers in North America north of Mexico ............................. 7Table I.3 The Conservation Status of Maine by GAP Status Levels. ................................................................................................. 9Table II.1 Estimated or known numbers of species in various taxonomic groups and their status in Maine................................... 16Table II.2 Plants believed to have been extirpated from the state, and those that have not been documented in Maine in 20 years........................................................................................................ 17Table V.1 Species scoring highest in the Maine Focal Species Survey.*.......................................................................................... 36Table V.2 Focal species profiled in the Maine Wildlands Network Vision. ................................................................................... 37Table VI.1 Road classes in Maine. .................................................................................................................................................. 39Table VI.2 Focal species data used in SITES analysis...................................................................................................................... 42Table VI.3 Relationship of the 37 GAP-ME Land Use Land Cover Types to the collapsed the land cover classes used in this analysis. ....................................................................................................... 45Table VI.4 Digital data layers used in the SITES analysis. ............................................................................................................. 47Table VI.5 Element target values used for the SITES analysis ........................................................................................................ 50Table VII.1 Digital data layers used in the Maine Wildlands Network Design process. ................................................................ 53Table VIII.1 Area and percentage of state represented by proposed core wild areas, aquatic linkages, aquatic buffers, and study areas in the Maine Wildlands Network together with existing conservation lands ....................................... 56Table VIII.2 General relationship of Wildlands Network Components to GAP status levels.1 ...................................................... 57Table VIII.3 Focal species and special element conservation goals included in different protection alternatives, in hectares. ....... 63Table VIII.4 Land use land cover types included under different protection alternatives, in hectares............................................. 65Table VIII.5 Ecoregions of Maine included under different protection alternatives, in hectares..................................................... 67Table VIII.6 Major watersheds included under various protection alternatives, in hectares. .......................................................... 69Table VIII.7 Representation of elevation categories under various protection alternatives, in hectares. ......................................... 70Table IX.1 Conservation actions components and operational priority........................................................................................... 75


List of FiguresFigure I.1 Population density in Maine in 2000............................................................................................................................... 6Figure I.2 Maine’s major cities, roads, rivers, and lakes, GAP class 1 and 2 protected lands, large (>5000 ha) class 3 lands containing easements, and other managed conservation lands. ............................................................ 11Figure II.1 Centers of human population and human population density by town......................................................................... 19Figure II.2 Roads in Maine (by class).............................................................................................................................................. 19Figure II.3 Past, current, and projected development in southern Maine. ...................................................................................... 20Figure IV.1 Components of a Wildlands Network......................................................................................................................... 28Figure VI.1 Focal species habitat included as inputs in SITES 1.0 analysis .................................................................................... 41Figure VI.2 Schematic of SITES analysis ........................................................................................................................................ 43Figure VI.3 SITES planning units showing units locked in and out............................................................................................... 43Figure VI.4 Special elements included as inputs ............................................................................................................................ 44Figure VI.5 Land cover classes used in SITES analysis.................................................................................................................... 46Figure VI.6 Sum Runs results from SITES analysis ........................................................................................................................ 49Figure VI.7 Sum Runs alternatives one and two............................................................................................................................. 51Figure VI.8 Sum Runs alternatives three and four.......................................................................................................................... 52Figure VII.1 Subwatersheds of major rivers and rivers containing runs of native Atlantic salmon. ................................................ 55Figure VIII.1 Proposed Maine Wildlands Network ...................................................................................................................... 58Figure VIII.2 Two elements assessed for representation within the proposed Maine Wildlands Network: major watersheds and elevation classes........................................................................................... 60Figure VIII.3 Biophysical regions (Maine ecoregions) assessed for representation within the proposed Maine Wildlands Network. ...................................................................................................................... 61Figure VIII.4 Percentage of focal species and special element conservation goals included in different protection alternatives...... 63Figure VIII.5 Percentage of the land use land cover types included under different protection alternatives................................... 64Figure VIII.6 Percentage of ecoregions of Maine included under different protection alternatives................................................. 66Figure VIII.7 Percentage of major watersheds included under various protection alternatives (e.g., core areas of proposed Wildlands Network encompass 85.5% of the Upper Androscoggin watershed). ................................................... 68Figure VIII.8 Representation of elevation categories under various protection alternatives (e.g., core areas of the proposed wildlands network include 15.6% of all lands having elevations of between 1 and 200 meters). ................................ 70

WILDLANDS PROJECT MAINE WILDLANDS NETWORK VISION 1

EXECUTIVE SUMMARYMAINE WILDLANDS NETWORK VISIONA Bold Vision

What would it take for wolves, salmon, cougars, andnorthern goshawks to thrive in the Maine woods? Whatconservation lands are needed to sustain all species in theland community? What natural processes that create andshape diverse habitats must be restored and allowed tooperate freely? If, in the coming decades, conservationaction in Maine focused on creating a network ofconservation areas that could protect the state’s rich naturalheritage for generations to come, what would that systemlook like?

The Wildlands Project, an international member-supported conservation organization, posed these questionswhen drafting a vision for land protection in Maine. Aslarge as the other five New England states combined,Maine’s 20 million acres encompass the dramatic Atlanticcoastline, thousands of lakes, rivers, streams, and wetlands,and the most heavily forested landscape in the nation. Thenorthern part of the state represents the largest block ofundeveloped, primarily unpopulated land east of theMississippi. Maine presents an extraordinary opportunityfor wilderness recovery.

Maine’s natural legacy has been severely compromised,however, by industrial forestry, development, anddisruption of natural processes. While scientists and somepolicymakers recognize these threats, strategies to restoreand protect Maine’s biological diversity have not yet beenimplemented at a scale that comprehensively addresses theecological wounds the land has suffered.

In the mid-1990s, a landmark scientific report (Gawleret al. 1996) noted the inability of Maine’s existingconservation areas to fully protect biodiversity. Thisrealization among scientists, and increasing skepticismabout the dominant “working landscape” model of landconservation that emphasizes continued industrial forestry,prompted the Wildlands Project to begin drafting ablueprint for a system of natural areas in Maine.

This document sets out that blueprint—a proposedMaine Wildlands Network comprised of core wild areasand wildlife linkages designed to protect Maine’secosystems in perpetuity. This wildlands network addressesthe human impacts—or ecological wounds—to the land,and applies the principles of regional and continentalrestoration to begin healing them. The primary goals of the

network are to protect the state’s diverse naturalcommunities, sustain healthy populations of native wildlifethat depend upon those habitats, and restore key naturalprocesses across the landscape.

To determine the size and location of the cores andlinkages, we employed a three-track planning approachthat attempted to represent in the network certainimportant features, including:

• physical and landscape diversity (e.g., biophysicalregions, major watersheds, and vegetation types);

• special elements (e.g., large wetland complexes androadless areas);

• adequate habitat for populations of “focal species”(organisms used in planning and managing naturereserves because their requirements for survivalrepresent factors necessary to maintaining ecologicallyhealthy conditions).

For regional conservation planning, it is useful tochoose focal species that need large areas, are sensitive tohuman disturbance, or are critical ecological actors. If wellchosen, focal species can help measure whether a wildlandsnetwork is truly functional. Without focal species analysis,there is no practical way of knowing how big reservesshould be, where they should be placed on the landscape,and where linkages between protected areas are necessaryfor wildlife movement.

The Wildlands Project emphasizes planning for theneeds of large carnivores and other “keystone” species. (Justas an arch crumbles if its keystone is removed, ecosystemscan unravel when key members of the land community areabsent.) There is persuasive scientific evidence that largecarnivores such as wolves and cougars are vitally importantto healthy ecosystems.

To ensure the long-term viability and functional role ofsuch species, it is necessary to protect and restore big,interconnected wildlands. Because large predators requireextensive space and connectivity, the modeling of theirhabitat requirements is a key tool in network design. Weincorporate the habitat needs of wolves, lynx, and Atlanticsalmon in the proposed Maine Wildlands Network.

I n sh o r t , ou r app r o a c h is to he l p Na t ur e he a l , to put th e p i e c e s of he a l t hy ec o s y s t e m s ba c k to g e t he r —i nc l u d i n gm i s s i ng sp e c i e s and na t u r a l pr o c e s s e s . We ca l l th i s r e w i l d i n g .


ResultsThe resulting wildlands network design includes

approximately 28,725 square kilometers of core wild areas,1,175 km2 of aquatic linkages, and 4,671 km2 of aquaticbuffers. We propose that the vast majority of lands withinthe network be granted a high level of ecological protection(that is, managed for biodiversity, not resource production).This proposed Maine Wildlands Network meets a broad setof conservation objectives, ranging from protecting at least50% of potential lynx and wolf habitat, to more than 30%of each of the major forest vegetation types in Maine.

We recognize that criteria used in this design, such asspecies distributions and restoration of large-scaleecological processes, are more meaningful within a larger,regional context, and are not constrained by politicalboundaries. The Wildlands Project has initiated a regionalwildlands network design process that will stretch from theTug Hill plateau and Adirondacks in north central NewYork, north into Ontario and east to Vermont, NewHampshire, Maine, New Brunswick, and Nova Scotia.Nevertheless, Maine’s size (relative to other northeasternstates), central location within the northernAppalachian/Acadian ecoregion, and large expanses ofunpopulated territory justify the attention that thisproposal represents. This conceptual network alsorepresents a test of the analytical techniques that will beapplied at a much larger scale.

Conservation ActionCan such a bold vision for protecting wilderness and

wildlife in Maine ever become reality? We believe theanswer is yes. The backdrop for all contemporaryconservation planning is a global extinction crisisunparalleled in the past 65 million years. Thisextraordinary moment in time requires us to think larger inspace and time about the needs of wild Nature.

A wildlands network on the scale we envision could becreated incrementally over the coming decades, or evencenturies, if local, regional, and national conservationistsbelieve it is a vision worth working toward. It will requirenetworks of people working to protect networks of land.The Wildlands Project supports democratic, participatoryactions at every scale—from local to national—that helpconserve big, wild, interconnected places throughoutMaine and the greater Northern Appalachians.

High priorities for advancing such efforts includedeveloping a statewide network of individuals andorganizations that share a common vision for large-scalewilderness recovery; establishing an outreach and educationprogram focused on decision-makers, government agencies,land trusts, and local citizens that communicates landscapeecology principles; and advocacy work that champions the

protection of specific places identified in the proposednetwork.

There is justifiable concern that the transformation ofnorthern Maine from a landscape dominated by commercialtimberland to one where large wilderness areaspredominate would harm the rural economy. Clearly, theregion’s economy is changing as global economic forcesaffect commercial forest management and the pulp andpaper industries; the forest products sector cannot becounted upon to support a vibrant rural economy in thefuture. Economic diversification, including job creationbased on the amenities offered by a network of conservedareas, may offer the best development option for ruralMaine.

For any wildlands network to become reality, the veryreal concerns of local citizens, government, and othersectors must be addressed. Indeed, a successful wildlandsnetwork in Maine will demand the full involvement ofcitizens government agencies at the federal, state, and locallevels, and it is these agencies, along with the donorcommunity, private conservation organizations, andacademic institutions that are the prime audience for thisdocument.

Now is the right time to begin making the MaineConservation Vision a reality. Massive land sales ofindustrial forest offer great opportunity for conservation;many of the new owners have different objectives than theold paper and lumber companies and have longer-termplanning horizons, both of which may aid conservation.New emphasis on green certification and value-added forestproducts also signal a potentially positive shift in resourcemanagement towards more ecologically sound forestry.Nevertheless, even timberlands subject to state-of-the-artforestry are no substitute for core wild areas withoutresource extraction, where natural processes areuntrammeled. The chance has never been better to set asidelarge tracts of land to protect and restore ecologicalintegrity, whether through public acquisition, purchase of“forever wild” conservation easements, or othermechanisms.

The vision of the Wildlands Project is one of thrivinghuman communities imbedded in a landscape of diversity,ecological richness, and good health. The challenge is tochart the way ahead on a broad, landscape scale, and theMaine Wildlands Conservation Vision seeks to do just that.This is an ambitious proposal based on science andopportunity; it assumes that political reality is ever-changing, but that ecological reality—Nature’s bottomline—is immutable. Through careful analysis and visionaryplanning, it seeks to answer the fundamental question ofwhat long-term changes are necessary for the land torecover. It offers a chance for natives like the wolf, lynx,and salmon to return and thrive in Maine—to come home.


I. INTRODUCTION AND BACKGROUNDOver the past three decades, scientists have come to the

conclusion that our planet now faces the sixth majorextinction event in the history of life. Earlier massextinctions were the result of a major natural disaster orsignificant climatic changes (Eldridge 1991; Ward 1997).The current extinction wave can be directly linked to ourown behavior. We have altered the air, soil, water, andbiota through millions of individual human acts. As aresult, the earth is a less hospitable place for hundreds ofthousands of species. E.O. Wilson (2001) estimates that:

Before humans existed, the species extinction ratewas (very roughly) one species per million speciesper year (0.0001 percent). Estimates for currentextinction rates range from 100 to 10,000 timesthat, but most hover close to 1,000 timesprehuman levels (0.1 percent per year), with therate projected to rise, and very likely sharply.

In Maine, and elsewhere in the mixed deciduous forestsof northeastern North America, many native species havedeclined, and the wild places in which they once flourishedhave been degraded and fragmented by human activity.These practices include unsustainable forest management,construction of dams, pollution of waterways, introductionof exotic species, poaching and predator control, andsuburban development.

Faced with these kinds of damaging activities on thelandscape, the conservation community often finds itself onthe defensive, reacting to the sales of large tracts offorestland, subdivisions, and new logging roads on anindividual basis. Under these circumstances, it is difficultto craft a positive vision that can garner significant publicsupport. A more strategic, less reactive approach is needed.It is the goal of this document to lay out such an approachfor Maine.

Wildlands Network Design: From Science toOn-The-Ground Protection

A strategic approach depends on clearly articulatedgoals. The overarching goal of the Wildlands Project inMaine is to restore and protect the ecological integrity ofthe state. Noss (1992) and Noss et al. (1997) describe thefour key conservation goals necessary for the establishmentand maintenance of lasting ecological integrity:

• Represent, within the system, all native ecosystemtypes and seral stages across their natural range ofvariation;

• Maintain, within the system, viable populations of allnative species in natural patterns of abundance anddistribution;

• Maintain, within the system, ecological andevolutionary processes, such as disturbance regimes,hydrological processes, nutrient cycles, and bioticinteractions, including predation; and

• Design and manage the system to be responsive toshort-term and long-term environmental change andmaintain the evolutionary potential of lineages.

A protected areas system is essential to achieve thesegoals. The purpose of this document is to propose such asystem for Maine—a “Maine Wildlands Network”—thatwill satisfy the biodiversity conservation goals noted above,as adapted to Maine. The specific goals for the MaineWildlands Network are discussed in detail in Section III.This document also outlines the means by which thenetwork could be established and maintained (Section IX).

The Maine Wildlands Network design is an extensionof the biosphere reserve concept (UNESCO 1974) asrefined by Noss (1983, 1987b, 1992, 1995), Noss andHarris (1986), Noss and Cooperrider (1994) and Noss et al.(1997). These refined reserve concepts employ three basiccategories of protected area or network components:

• Core wild areas (with a high degree of protection);• Wildlife linkages; and• Compatible-use lands.

Under this scheme, core wild areas remainundeveloped sanctuaries for wildlife and natural processes.Wildlife linkages, necessary to overcome habitatfragmentation, help connect core areas, enabling wildlifeand ecological processes to move across the landscape. Thecompatible-use lands that surround and buffer the coresaccommodate increasing levels of human activity.Compatible-use activities such as ecological forestry andagriculture should complement the functions of core areasand landscape linkages1.

To determine the location and size of the networkcomponents we employ a three-track approach (Noss et al.1997; Noss et al. 1999). This approach incorporates:

1 In this network design for Maine, we do not propose compatible-useareas, though it is hoped that commercial timberlands across much ofthe state can—through gentler, more ecologically informedmanagement—serve to buffer the cores and linkages from the impacts ofindustrial activities and development.


• Protection of “special elements,” such as rare speciesoccurrences, roadless areas, endangered or criticalecosystems, and critical watersheds;

• Representation of key physical and landscape features,such as natural communities, physiography, andvegetation, in protected areas; and

• Conservation of focal species, especially wide-rangingspecies and others of high ecological importance orsensitivity to disturbance by humans.

The most important feature of this three-trackapproach, and one that distinguishes the WildlandsProject’s plans from most of the ecoregional plansdeveloped by other organizations, is the modeling ofhabitat requirements and population viability of wildlands-associated focal species, such as large carnivores (e.g., lynx)and forest mesocarnivores (e.g., American marten (Martesamericana) (Miller et al. 1998; Carroll et al. 2001; Noss etal. 2002). Focal species analysis complements the specialelements and representation tracks by addressing questionsconcerning the size and configuration of reserves and otherhabitats necessary to maintain populations over time.

The emphasis that this plan, and the WildlandsProject in general, places on focal species stems frompersuasive evidence that large predators and other“keystone” species2 play an extraordinarily important rolein ecosystem functioning. As Soulé et al. (submitted)observe, “where the density of a keystone species falls belowsome threshold, the diversity of species in an ecosystemvirtually always decreases, triggering ecological chainreactions ending with degraded or simplified ecosystems(Estes and Palmisano 1974; Terborgh et al. 1999; Crooksand Soulé 1999; Jackson et al. 2001; Terborgh et al.2001).” Among such “keystones” are large predators(Carpenter and Kitchell 1993; Estes and Palmisano 1974;McLaren and Peterson 1994; Terborgh et al. 1999; Rippleand Larsen 2000).

In order to ensure the long-term existence andfunctioning of populations of focal species, especially largepredators, it is necessary to protect and restore big,interconnected wilderness areas. This argument forrestoring and protecting big wilderness based on theregulatory roles of large predators is called “rewilding.”And because large predators require extensive space andconnectivity, the modeling of their habitat requirements onthe landscape is a key tool in conservation area design(Soulé and Noss 1998).

At its root, a wildlands network is an approach toprotecting all native flora and fauna that applies the science

2 Keystone species are defined as relatively rare (low density), and havingparticularly strong, ramifying interactions (Paine, 1969; Mills et al.1992; Power 1992; Power et al. 1996), with impacts that aredisproportionate to their population densities (Mills, et al. 1992, Poweret al. 1996) and not wholly duplicated by other species (Kotliar 2000).

of conservation biology to land management. Thewildlands network design that results from the applicationof conservation biology principles can guideconservationists as they seek to address a seemingly endlesslist of short-term threats. Such a network design provides astrategic framework against which proposed actions can bejudged. Land sales, forestry practices, and other land andwater uses can be weighed against ecosystem needs.

A successful wildlands network in Maine will demandthe full involvement of government agencies at the federal,state, and local levels; and it is these agencies, along withthe donor community, private conservation organizations,and academic institutions, who are the prime audience forthis proposal. Local communities and the general publicmust also be involved in the design and establishment of awildlands network. Outreach to local groups and thegeneral public is an important element of the conservationaction plan presented in Section IX.

The Natural Heritage of MaineBefore discussing the application of conservation

biology principles to Maine, it is essential to take a measureof where Maine stands today regarding its natural resourcesand biodiversity. The discussion below provides anoverview of the biological heritage of Maine and the statusof conservation in the state. Section II then explores thehuman impacts on the natural environment in more detail,while Section III lays out the goals of the Maine WildlandsNetwork.

Maine is renowned for its rich natural diversity. Aslarge as the other five New England states combined,Maine’s more than 20 million acres comprise a broad rangeof physical features, including the dramatic Atlanticcoastline, the legendary Maine Woods, extensive peatlands,an estimated 32,000 miles of flowing waters, and nearly6,000 lakes (Table I.1).


Table I.1 Maine physical statistics.

Attribute ValueLatitudinal range1 42°58’40’’ - 47°27’33’’

~ 450 km (280 miles)Longitudinal range1 66°56’48’’ - 71°06’41’’

~ 320 km (200 miles)Altitudinal range1 0-1,596 meters

(0–5,267 feet)Size2 8,463,000 hectares; 84,630 km2

(20,912,640 acres; 32,676 mi2)Land area2 (Total area of state minusopen water)

7,998,510 hectares; 79,985 km2

(19,764,658 acres; 30,822 mi2) Includes wetlands,which cover ~ 820,000 hectares (2,025,400 acres)

Water area2 (Shallow and openwater)

464,490 hectares; 4,644 km2

(1,147,290 acres; 1,793 mi2)Lakes1 5,785Rivers and streams1 50,700 km. (31,700 mi.)Public conservation lands3 443,977 hectares; 4,440 km2

(1,096,624 acres; 1,714 mi2;) ~ 5.6 % of Maineland area

Forests2 7,067,830 hectares; 70,678 km2

(17,457,540 acres; 27,277 mi2) ~ 88% of land areaCounties1 16Organized towns1 494Unorganized towns1 4502000 population4 1,274,9231 Gawler et al. 19962 Krohn et al. 19983 McGrory Klyza 20014 U.S. Department of Commerce 2001

The forest ecosystems of northern Maine arecharacterized by a cold temperate climate, abundantmoisture, poorly drained soils, and a short growing season.Conifers are well suited to this harsh environment anddominate the northern two-thirds of the state.Predominant species include balsam fir (Abies balsamea) andred (Picea rubens), black (Picea mariana), and white spruce(Picea glauca), which together comprise an estimated 70%of the evergreens. Northern hardwoods, especially sugarmaple (Acer saccharum), American beech (Fagus grandifolia),and yellow birch (Betula alleghaniensis), are scatteredthroughout. In the milder conditions of southern Maine,dominant oaks are accompanied by other hardwoods suchas shagbark hickory (Carya ovata), red maple (Acer rubrum),and gray birch (Betula populifolia)(Bennett 1988).

The Maine Natural Areas Program has cataloged atleast 104 natural community types in the state (includingseven provisional types), which in turn are grouped into 29types of wooded uplands, 14 of wooded wetlands, 20 types

of open uplands, and 34 types of open wetlands3. The 104natural communities are also grouped into 24 ecosystemtypes (Gawler 2001). Two or three regional-scaleecoregions4 depending on the system used—also fallpartially within Maine (Keys et al. 1995; Olson et al.2001).

Maine’s significant biodiversity is linked to the steepenvironmental gradients in the state, which in turn are afunction of its location in the temperate-to-borealtransition zone. As a result, many species reach theirnorthern and southern range limits in the state. For

3 These are vegetated communities only and do not include aquaticcommunities that are not characterized by vascular plants – includingthe non-vegetated Lacustrine, Riverine, Estuarine, and Marine types inthe 1991 (Gawler 1991) natural community classification (Gawler2001).4 Ecoregional classifications are based on broad geographic patternsincorporating major vegetation types, climate, and to some extent,geomorphology (Gawler 2001).


example, of the 198 species of birds that breed in inlandMaine, 59 reach their southern or northern limit in thestate. These peripheral populations may serve as importantreservoirs of genetic diversity (Gawler et al. 1996).

This diverse region provides habitat for an estimated50,000 species of wildlife (Maine Forest Service 1999),including 54 extant land mammals, 218 breeding birds,and 17 each of native amphibians and reptiles (Gawler etal. 1996). Indeed, Maine has a relatively large proportion ofNorth America’s birds, lichens, and mosses—30% of thebryophytes (mosses and liverworts) north of Mexico can befound in Maine (Gawler et al. 1996; see Table I.2).

The diversity of Maine’s natural communities andlandscapes is also a function of the fact that Maine is theleast densely populated state east of the Mississippi River,with a population of less than 1.3 million people (U.S.Department of Commerce 2001). Roughly half of thepopulation lives in eight coastal counties, mostly in thesouthwest portion of the state (Maine EnvironmentalPriorities Council 1999; see Figure I.1). In these coastalcounties, population density ranges from about 39 to over116 persons per square kilometer (100 to over 300 personsper square mile). In contrast, the northern part of the staterepresents the largest blocks of undeveloped, largelyunpopulated land in the eastern United States. This areahosts fewer than 12,000 year-round residents—less thanone-tenth the population of New York’s 6 million-acreAdirondack Park. Many townships in the heart of theMaine Woods have no permanent residents (Land UseRegulation Commission 1997). Population densitiesthroughout northern Maine are generally lower than 13persons per square kilometer (33 persons per square mile).

Finally, much of Maine, especially the north and west,has not been well suited to agriculture:

Maine farms were often remote from goodmarkets, and, aside from the sea lanes,transportation was abysmal. In addition, soils werenot as productive as they were in the states furthersouth. Wheat, a primary export crop in colonialMaine, grew thick and yielded heavily on newlycleared lands, but after a few years yields declined.By 1800, troubles with a blight almost drovewheat from cultivation. Wheat growingrebounded briefly in the Penobscot Valley between1807 and 1814, when Maine farmerscircumvented federal trade embargoes againstFrance and Great Britain by smuggling farm cropsto the Canadian provinces. But these advantagesdisappeared after the War of 1812. Wheat fromNew York and Pennsylvania, grown under morepropitious conditions, replaced native flour even inMaine’s own farmhouses.

In addition to poor markets and sparse soils,Maine’s climate was fickle. In 1816, the yearknown as “Eighteen-hundred-and-froze-to-death,”there was a frost each of the twelve months. Cropswere not harvested until October, and settlers wereforced to rely on fish and game for survival. Thesehardships discouraged thousands and prompted amass migration to the Ohio Valley, the beginningsof a long New England exodus that reachedepidemic proportions after mid-century (Vickeryet al. 1995).

As a result of its low population density, cold climate,and generally poor agricultural soils, nearly 90% of Maine’slandscape remains covered by forest, making it the mostheavily forested state in the nation. Further, many ofMaine’s diverse ecosystems, though affected to varyingdegrees by human activities, remain at least partially intact(McMahon 1998). Maine thus presents an extraordinaryopportunity for ecosystem restoration on a large scale.

Figure I.1 Population density in Maine in 2000.

Source: U.S. Department of Commerce 2001


Forest Management HistoryNative peoples lived throughout much of the land we

know as Maine prior to European settlement, thoughhuman-caused disturbance was relatively localized andinfrequent (Krohn et al. 1998). European explorers such asBritish physician and botanist John Josselyn encounteredwhat they perceived as endless wilderness when they firstentered the North Woods: “[The land beyond the WhiteMountains is] full of rocky Hills, as thick as Mole-hills in aMeadow, and cloathed with infinite thick Woods,”(Josselyn 1672, cited in Bennett 1996). The forests of theseearly days were largely mature; in north central Maine, anestimated 59% of the forest featured trees at least 150 yearsold, and 27% consisted of an all-aged old-growth mosaic,with some trees older than 300 years (Lorimer 1977). Butby the late 1700s, extensive harvesting of Maine’s forestshad begun in earnest (Bennett 1996). Upon reaching theMattaseunk stream and mill in 1846, during his excursionto Mount Katahdin, Thoreau (1865) was moved to write:“here were thousands of cords...which only cumbered the

ground and were in the settler’s way. And the whole of thatsolid and interminable forest is doomed to be graduallydevoured thus by fire, like shavings, and no man bewarmed by it.”

Maine produced an enormous quantity of lumber inthe last decades of the 19th and the first decades of the 20th

century. The private timber barons who accumulatedextensive holdings during the 1800s were largelysucceeded by corporate industrial landowners during the1900s. Many forest products firms achieved verticalintegration, owning the forests and mills, as well asemploying the workers who cut the wood, moved it to themills, and turned the trees into pulp and paper or lumber.This scenario provided (and continues to provide) industriallandowners great leverage over costs and prices (Falk 1973;Osborn 1974; St. Pierre 1976; Lansky 1992), andenormous political and economic influence in fending offrecurring calls for more land to be protected in publicreserves (Hakola 1981; St. Pierre in press).

From 1952 until 1992, industrial pulp, paper, andlumber companies added 1.4 million acres to theirholdings in Maine (Irland 1999). Today, industrial and

Table I.2 Comparison of known numbers of species in Maine with numbers in North America north of Mexico.

Taxonomic Group # in Maine # in N. AmericaNorth of Mexico

% inMaine

Mammals 60 380 16%

Birds 226 650 35%

Reptiles 17 280 6%

Amphibians 18 228 8%

Fish 69 ~800N. America including Mexico

9%

Insects 15,000+ Described: 90,968 est.undescribed: 72,500

~16%

Earthworms ~15 156N. America including Mexico

5%

Crayfish 7 253 3%

Mollusks 110+ ~765N. America including Mexico

14%

Vascular Plants 2107 21,757 10%

Bryophytes 550+ 1850 30%

Lichens 550-700 3499 16-20%

Fungi 3500+ 29,000 microfungi5,000-10,000 macrofungi in U.S.

~10%

Adapted from Gawler et al. (1996)


other large private landowners with timber interests holdroughly 60%, or nearly 10 million acres, of Maine’scommercial forest (Maine Forest Service 2001). This is thegreatest such concentration of land ownership of any statein the U.S. With the sale of approximately 5 million acresof Maine’s forestland in the last two years, the ownershipmix is shifting, but the total large timberland holdingshave not declined dramatically (St. Pierre 1999a, 1999b).The historical pattern of large land ownership in Maine hasbeen a double-edged sword for conservation: it hasminimized development in vast areas of the landscape, butintensive forest management practices, including clearcutting and road building, have resulted in extensivehabitat fragmentation (Maine Forest Service 1999).

The “Mega-easement” PhenomenonThe latest chapter in the environmental and land use

history of Maine involves the acquisition by conservationgroups of the development rights on enormous tracts offorestlands. Between October 1998 and early 2001, 22percent of Maine’s lands changed hands. Many Maine-basedpaper and lumber companies have left, replaced byinternational corporations, pension funds, family trusts,and wealthy individuals. In the course of these transactions,millions of public and private conservation dollars havebeen spent to acquire the development rights on hundredsof thousands of acres of commercial timberland, primarilyin northern Maine. Indeed, the largest conservationeasement deal in the United States to date involved thepurchase of development rights on nearly 760,000 acres offorestland in Maine. Other major deals are in the works,which could easily bring the total amount of land subjectto conservation easements to well over one million acres inthe near term (Curtis 2001).

Although these deals curtail or prevent residentialdevelopment on large tracts of land, there is concern thatthey may be an inefficient use of conservation funds.Central to this concern is the threat being addressed by theeasements. Lewis (2001) notes that “a conservationeasement is a partial interest in land as opposed to full feeownership, and can be an effective tool for regionsattempting to preserve small parcels of undeveloped land inthe face of high growth.” Yet many of the existing or

proposed “mega-easement” areas, with the exception ofsome lake and river frontage, face little developmentpressure. Moreover, most of the largest easements appear todo little to promote best management practices on theselands. Timber can still be harvested unsustainably,herbicides can continue to be used, and land can even besubdivided as long as it stays in forestry (See Lewis 2001,New England Forestry Foundation 2000, Curtis 2001,Pidot 2001, and Walsh 2001 for greater discussion.).

Despite these concerns, conservation easements are avery attractive option for many government officials. “Forme these easements are a terrific value for the publicbecause we only buy what we need” said Governor AngusKing. “The right to use the land and keep it free fromdevelopment can be purchased for one-quarter, one-third toone-half the price of buying the same land outright (Walsh2001).” U.S. Representative Tom Allen, in discussing $4million in federal money for the West Branch of thePenobscot project, noted that the “money included in theInterior appropriations bill will go a long way towardpreserving and protecting Maine’s natural resources foryears to come (Allen 2001).”

Given the tremendous amount of land involved incompleted and planned transactions, the diminutiveportions of these lands that have been set aside as coreareas, the large amounts of private and public conservationfunds devoted to these deals, and the likelihood that theeasements will not significantly improve forestry practices,it is worth considering how much conservation benefit isreally being generated by these mega-easements. Althoughbeyond the scope of this document, such an analysis wouldbe useful for regional conservationists and policy-makers.

The Status of Conservation in MaineAlthough Maine has outstanding conservation

potential, it has the smallest proportion of public land ofany forested state (Irland 1996), with roughly 94% inprivate and corporate ownership (Krohn et al. 1998,McGrory Klyza 2001). Taken together, public andprivate conservation lands about 11.2% of Maine(Table I.3, Figure I.2).


Table I.3 The Conservation Status of Maine by GAP Status Levels.

GAP Status Level DescriptionAmount in

Maine(Hectares)

Percent ofTotal

Land Area

Status 1(“Wild Lands”)

A parcel totally protected from conversion ofnatural land cover and with a conservation planin operation to maintain land in a natural state.Natural processes are allowed to proceedwithout interference or are mimicked throughmanagement practices. Examples include partsof Baxter State Park, Big Reed Preserve, etc.

Status 2(“Low Use”)

A parcel totally protected from conversion ofnatural land cover and with a managementplan in operation to maintain a primarilynatural state, but where uses or suppression ofnatural processes may degrade the quality ofexisting natural communities. Examples includethe newly established ecological reserves onPublic Reserve Lands.

188,025 2.2%

Status 3(“Medium Use”)

A parcel protected from conversion of naturalcover for more than 50% of area, but subject toextractive uses such as timber harvest ormining. Examples include Pingree easementlands.

768,180 9.0%

Status 4(“High Use”)

A parcel with more than 50% of area plannedor in use for agriculture or as "open space" foractive recreation purposes (e.g., ballfields, golfcourses). Natural processes are altered orreplaced by human use and management ofland.

--- ---

1These categories are similar to those used in the GAP Mapping and Categorizing Land StewardshipHandbook, by Crist et al., available on the World Wide Web athttp://gap.uidaho.edu/gap/AboutGAP/Handbook/SMC.htm .

A recent study by the Maine Natural Areas Program(Gawler et al. 2001) reports that 34 of the 97 naturalcommunity types in Maine are rare (S1 or S2; G1, G2 orG35). A related earlier study found that of those that arenot rare, “high quality natural examples are rare (Gawler etal. 1996).” Although a number of the rare types arethought to have been rare prior to European settlement, allhave since been further depleted by agricultural conversionand development in southern Maine and by timberharvesting in northern Maine. Gawler et al. (1996) statefurther that “land use trends point to increasingfragmentation from development in the southern part of

5 NatureServe, The Nature Conservancy and the Natural HeritageNetwork have developed a conservation status ranking system todescribe the relative imperilment, or conservation status, of plants,animals, and ecological communities (elements) on a global (G), national(N), and subnational (state/provincial - S) level. Rankings are assigned,reviewed, and revised according to standard criteria, on a scale from 1 to5, with 1 being “critically imperiled” and 5 being “demonstrablywidespread, abundant, and secure” (NatureServe Explorer 2002).

the state and increasing fragmentation and forestsimplification from harvest activities…in the northern partof the state.” At least 20% of Maine’s native plants areRare, Threatened, or Endangered (RTE). Maine’s nativeplant diversity has declined over the past century, with atleast 29 species having been extirpated. Exotic speciescomprise 25% and 31% of fish and plant speciesrespectively (Gawler et al. 1996).

Mostly as a result of large-scale commercial logging, asopposed to clearing for agriculture and livestockgrazing—as occurred in Vermont, New Hampshire andother parts of the northeast with better agricultural soilsand access to markets—less than 0.5% of the state’s oldgrowth forests remain (Gawler et al. 1996; Davis 1996). Indiscussing the pre-European landscape of New England,Foster (2001) concludes that:

Overall, we had a largely forested landscape thatcontained a relative abundance of big old trees, and


forests that were driven by relatively infrequentdisturbances interjected into a landscape controlledby broad physical and biological processes. Thatview doesn’t argue for extensive grasslands,shrublands, heathlands, and other kinds of openlandscapes other than those that are generated on avery local scale by small numbers of mobile peopleand by animals.

Lorimer (1977) reaches very similar conclusionsregarding the composition and natural disturbance cycles ofpresettlement forests in northeastern Maine. Through ananalysis of early survey records, he found that in an area ofabout 1.6 million hectares, the forests were largely in a“climax state as indicated by the dominance of shade-tolerant species and the small percentage (8%) of intolerantor early successional species.” Lorimer found further thatthe average recurrence interval of major fire was 800 years,and of catastrophic windstorm 1,150 years. Irland (1999)concurs, suggesting that “uneven-aged stand structureswould have been common in the presettlement forest. Thisresult is supported by analyses of the few remaining[presettlement/old growth] stands that have beenmeasured, and by the modest presence of short-lived orearly successional species in such stands.”

The forests described by Foster, Lorimer, and Irlandhave been replaced by a much younger forest in variousstages of regeneration (Maine Forest Service 2001), theresult of 150 years of systematic resource extraction.Because of the nearly complete conversion of the old-growth forests that dominated Maine, we have also lost thefeatures (dominance of shade-tolerant species, large tracts ofunfragmented interior forest, large standing snags andcoarse woody debris) and natural processes (insectoutbreaks, fires, windthrow) that characterized them.

Maine’s forests reflect the realities in other parts of theNorthern Forest of New York, and New England. TheNorthern Forest Alliance (2002) observes that:

Today the Northern Forest is biologically young,with a median tree age of 55 years among nativetree species that can live 250-400 years on average.More than half the growing stock is in the smallestof four size classes, under 13 inches [33 cm] indiameter [USDA 1997], and less than 10% of thestanding timber in the region would be consideredof good commercial quality [National WildlifeFederation 1995].

In 1998, the Maine Forest Biodiversity Project (MFBP)published a report identifying potential benchmark reservesites on the state’s public land and on private conservationareas, encompassing 498,700 acres (2,019 km2) or about2% of the state’s total land area (McMahon 1998). TheMFBP report cautions that even if all of their qualifyingecoreserves were protected, they alone would notmaintain—much less restore—biodiversity. In January2001, Maine acted on the recommendations of the MFBPand mandated the establishment of a series of ecologicalreserves on existing Public Reserved Lands. Despite thecautionary words of the final MFBP report, only 70,000acres (roughly 0.3%) of the state’s land was protected.Further, the State Legislature stipulated that no more than15% of Maine’s Public Reserved Lands, and no more than6% of such land that is capable of sustaining forestryoperations, be designated as ecological reserves.

Even with the establishment of these reserves, onlyslightly more than two percent of the state is strictlyprotected (the equivalent of GAP Status 1 and 2 in Krohnet al. 1998; McGrory Klyza 2001; see Table I.3). Most ofthese conservation lands are in parcels that are less than200 hectares in size and are not interconnected (Krohn etal. 1998). Further, most remaining state lands aredeveloped for intensive recreation, used for logging, oropen to other consumptive uses (Maine Forest Service2001).


Figure I.2 Maine’s major cities, roads, rivers, and lakes, GAP class 1 and 2 protected lands, large (>5000 ha) class 3 landscontaining easements, and other managed conservation lands.


Most of the major conservation initiatives carried outin the late 1990s are merely maintaining managedtimberlands, or “working forests.” Although thesetransactions aim to meet the demands for public tourism,maintain the appearance of natural beauty, and manage forsustained production, most have not sufficientlyincorporated preservation of biodiversity and ecologicalintegrity into the equation, as Gawler et al. (1996) observe:

Evidence indicates that the working forest can bemanaged to maintain many aspects of biologicaldiversity in an economically viable framework, butthat certain aspects—in particular, representative,unmanaged ecosystems and some habitatspecialists—will be incompatible with forestry asit is generally practiced.In sum, as of mid 2002, over 97% of Maine’s landscape

continues to be managed for forestry or agriculture, or isused for residential, commercial, or industrialdevelopment. By nearly any measure, existing efforts toprotect and restore Maine’s biodiversity and ecologicalintegrity are inadequate. The problems, and opportunities,confronting conservation in Maine are summarized here:

• Little of Maine remains in its presettlement state. As inmost of the rest of the Northeast, Maine’s presentforests have been heavily influenced by human action,intentional or accidental. This has led to forestfragmentation, changes in forest stand dynamics (e.g.,tree species composition, decreasing forest age-classdistribution), and disruption of natural ecologicalprocesses.

• Pockets of truly wild habitat are too small, tooisolated, and represent too few types of ecosystems tomaintain native biodiversity in all its forms.

• A number of native species, notably large carnivores,have been nearly eliminated from the landscape by theindirect and direct effects of human actions and habitatalteration. The eastern timber wolf (Canis lycaon),mountain lion (Felis concolor), and wolverine (Gulo gulo)have been extirpated from Maine, while others, such asthe Canada lynx (Lynx canadensis), bobcat (Felis rufus),and American marten (Martes americana) occupy only afraction of their historical range. Some of these wide-ranging predators have shown signs of recolonization,but much work needs to be done to restore viable andecologically effective populations of carnivores (Soulé etal. submitted).

• Extensive residential and commercial development insouthern Maine has fragmented and destroyed wildlife.

• The construction and operation of dams, waterwithdrawals, certain agricultural and forestry practices,and chemical pollution have diminished freshwaterecosystem integrity and connectivity.

• The introduction and spread of exotic species hasdisrupted ecological processes and displaced nativespecies.

• Few state, federal, or private organizations are willingto propose large-scale conservation measures thatinclude high-level wildlands protection, despite therarity of many of Maine’s natural communities andspecies, and the projected increase in forestfragmentation and destruction.

• Large tracts of industrial forestland are being boughtand sold at an astonishing pace in northern, westernand eastern Maine. The instability of this situationmakes it difficult to establish long-term managementpractices oriented toward restoration and biodiversity.But the availability of such parcels also offersopportunities to acquire, protect, and restore largeamounts of habitat.

• The science of conservation biology has matured to thepoint where it can accurately guide the design ofprotected areas networks. These principles advocate forthe protection of large blocks of habitat. Further,functional connectivity between habitat patches appearsto be essential to preventing or mitigating thedeleterious effects on animal populations associatedwith fragmentation. Connectivity should also serve torestore viable and ecologically effective populations ofcarnivores (Soulé et al. submitted).

As this list makes clear, there is a need for systematic,landscape-scale planning that will yield a conservation areadesign for Maine that can satisfy the four key goalsarticulated by Noss (1992) and Noss et al. (1997), asadapted to Maine. This document proposes such adesign—the Maine Wildlands Network. To providefurther background on the bases for the design, thefollowing section (Section II) explores the human impactson the natural environment in more detail. Section III laysout the objectives of the Maine Wildlands Network.Section IV describes the rewilding concept; the core,linkage, compatible use model; and the three-tracknetwork design methodology in detail. Section V andAppendix 3 provide more information on the focal speciesused in the analysis. Sections VI, VII, and VIII describe theproposed Maine Wildlands Network. Section IX thenoffers recommendations for the next steps needed toestablish the network.

This network design was developed with input frommore than 30 regional scientists, naturalists, andconservationists, and balances scientific rigor with thetimeliness necessary to address current trends inbiodiversity loss. Practical while visionary, this design isintended to stimulate discussion and provide an ecologicalframework for regional and local conservationists andplanners. This said, the network design is the result of astep-wise, empirical, and iterative process undertaken withthe understanding that it will require revision andadaptation as new information becomes available, and asconservation measures are carried out on the ground.


II. ECOLOGICAL WOUNDSThe introduction and background provided an

overview of the key issues related to landscape-scaleconservation in Maine. We now turn to a closerexamination of several of these issues, namely the majorhuman impacts that have altered or degraded Maine’secosystems.

Aldo Leopold recognized that we have done greatdamage to the natural world. He wrote:

One of the penalties of an ecological education isthat one lives alone in a world of wounds…Anecologist must either harden his shell and makebelieve that the consequences of science are none ofhis business, or he must be the doctor who sees themarks of death in a community that believes itselfwell and does not want to be told otherwise(Leopold 1972).

In the process of developing the Maine WildlandsNetwork design, we identified five primary human impactsor “ecological wounds” to the natural ecosystems of Maine.We focus on these impacts, below, because the reversal oftheir effects on the ecosystems of Maine will be essential toachieving the ultimate goals of the Maine WildlandsNetwork. We discuss the subject of “healing thewounds”—and wildlands restoration more generally—inSections III and IV.

1. Elimination of old growth, fragmentation offorest habitat, changes in forest stand dynamics(e.g., tree species composition, truncated forestage-class distribution), and disruption of naturalecological processes as a result of forestmanagement practices.

When Thoreau made his epic journey to the MaineWoods over a century and a half ago, he encountered adiverse forest where the average age of trees was 125 yearsand the average diameter was close to two feet. Althoughmany of the largest trees, especially pines and othersoftwoods close to rivers, had already been cut, the forest asa whole was intact, and a closed canopy spread acrossthousands of square miles of unfragmented forest(Thoreau 1865).

The forest that Thoreau saw is gone, with theexception of a few small patches, replaced by forests invarious stages of regeneration that have been subjected tointensive forest management for decades. Although woodhas been harvested commercially since at least the 1850s,the introduction in the 1960s of skidders and other

mechanized means of harvesting wood, combined withchanges in mill technologies, brought about significantchanges and increases in timber harvesting in the Mainewoods (Irland 1999). Between 1982 and 1994, more than2.2 million acres of forest were “clearcut” (more than 80%of the volume of trees was removed) (Griffith and Alerich1996). The rights-of-way from the tens of thousands ofmiles of new logging roads (ignoring yards and trails) inthe unorganized territories converted more than nine timesas much forest land as did all the houselots in that regionbetween 1972 and 1993, and created far morefragmentation (Land and Water Associates et al. 1994).Even though many logging roads are temporary and aregenerally smaller and less heavily traveled than permanentpaved roads, their effects can still be considerable inproviding access and otherwise compromising forestinterior habitat (Noss 1996). Direct effects of roads onwildlife include roadkill, road aversion and other behaviormodifications, fragmentation and isolation of populationsof various species, pollution (e.g., noise, heavy metals, salt),direct habitat loss, facilitated invasion of exotic species, andalterations to the hydrology of watersheds. The mostprofound effect, however, is indirect, that of facilitatedaccess by humans (Noss 1996).

Lansky (1999) observes that generally on industrialforestlands in Maine, “large absentee landowners…duringthe period 1982-1995, cut twice as much as net growth,decreased the acreage of spruce-fir, vastly increased theacreage of seedlings and saplings, increased the ratio oflow-value species, and lowered hardwood sawlog quality.”Changes in technology made these unsustainable practicespossible and, as Irland (1999) notes, “these technologieshave transformed the opportunities for forestry.” But Irlandalso acknowledges that “it is a great tragedy…that theseopportunities to improve forest management have so rarelybeen taken.” Indeed, while development is definitely theprime cause of habitat fragmentation in the southern partof the state, timber management is the prime cause in thenorthern part (Gawler et al. 1996). Although clearcuttinghas dropped dramatically as a harvesting method since1989—from 44% of harvest in 1989 to 3.5% in1999—the total number of acres harvested annually hasincreased by 64%, from 325,000 in 1989 to 532,000 in1999 (Maine Forest Service 2001).

High rates of cutting have also reduced standingtimber volumes significantly. Industrial lands in Maineaverage about 16.3 cords/acre (Maine Forest Service 2001).Recent harvesting has also produced compositional changesin species type and age-class, skewing forest structuretoward younger age classes. In the last two decades, for


example, seedling and sapling acreage has increased to 4.2million acres—an area equal to nearly 25% of Maine'sforests (Griffith and Alerich 1995

The conversion of older, late-seral stage forests tosignificantly younger stands has likely had many ecologicaleffects yet to be documented. A naturally functioning forestecosystem is the interaction of periodic large-scaledisturbances like fires, ice storms, and insect epidemics,with effects played out over longer periods of time due toclimatic and other environmental changes. Naturaldisturbances typically leave coarse woody debris behind inthe form of down logs, fallen leaves, branches, and snags.These woody materials provide a long-term source ofnutrients, as trees rot and are returned to the soil. Thedebris also provides an important structural componentthat is essential to a number of species, including focalspecies such as lynx and marten (Ray 2000; Payer andHarrison 1999, 2000).

While nutrients removed by timber harvest canpotentially be replaced over time, such nutrient recharginggenerally requires hundreds of years—far longer than thestandard rotation period between cuttings on industrialtimberlands. As eloquently stated by Leopold (1949) half acentury ago:

Many forest plantations are producing one-log ortwo-log trees on soil which originally grew three-log and four-log trees. Why? Thinking forestersknow that the cause probably lies not in the tree,but in the micro-flora of the soil, and that it maytake more years to restore the soil flora than it tookto destroy it.

A number of foresters in Maine acknowledge that thereare problems with current management practices (Irland1999). In response, alternative strategies are emerging thatwould dramatically improve forest management practiceswhile also improving forest health and providing asustainable economic base for the rural economy. TheMaine Low-Impact Fores t ry Pro ject ( seehttp://hcpcme.org/home.html ), for example, advocates foran ecologically viable forestry system that would:

Ensure that, over time, all the habitats capable ofsupporting viable populations of all the nativespecies are present, despite natural and human-induced changes. It is easy enough to create earlysuccessional habitats – that can be done with atimber harvester in a day. The real challenge is toensure the presence of late-successional interiorforest, which takes more than a century to grow.There needs to be adequate time for recovery fromdisturbance and adequate sources nearby forrecolonization of disturbed areas. Managed forests

should, in this regard, complement the processestaking place in reserves or wilderness, rather thanisolate or fragment them.

There also has to be a way to make sure thatthe land is going to stay permanently forested.One cut, however sensitive, does not mean that theforestry is sustainable, because the landownercould, in a few years, sell to a liquidator ordeveloper. There are ways to make such long-termcommitments now with easements, and landtrusts, or changes in deeds (Lansky 1999).

The Maine Forest Biodiversity Project (MFBP) has alsodeveloped a set of comprehensive “voluntary forestrypractices that can help maintain forest biodiversity inMaine…the suggested practices are intended to maintaincurrent biodiversity, but they can also be used to enhancecomponents of biodiversity that have become locally orregionally uncommon (Flatebo et al. 1999).” Theguidelines are quite specific, focusing on five standcharacteristics, including vertical structure and crownclosure, native species composition, and down woodmaterial, snags and cavity trees, as well as 10 specialhabitats and ecosystems, such as riparian and streamecosystems, old growth and primary forests, and rarenatural communities.

Given that the vast majority of the Maine woods is stillunder private commercial management, the guidelinesoffered by the Maine Low-Impact Forestry Project and theMaine Forest Biodiversity Project are welcome andproductive. The new recommendations are emerging at at i m e whe n the fo r e s t pr o duc t s ind us t r y in Ma i ne is und e r go i n g ra pi d ec o n o m i c cha nge . Po we r (2001 ) fo un d tha t :

In 1998 only about 4 percent of all Maine personalincome and jobs were directly derived from thelumber and wood products and pulp and paperindustries combined. Employment in forestproducts across Maine has fallen significantly andis projected to continue to fall. Adjusted forinflation, wages in wood products, includinglogging, have also been in decline. Thecombination of dwindling jobs and decliningwages has hit timber-dependent communitiesdoubly hard.

These declines during past decades have takenplace during a period of rising timber harvest. Butharvest of many commercial tree species in recentyears has exceeded net forest growth by aconsiderable margin. To stay within sustainedyield, Maine forest harvests will have to bereduced. This could further accelerate the declinein forest products employment in Maine.


The changing nature of the forest products industry inMaine actually provides an opportunity to diversify therural economy, increase employment, raise wages, and conserve vast tracts of northern Maine. Indeed,“communities concerned about economic stability havelittle choice but to play to their strengths and uniqueness.Around the country, including in other parts of Maine, asnatural resource industries have declined as sources ofemployment and income, high quality natural landscapeshave successfully supported the development of alternativeeconomic bases (Power 2001).”

2. Loss of native species as a result of habitatalteration, hunting, and introduced species.

Maine’s native biological diversity has sufferedextensive losses since European settlement. In the past 100years, for example, human-induced changes have resultedin the regional extirpations of at least four mammals(eastern cougar, eastern timber wolf, sea mink [Mustelamacrodon], and woodland caribou [Rangifer taranduscaribou]), three birds (Labrador duck [Camptorhynchuslabradorius], passenger pigeon [Ectopistes migratorius],loggerhead shrike [Lanius ludovicianus]), one reptile (timberrattlesnake [Crotalus horridus]), and 29 plants (see TablesII.1 and II.2). At least 100 additional plant species havenot been documented in over 20 years; roughly half of thesemay be extirpated as well. Little is known about Maine’sinvertebrates, but it has been speculated that major lossesor reductions in abundance have probably occurred in boththe terrestrial and aquatic environments (Gawler et al.1996).

Native fish have also been severely affected by humanactivities (e.g., dam construction, pollution, and overfishing), with exotic species—comprising 25% of Maine’scurrent fish species—likely to be displacing some nativepopulations. Of the 52 native freshwater fish speciesrecognized in Maine, 17% are considered rare at some level(Gawler et al. 1996). For example, the shortnose sturgeon(Acipenser oxyrhynchus), historically found in tidal riverbasins throughout the year (Bennett 1988), is currentlylisted as Endangered under the Endangered Species Act(ESA). Likewise, Atlantic salmon (Salmo salar) werehistorically numerous in Maine, yet nowadays fewer than300 wild salmon return statewide each year (Stevens 1999).In recognition of their imperiled status, wild Atlanticsalmon in Maine were listed under the Endangered SpeciesAct (ESA) in November 2000 (USFWS 2000). In early2002, the National Academy of Sciences bolstered thedecision to list the species as Endangered, confirming thatMaine’s wild Atlantic salmon form a distinct populationsegment, one genetically distinct from Canadian salmon(Young 2002).

Numerous anthropogenic factors have been identifiedas primary causes of species loss in Maine. Some of theseinclude: hunting and trapping, habitat fragmentation,intensive timber management, permanent landconversion/development, introduced species, collection (ofplants and invertebrates), recreational land use, damconstruction, pollution, nutrification and pesticide use.Long-term climate change may also have significantimplications for Maine’s biological diversity in the future.


Table II.1 Estimated or known numbers of species in various taxonomic groups and their status in Maine.

Taxonomic Level& Group

Native&

extant

Introduced Extirpated FederallyListed1

StateListed2

StateRare3

Endemic State ofKnowledge

KINGDOM ANIMALIA

Mammals 54 2 4 2 1(3) 7 0Thorough for

speciesdistribution

Birds(breeding only)

198 5 3 6 9(22) ? 0Thorough for

speciesdistribution

Reptiles 17 0 1 0 4(0) 7 0 Fairly wellknown

Amphibians 17 1 0 0 0(3) ? 0 Fairly wellknown

Freshwater Fish 52 17 ? 1(1) 0(2) 10 0 Fairly wellknown

Insects 15,000+ 10+ 7+ 1 0(15) 38+ ?

Few groupsfairly wellknown,majorityunknown

Crustacea(crayfish only)

4 3 0 0 0 1 0Certain groups

fairly wellknown

Arachnids 500+ 6+ ? 0 0 ? 1 Not wellknown

Molluscs110 ? ? 0 0(4) 3 ?

Most speciesnot wellknown

OtherInvertebrates ?? ? ? 0 0 ? ?

Very poorlyknown

KINGDOM PLANTAE

Vascular Plants 1432 643 32+ 3 175 254 2Well known,esp. for rare

species

Bryophytes 550+ 0? ? 0 0 1+ ? Not wellknown

KINGDOMFUNGI 3500+ ? ? 0 0 ? ? Very poorly

knownLecanorales

(lichens)550-770 0? ? 0 0 ? ? Not well

known4

KINGDOM PROTISTA

Algae, protozoa ? ? ? 0 0 ? ? Very poorlyknown

1: Federally listed Endangered or Threatened species only.2 Animals state Endangered or Threatened species, 1986 list; 1995 proposed additions in (). Plants: state Endangered or Threatened species: 1989 list plus S1 plants not yet listed.3 Tracked by the Natural Areas Program as rare in the state: rarity ranks of S1-S3.4 Studies by Steve Selva and University of Maine colleagues suggest, however, that many native lichen species need old trees and hence may be diminishing in Maine’s unnaturally young forest.Source: Adapted from Gawler et al. 1996. Table excludes marine and estuarine species.


Table II.2 Plants believed to have been extirpated from the state, and those that have not been documented in Maine in20 years.

Extirpated Plants Possibly Extirpated Plants(Not known to currently exist in Maine; not field-verified[or documented] in Maine over the past 20 years.)

Anticosti aster American winter-cress Marsh bulrushBlunt-leaved bedstraw Awned sedge Marsh milkwortBush’s sedge Beaked spikerush Mountain hairgrassButterfly weed Bicknell’s sedge Netted chain-fernCarpenter’s square Bitter fleabane Nuttall’s bush-cloverDisjunct eyebright Blunt mountain-mint Palmate-leaved violetEastern Sycamore Boreal bedstraw Prairie wedge-grassElegant milk-vetch Boundary meadow-rue Red-stemmed gentianElm-leaved goldenrod Canadian wormwood Richardson’s tansy-mustardGreat blue lobelia Cat-tail sedge Robinson’s hawkweedHairy hawkweed Cliff muhly SamphireScarlet painted-cup Climbing hempweed Sand-dwelling dewberryLarge whorled pogonia Coast violet Secund rushLate purple aster Comb-leaved mermaid-weed Shore quillwortMoss campion Common mare’s-tail Slender false pimpernelNorthern meadow sedge Dwarf dandelion Slender knotweedPennsylvania pellitory Engelmann’s spikerush Slippery elmPurple cudweed Estuary boneset Small dropseedRobbin’s milk-vetch Falkland Island sedge Small rabbit tobaccoRock goldenrod Foxtail sedge Smooth-sheathed sedgeRound-leaf boneset Gaspe arrow-grass Southern bog clubmossRue-anemone Great St. John’s-wort Southern naiadScarlet painted-cup Hair-like sedge Southern slender ladies tressesSchreber’s aster Hairy boneset Stiff gentianSharp-lobed hepatica Hairy bush-clover Straight-leaved pondweedSlender pinweed Hairy ground-cherry Teal love grassUnicorn root Hairy wood brome-grass Thread-like naiadWalking fern Houghton’s umbrella sedge Three-seeded mercuryWild lupine Lake-cress Torrey’s rushYellow star grass Large toothwort Water-plantain spearwort

Leonard’s skullcap White bluegrassLesser wintergreen White vervainLimestone swamp bedstraw Wild chervilLoose-flowered sedge Wild chessLopseed

Source: Adapted from Gawler et al. 1996 and Maine Department of Conservation 2001


Approximately 41% of Maine’s mammals are limitedin distribution because they occupy geographically limitedhabitat types. These species and other habitat specialistsmay be especially vulnerable to extirpation if associatednatural habitat types are eliminated. Indeed, most ofMaine’s known extirpated species were habitat specialists(Gawler et al. 1996).

Two of Maine’s top predators, the eastern timber wolfand the eastern cougar, were heavily hunted and trapped byearly settlers and eliminated from the landscape by the late1800s. The Canada lynx, whose state bounty was removedin 1967 (Bennett 1988), is currently a species of specialconcern in Maine due to its low numbers and a lack ofreliable information on its population size and trend (seeSection V and Appendix 3 for more background on wolf,cougar, and lynx). Some wildlife species that were huntedinto extirpation by Maine’s early settlers later went extinctglobally. For example, the last passenger pigeon in Mainewas reportedly shot in 1896 and the species went extinct in1914; and the sea mink, whose fur was sought by pelthunters, was exterminated from Maine by about 1860, andwent extinct soon after. The Labrador duck, whichprobably wintered off the coast of Maine, was exploited toextinction for its plumage. Other species, like the beaver(Castor canadensis) and the wild turkey (Meleagris gallopavo),have made strong comebacks in Maine after having beenvirtually eliminated in the 1800s (Bennett 1988).

The woodland caribou, today extirpated from northernNew England, was once a permanent resident of northernMaine. Until the late 1800s, there was little regulation ofcaribou hunting. At this time, Maine officials noted therapid disappearance of the species, and banned hunting ofcaribou in 1899. The caribou population was unable torecover, however, and the last sightings were made in theMt. Katahdin area in the early 1900s. Overhunting andpossibly infestation by brain worm carried by white-taileddeer (Odocoileous virginianus), which extended its range intonorthern Maine in the mid-1800s, are cited as causes of theloss of caribou in Maine. Two reintroduction efforts, in1963 and 1989, were unsuccessful (Bennett 1996). The1989 reintroduction probably failed because of mortalitycaused by the brain worm parasite and black bearpredation. Although 39% of the mortalities investigatedwere likely attributable to neurological disease, theevidence suggests that infections were contracted incaptivity (McCollough 1991). Reintroductions into areaswith white-tailed deer are quite risky; all seven cariboureintroduction attempts in white-tailed deer range havefailed, but 24 out of 29 reintroductions succeeded innorthern regions with no deer present (McCollough 1987).Despite these successes, McCollough (1991) concludes thatattempting to restore woodland caribou to habitats ineastern North American that contain moderate populationsof white-tailed deer infected with brain worm, and black

bears, is a high-risk and extremely costly venture with alow probability of success. Even with ecological restorationof the forests of Maine, including road removal, restorationof wolves (with subsequent decrease in deer populations)and landscape connectivity, and an increase in the totalamount of secure habitat, successful reintroduction isprobably not possible (J. Ray pers. comm.). The reasons forthis are several and interrelated. Among them are theexpectation that climate change would favor deerpopulations in northern Maine, which in turn wouldincrease the likelihood of brain worm infection (G. Forbespers. comm.). Climate change would also likely reducepreferred caribou habitat in Maine—lichen-rich matureforests, open muskegs and bogs, and alpine or high latitudetundra which the caribou prefer (J. Ray pers. comm.).

Although the historic distribution of the wolverine is atopic of debate in Maine, its presence is well documentedin the historical literature. For example, Seton (1929, ascited in Bennett 1996) suggested that the wolverineinhabited the northern two-thirds of the state. Thewolverine was pushed north by hunting and trapping,however, and by the mid-1850s, was rarely sighted in NewEngland (Bennett 1996).

3. Habitat fragmentation and destructionin southern Maine as a result of extensiveresidential and commercial development.

Approximately 87% of Maine’s human populationresides in southern and coastal Maine, resulting inrelatively high population densities in these regions (mean46 people/km2, range 0-1,107 people/km2; Figure II.1).Throughout southern Maine, road densities are alsorelatively high (mean 1.2 km/km2, range 0.1–6.7 km/km2;Figure II.2). Related development trends have led to foresthabitat loss and significant fragmentation of remainingforested habitat. Notably, southern and coastal Mainesupport the state’s highest richness of terrestrialvertebrates, Threatened and Endangered vertebrates, andwoody plants, despite the dearth of protected areas (Krohnet al. 1998). These areas have relatively productive soils andthus high biological productivity; so it is little surprisethat both humans and wildlife are attracted to them(Publicover 2002).

This situation, already of concern, seems destined tobecome worse. Over the last 30 years, the fastest growingtowns in Maine have been the outer suburbs, locatedwithin 10-25 miles (16-40 km.) of metropolitan areas. Thistrend continues today, with virtually all of the statespopulation growth occurring in these areas (Maine StatePlanning Office 1997). Projections suggest that much ofsouthern Maine could be urban or suburban by 2050(Figure II.3, also refer to Figure II.1).


Figu

re II

.1 C

ente

rs o

f hum

an p

opul

atio

n an

d hu

man

popu

latio

n de

nsity

by

tow

n.Fi

gure

II.2

Roa

ds in

Mai

ne (

by c

lass

). S

ourc

e: M

aine

offi

ce o

f GIS

, 199

9


Figure II.3 Past, current, and projected development in southern Maine.

Source: Maine State Planing Office-Expansion of Development application

Beyond the indirect costs of sprawl, such as the effectof automobile air pollution on natural ecosystems andspecies, the direct costs of sprawl are substantial. Habitatfragmentation and loss have likely contributed topopulation declines of species such as bobcat, owls, hawks,and certain songbirds (Maine State Planning Office 1997).These species require large blocks of undeveloped land tomaintain their populations. In southern Maine, nestingsites for endangered birds such as the piping plover(Charadrius melodus) and least tern (Sterna antillarum) havebeen lost to development (Maine State Planning Office1997). In 1985, a study of 8 towns in southern Mainefound that 76% of all wetlands were visible from a road orwithin 2,000 feet of a road, and thus of limited habitatvalue (Maine State Planning Office 1997).

4. Loss of freshwater ecosystem integrity andconnectivity as a result of the construction andoperation of dams, water withdrawals, certainagricultural and forestry practices,and chemical pollution.

Freshwater ecosystems throughout Maine arethreatened by the continued presence of dams, invasions ofexotic species, escaped fish from the aquaculture industry,the withdrawal of water for ski areas and agriculture, andcertain management practices in forestry and agriculture.

Between 1800 and 1992, nearly 80 dams wereconstructed in Maine. Of these, 82% were created togenerate electric power, 12% for flood control, 3% tocreate lakes for recreational use, and 3% to fulfill watersupply needs (Free World Data Foundation 1999). Damscan have many different effects on the immediatelandscape, but can also affect ecosystems and species farfrom their location.

The following list, compiled by American Rivers(2001), identifies some of the ways that dams can damageriver ecosystems.

1. Dams reduce river levels. By diverting water forpower, dams remove water needed for healthyin-stream ecosystems. Stretches below dams areoften completely de-watered.

2 . Dams block rivers. Dams prevent the flow ofplants and nutrients, impede the migration offish and other wildlife, and block recreationaluse. Fish passage structures can enable apercentage of fish to pass around a dam, butmultiple dams along a river make safe travelunlikely.

3. Dams slow rivers. Some ecologically importantfish species (e.g., salmon) depend on steadyflows to flush them downriver early in their lifeand guide them upstream years later to spawn.Stagnant reservoir pools disorient migratingfish and significantly increase the duration oftheir migration and expose them to predationfor longer intervals.

4. Dams alter water temperatures. By slowing waterflow, most dams increase water temperatures.Alternately, some dams decrease temperaturesby releasing cooled water from the reservoirbottom. Fish and other species are sensitive tosuch temperature irregularities, which oftendestroy native populations and favor exoticspecies.

5. Dams alter timing of flows. By withholding andthen releasing water to generate power for peakdemand periods, dams cause downstreamstretches to alternate between no water andpowerful surges that erode soil and vegetation,thus flooding or stranding wildlife. Theseirregular releases destroy natural seasonal flowvariations that trigger natural growth andreproduction cycles in many species.

6 . Dams—not natural processes—control reservoirlevels. Peaking power operations can causedramatic changes in reservoir water


levels—often up to 40 ft—which degradeshorelines and disturb fisheries, waterfowl, andbottom-dwelling organisms.

7 . Dams decrease oxygen levels in reservoir waters.When oxygen-deprived water is released frombehind the dam, it kills fish downstream.

8 . Dams hold back silt, debris, and nutrients. Byslowing flows, dams allow silt to collect onriver bottoms and bury fish spawning habitat.Silt trapped above dams accumulates heavymetals and other pollutants. Gravel, logs andother debris are also trapped by dams,eliminating their use downstream as food andhabitat.

9 . Dam turbines harm or kill fish. Followingcurrents downstream, fish are drawn into andwounded or killed by power turbines. Whenfish are trucked or barged around the dams,they experience increased stress and disease anddecreased homing ability.

10. Dams increase predator risk. Warm, murkyreservoirs often favor predators of naturallyoccurring species. In addition, passage throughfish ladders or turbines injure or stun fish,making them easy prey for flying predators likegulls and herons.

While not all of these effects occur on all dammedrivers in Maine, as few as two or three acting in concertmay result in conditions that jeopardize native species,natural communities, and natural processes along thelength of the river. Dams are cited as one of the primarycauses of Atlantic salmon decline in Maine, excludingsalmon from about one-half of existing habitat (USFWS1999a). Acidic precipitation and toxic chemicals also pose arisk to freshwater ecosystems. The EPA (2001) notes that:

The most obvious environmental effects of acidrain are the loss of fish in acid sensitive lakes andstreams. Many species of fish are not able tosurvive in acidic water. Acid rain affects lakes andstreams in two ways: chronic and episodic. Chronicacidification results from years of acidic rainfall. Itreduces the alkalinity (buffering capacity) andincreases the acidity of the water. Chronicacidification may reduce the levels of nutrientssuch as calcium, which, over time, may weaken thefish and other plants and animals in an aquaticecosystem. Episodic acidification is a sudden jumpin the acidity of the water. This can result from aheavy rainstorm. It also happens in the spring,because the sulfates and nitrates will concentrate inthe lowest layers of a snowpack. Episodicacidification can cause sudden shifts in water

chemistry. This may lead to high concentrations ofsubstances such as aluminum, which may be toxicto fish.

Acid rain also deposits nitrates that can lead toincreases in nitrogen in forests. Nitrogen is an importantplant nutrient, but some forest systems may not be able touse all they receive, leading to nitrogen saturation. In theeastern United States, there is evidence of nitrogensaturation in some forests. Excess nitrogen may causeeutrophication in areas where rivers enter the ocean. This isa concern in places such as Casco Bay, where as much as40% of the total nitrogen entering the bay may come fromatmospheric deposition (EPA 2001).

Toxic elements are also entering aquatic ecosystems viaprecipitation. Scientists from state and federal agencies andmajor universities have sampled rain (and snow), and havediscovered that New England’s precipitation containsmercury in concentrations that exceed EPA’s safe level forthis element in lakes and streams. Mercury contaminationof rain has been documented in Maine’s parks and ruralcommunities, including Acadia National Park, wheremercury levels in rain are up to four times as high as theEPA’s aquatic life standard for mercury in surface water.That standard is also used to protect wildlife (NationalWildlife Federation 2000).

Aquatic ecosystem integrity and water quality are alsoaffected by forest management practices. Thousands ofmiles of small streams course through the forests of Maine.Indeed, the Boundary Mountains of western Maine maycontain one of the highest densities of headwater streamson earth. Because there are so many more small streamsthan big ones, and because heavy wood harvestingmachinery frequently encounters such streams, it isimportant to determine the effects of harvesting on thosestreams and how those impacts can be mitigated oreliminated (Hagan 2001). Kahl (1996) summarizes some ofthe effects of forest management on freshwater systems:

Water quality integrates upstream watershedprocesses and disturbances....The most importantfactor for protection of water quality ismaintaining a forested landscape. Water quality isdegraded when eroded soils cause sedimentation,excessive amounts of nutrients are leached from asite, or temperature of water is raised above levelssuitable for existing aquatic biota. Chief causes ofwater quality degradation from forested watershedsare excessive soil disturbance (modest soil disturbanceimproves the seedbed for later germination) andsubsequent erosion, stream crossings by roads orskidder trails, increased water temperature causedby removal of near-stream shade, and inadequate


forest regeneration for nutrient uptake afterharvesting.

Fundamental changes have occurred in themanagement of industrial forests in Maine, withpotential consequences for water quality. Withinthe past 40 years, year-round mechanizedharvesting has replaced winter harvesting usinghorses. Within the past 30 years, the network offorest roads has grown exponentially due to thedemise of the river drives. These changes havecreated new issues for water quality. While thequality of surface waters draining forestedwatersheds in Maine is generally excellent, timberharvesting near water usually affects downstreamwater quality and aquatic habitat in some manner.

Studies in Maine have documented thatindustrial-scale harvesting on unfrozen soils mayexcessively disturb the soil surface, increasing thepotential for erosion and runoff of sediments.Nitrate concentrations in streams have beendocumented as high as 90 mg/l after harvesting,compared to near zero values before harvesting.Nitrification can also lead to soil and streamacidification, which in turn results in highaluminum concentrations in soil solutions andsurface waters; concentrations which may be toxicto seedlings, roots, and aquatic organisms.Depending on compliance with Best ManagementPractices and the amount of biomass removed,harvesting may increase water yield by up to 80%,and increase summer surface water temperaturesby 5 to 27oF. Studies in Maine suggest thatexcessive disturbances can inhibit forestregeneration under some circumstances, whichmay prolong impacts on water quality.

Finally, the competition for water from Maine’s manywatercourses will inevitably increase over time. Maine’swaterways already support recreational and commercialfishing, boating, swimming, subsistence fishing, wildlife,industry, agriculture, development, municipal watersuppliers, the ski industry, golf courses, and commercialwater bottling. Day (2001:2) observes that:

For fisheries, enough flow at the right times isrequired to support nursery and spawning areas,migration, adequate temperatures andoxygenation. Wildlife such as birds rely on healthystreamside wetlands for food and nesting habitat.Adequate flows also provide increased assimilativecapacity, maintenance of natural streamflowpatterns, recharge groundwater aquifers, supportwetlands, maintain riparian and floodplainstability and habitat, and prevent the decline of

rare aquatic and riparian-habitat dependentspecies.

However, as Day (2001) cautions, water diversions andwithdrawals can dramatically alter the natural timing andquantity of streamflow, affecting:

• Instream and streamside habitat for fish and otheraquatic life, which in turn changes water quality andreduces spawning and rearing habitat and protectivecover from predation;

• Recreational uses; and• The state’s fisheries management goals and efforts to

manage and restore important fisheries such as brooktrout, bass, and Atlantic salmon.

The issue of water withdrawals and diversions clearlywarrants much greater attention as part of a comprehensiveplan for the conservation of Maine’s watercourses.

5. Disruption of ecological processes anddisplacement of native species as aresult of the introduction and spreadof exotic species.Note: The primary source for this information is Gawler et al.[1996] unless otherwise noted.

Exotic species have been ranked second only to habitatdestruction as a leading cause of extinction (Wilcove et al.1998). Exotics can pollute the integrity of regional faunasand floras, and alter fundamental ecological processes suchas nutrient cycles, predator/prey dynamics, and firefrequency and intensity (Noss and Cooperrider 1994). InMaine, almost one-third of the flora consists of introducedspecies. Some introduced plants (e.g., purple loosestrife[Lythrum salicaria]) have displaced native plants, thuslowering native biodiversity and disrupting nativeecosystems. In addition, some exotics may becompromising the genetic integrity of certain native plantspecies via hybridization. Indeed, at least five hybridsbetween introduced and native plant species have beenidentified in Maine.

Invasive aquatic species are of particular concern inMaine. A recent federal report indicated that 45 invasiveaquatic species have been reported in Maine (NationalInvasive Species Council 2000). Moreover, the opportunityfor the spread of these 45 current invasive species issignificant (Maine Invasive Aquatic Species Work Group2001).

Intensified use of Maine’s public waters (marine andfreshwater) for interstate and international commerce,recreational boating, fishing, aquaculture, tourism, andwaterfront development exposes Maine’s waters to invasive


species. The principal human activities contributing to thetransport and dispersal of these species in state watersinclude (Maine Invasive Aquatic Species Work Group2001):

• Transport and release from the bottom of boats, andboat trailers;

• Improper disposal or containment of aquarium plantsand animals;

• Movement or intentional release of aquaculture andfishery species along with their associated pathogens,attached free-living organisms and parasites;

• Recreational boating within and between waters;• Bait handling and movement throughout the state’s

inland waters;• Ornamental gardening and waterfront landscape

practices; and• Release of organisms from the ballast water of ships

and from hull surfaces.

Fish provide an especially striking example of theeffects of exotics on native biodiversity. Displacement byintroduced species has been implicated in the decline of68% of freshwater fish species that have become extinct orvulnerable to extinction in North America during the last100 years; hybridization with exotics is linked to thedecline of 38% of these extinct or vulnerable species(Wilcove et al. 1992).

In Maine, 25% of fish species are exotics, andintroduced species have been reported in almost all of thelakes in Maine. The native arctic charr (Salvelinus alpinus),for example, was eliminated from the Rangeley Lakesfollowing introductions of rainbow smelt (Osmerus mordax)

and landlocked salmon (Salmo salar sebago). Further,hybridization between wild Atlantic salmon and farmedsalmon as a result of aquaculture is of major ecologicalconcern in Maine today. In addition to changing thegenetic characteristics of wild Atlantic salmon, aquacultureelevates the risk of disease and parasite transfers. Infectioussalmon anemia and sea lice—widespread problems inEuropean salmon aquaculture—have recently appeared inNorth American salmon farms and could spread to wildsalmon (Naylor et al. 2001). Naylor et al. (2001) also warnthat:

In the New Brunswick-Maine region, farmedescapees vastly outnumber wild salmon in somespawning rivers (Goldburg et al. 2001). Theestablishment of farmed salmon in the wildincreases pressure on endangered native salmonpopulations. Even more pressure could arise iftransgenic salmon containing added growth-hormone genes are approved for commercial net-pen culture.

See the focal species profile for Atlantic salmon inSection V for more discussion of this species.

Minnows provide another salient example of impactsfrom exotic species: a recent study showed that, of 18native minnow species expected to occur in Maine lakes,the maximum number of species found in any one of the 47lakes assessed was 8. This finding was likely due to eitherinteractions between native and non-native minnows(which are dumped into the water as excess by anglers), orto predation by introduced game species.


III. MISSION, GOALS, AND OBJECTIVES

Section II profiled the major human impacts orecological wounds evident on the Maine landscape. Thesemay be summarized as follows:

1. Forest fragmentation, changes in forest stand dynamics(e.g., tree species composition, truncated forest age-class distribution), and disruption of natural ecologicalprocesses as a result of industrial forest managementpractices;

2. Loss of native species as a result of habitat alteration,hunting, and introduced species;

3 . Habitat fragmentation and destruction in southernMaine as a result of extensive residential andcommercial development;

4. Loss of freshwater ecosystem integrity and connectivityas a result of the construction and operation ofdams, water withdrawals, certain agricultural andforestry practices, and chemical pollution; and

5. Disruption of ecological processes and displacement ofnative species as a result of the introduction andspread of exotic species.

How do we address these impacts and link theirsolutions to the ultimate goal of the Maine WildlandsNetwork, which is to restore and protect the ecologicalintegrity of the state? It is useful to revisit the four keyconservation goals required to achieve the broader objectiveof ecological integrity (Noss 1992; Noss et al. 1997):

• Represent, within the system, all native ecosystemtypes and seral stages across their natural range ofvariation;

• Maintain, within the system, viable populations of allnative species in natural patterns of abundance anddistribution;

• Maintain, within the system, ecological andevolutionary processes, such as disturbance regimes,hydrological processes, nutrient cycles, and bioticinteractions, including predation; and

• Design and manage the system to be responsive toshort-term and long-term environmental change andmaintain the evolutionary potential of lineages.

Conservation biologists have debated for years themost effective means of achieving these ends. In a place likeMaine, which is missing several of its essential nativespecies and whose ecosystems have been highly altered, itwill be necessary to engage in an ambitious program ofrestoration. Unfortunately, much of what is now called

ecological restoration falls far short of the mark. Thecontributors to Continental Conservation write, “an aim of[the] Wildlands Project is to maintain the full range ofnative species and ecosystems. This is a much moreambitious goal than that of most restoration projects, andit is a pity that the field of restoration ecology cannot yetprovide a set of models and tools to ensure its attainment(Simberloff et al. 1999).”

Despite this lack of models and tools, the authorsnevertheless provide important guidance:

Whatever the restoration project, one must firstchoose a specific goal: the full range of nativespecies and ecosystems when? This “reference time”is the first decision. Second, given a reference time,one must determine what is acceptably close to thefull range of native species and ecosystems presentat that time…Restoration should be aimed atreturning to the point on this trajectory of changethat would have obtained in the absence of humandisturbance, rather than trying to replicate theprecise system that once was present (Atkinson1990; Simberloff 1990). Of course, specifyingwhat that trajectory would have been is anenormous scientific challenge. But in principle thisshould be our goal, and we must be able to statehow closely our goal must be approximated inorder for a restoration project to be called asuccess.

The work of Foster (2001), Lorimer (1977), and othersprovides a basic guide to the native species and ecosystemsof the presettlement forests of New England. While moreresearch is no doubt necessary, there is certainly enoughinformation to begin planning and putting into practicelandscape-scale restoration. Simberloff et al. (1999)recommend focusing wildlands restoration into three broadcategories:

• Control of invasive nonindigenous species;• Reestablishment of natural abiotic forces; and• Reintroduction or augmentation of extirpated or

imperiled native species.

Carried out at a broad, regional scale, these threecomplementary steps support the goal of restoringecological integrity in Maine. They also respond to severalof the major human impacts detailed in Section II. For thisreason, these restoration steps are incorporated into the


goals and objectives of the Maine Wildlands Network.Perhaps the most important of these steps is that ofreintroduction or augmentation of missing or depletednative species. There is increasing evidence that certain“keystone” species, such as large predators, play a criticalrole in ecosystem function. These predators, in turn, needlarge, connected protected areas in order to thrive. The“scientific argument for restoring big wilderness based onthe regulatory role of large predators (Soulé and Noss1998),” rewilding, is described in detail in Section IV.Because large predators and certain other native species areso important, they deserve special emphasis in the mission,goals, and objectives of the Maine Wildlands Network.

Mission of the Maine Wildlands NetworkThe mission of Maine Wildlands Network is to restore

and maintain the ecological integrity of Maine inperpetuity through the design and promotion of a systemof protected wildlands within Maine. This system willprotect the ecological integrity of existing wildlands andensure the restoration and protection of other areas,providing the habitats necessary for the long-term viability(i.e., hundreds to thousands of years) of native ecosystemsand species.

Goals of the Maine Wildlands NetworkThe following goals are essential to achieving the

mission of the Maine Wildlands Network. These goalsrespond to the human impacts described in Section II,incorporate the elements of regional and continentalrestoration as described in Simberloff et al. (1999), andreflect the four conservation goals set out by Noss (1992)and Noss et al. (1997):

• Restore and protect all native terrestrial and freshwaterecosystem types and seral stages across their naturalrange of variation;

• Recover and protect ecologically effective populations6

of all focal species throughout as much of the planningareas as possible and ecologically desirable;

• Restore and maintain ecological and evolutionaryprocesses, including wildfire, insect outbreaks,

6 Some species of animals—keystone species—are especially interactivewith their environment; “their disappearance or rarity leaves not just aphysical void in nature, but a functional one that can trigger chainreactions ending with degraded or simplified ecosystems…Recoverygoals for keystone predators and other highly interactive species mustconsider ecologically effective population density and distribution”(Soulé et al. submitted: 1). Ecologically effective is defined as “theattainment of densities sufficient to maintain ecological interactions or functionssuch that other species, ecosystems, and processes do not decline or disappear(M. Soulé pers. comm. 2002).” See Sections IV-V for a description of thefocal species methodology.

predator/prey dynamics, natural succession, and floodregimes; and

• Eliminate or control exotic species in terrestrial andfreshwater ecosystems.

Objectives of the Maine Wildlands NetworkThe goals above chart out a set of broad, strategic

changes needed at the landscape level. Many individualsand organizations working within their specific regions andwithin their areas of expertise (e.g., clean water, wilderness,endangered species, stewardship, land acquisition, organicand diversified farming) will be needed to bring aboutthese changes. Further, timelines for different projects willvary. Ultimately, the efforts needed to bring about afunctioning Maine Wildlands Network can be thought ofas an assemblage of many smaller initiatives, workingsynergistically to achieve the big picture.

The goals of the Maine Wildlands Network are long-term. Under each goal below is a set of medium-termobjectives. These objectives are oriented toward those withthe greatest control and influence over natural resourcesand their use in Maine, including state and federalgovernment agencies, private landowners, and land trusts.In order to achieve the long-term goals, these actorswill—in most cases—need to change how they nowmanage the natural resources under their control. Withadditional analysis, many of these objectives can bequantified and otherwise modified to make them morespecific. This process should help those carrying out theaction plan to focus their efforts. Incremental steps ormilestones should also be identified for each objective,aiding in monitoring and evaluation efforts. See Section IXand Appendix 3 for details on recommended conservationaction steps.

Goal 1: Restore and protect all native terrestrialand freshwater ecosystem types and seral stagesacross their natural range of variation.Objectives1. Significantly increase the amount of public and private

conservation land with a high degree of protection(core wild areas). Give protection priority to largeblocks of land with little or no human or road presenceand to ecosystem types that are not currently wellrepresented in the protected areas system of Maine.

2. Establish smaller core areas where necessary to improveecosystem representation and aid the recovery andmaintenance of focal species.

3 . Maintain or reinstate prohibitions on building andlogging in roadless areas of National Forests.

4 . Remove roads in areas of high ecological value onpublic lands and other conservation areas.

5. Increase the number of miles of officially designatedWild and Scenic rivers in the state.


6 . Protect significant portions of major rivers andwatersheds in the Wildlands Network, especiallyheadwaters and adjacent riparian zones.

7. Design and implement a removal strategy for the mostenvironmentally destructive dams.

Goal 2: Recover and protect ecologicallyeffective populations of all focal speciesthroughout as much of the planning areas aspossible and ecologically desirable.Objectives8 . Implement reintroduction programs and facilitate

natural recovery or recolonization for all identifiedfocal species.

9. Adopt hunting, fishing, and trapping regulations thatsupport the recovery of identified focal species; includeeliminating the incidental take of non-target species.

10. Increase general public awareness and acceptance ofwilderness and focal species—especially largecarnivores such as the wolf, Canada lynx, and mountainlion.

11. Significantly increase the amount of public and privateconservation land whose primary purpose is landscapeconnectivity (terrestrial linkages, aquatic linkages,aquatic buffers) and wildlife movement (feeding areas,stop-over points, and movement zones— especiallyareas with minimal human populationand road density).

12. Within these connection zones, establish managementpractices conducive to connectivity, includingmaintenance of appropriate habitat and minimizationof harassment of focal species.

Goal 3: Restore and maintain ecological andevolutionary processes, including wildfire, insectoutbreaks, predator/prey dynamics, naturalsuccession, and flood regimes.Objectives13. Carry out management actions as necessary to restore

the full functioning of abiotic and biotic processes incore wild areas.

14. On compatible-use lands (such as private commercialforestlands surrounding cores), establish high standardsfor forest integrity, and implement ecologically soundforest management practices, including restoration ofpreviously logged areas.

15. On state and federal lands, increase the use ofmanagement practices that promote natural ecologicaland evolutionary processes, as opposed to ones thatfavor timber and game production.

Goal 4: Eliminate or control exotic species interrestrial and aquatic ecosystems.Objectives16. Implement a comprehensive statewide management

program to control and prevent the spread of exoticplant and animal species.

As discussed more fully in sections VI, VII, and VIII,the goals and objectives of the Maine Wildlands Networkare aimed for now at the northern, western, andnortheastern portions of the state. Core areas proposed forsouthern Maine are insufficient in number and size tosatisfy the requirements of the focal species used in thedesign. This is, in turn, partially a result of limitations inour methodology’s ability to deal with this region’srelatively large human population, large-scaledevelopment, and high road density. Other conservationprojects currently underway in Maine are focusing on thissouthern region. Nevertheless, as the Wildlands Projectand others move forward with the Maine WildlandsNetwork, efforts will continue to address biodiversityconservation in southern Maine.


IV. WILDLANDS NETWORKDESIGN APPROACH

To satisfy the goals and objectives of the MaineWildlands Network, it is necessary to design andimplement a comprehensive protected areas system. Arewilding approach to conservation planning—with itsemphasis on large core areas, connectivity among thosecores, and maintenance and restoration of ecologicalprocesses—provides a guide for defining the system and itsvarious components. To determine the location and size ofthese components, we use a three-track approach thatincorporates 1) special elements, such as large wetlandcomplexes; 2) focal species planning; and 3) representationof key landscape features, in an attempt to capture thegreatest amount of biodiversity and provide for the long-term viability of populations at all scales (Noss et al. 1997;Noss et al. 1999).

RewildingRewilding is characterized by 1) large, strictly

protected, core reserves (“the wild”), 2) connectivity, and 3)keystone species and processes (Soulé and Noss 1998).Soulé and Noss (1998:22) propose that:

Three major scientific arguments constitute therewilding argument and justify the emphasis onlarge predators. First, the structure, resilience, anddiversity of ecosystems are often maintained by“top-down” ecological (trophic) interactions thatare initiated by top predators (Terborgh 1988;Terborgh et al. 1999; 2001). Second, wide-rangingpredators usually require large cores of protectedlandscape for secure foraging, seasonal movement,and other needs; they justify bigness. Third,connectivity is also required because core reservesare typically not large enough in most regions;they must be linked to insure long-term viabilityof wide-ranging species. In addition to largepredators, migratory species such as caribou andanadromous fishes also justify connectivity in asystem of nature reserves. In short, the rewildingargument posits that large predators are ofteninstrumental in maintaining the integrity ofecosystems; in turn, the large predators requireextensive space and connectivity.

In addition to its role in restoring terrestrialecosystems, Soulé and Noss (1998:24) propose rewilding asa “human responsibility,” and suggest that:

By insuring the viability of large predators, werestore the subjective, emotional essence of ‘thewild’ or wilderness. Wilderness is hardly ‘wild’where top carnivores, such as cougars, jaguars,wolves, wolverines, grizzlies, or black bears, havebeen extirpated. Without these components,nature seems somehow incomplete, truncated,overly tame. Human opportunities to attainhumility are reduced.

Much of the support for the rewilding approach isbased on recent studies suggesting that ecosystem integrityis often dependent on the presence of large carnivores(Terborgh et al. 1999; 2001; Estes et al. 1978; Estes et al.1998; Crooks and Soulé 1999; Henke and Bryant 1999;Ripple and Larsen 2000). Soulé and Noss (1998:22)explain:

Studies are demonstrating that the disappearanceof large carnivores often causes these ecosystems toundergo dramatic changes, many of which lead tobiotic simplification and species loss (Mills et al.1993). On land, these changes are often triggeredby exploding ungulate populations. For example,deer, in the absence of wolves and cougars, havebecome extraordinarily abundant and emboldenedin many rural and suburban areas throughout theUnited States, causing both ecological andeconomic havoc (McShea et al. 1997; Nelson 1997;McLaren and Peterson 1994).

From studies conducted in South America, Terborgh etal. (1999; 2001) have shown that the absence of carnivorecontrol on herbivores (e.g., tapir, monkeys, rodents, insects)can precipitate a rapid loss of plant species diversity. Forexample, construction of a reservoir in Venezuela causedflooding of a vast area, now known as Lago Guri. Many ofthe newly created islands lack larger predators such asjaguar, puma and harpy eagle, and subsequently, thereproduction and replacement of many species of canopytrees have come to a halt. Further, on middle-sized islands,even though 60-70 species of trees coexist in the canopy,only a handful of species are represented in young recruits.Terborgh et al. (1999; 2001) speculate that the primaryfactor in the failure of canopy trees to reproduce is thesuperabundance of herbivores (e.g., leaf-eating monkeysand ants, rodent seed predators). The herbivores have


apparently been “released” from the population controlimposed, directly or indirectly, by large predators. As aresult, the entire island ecosystem is crashing.

Ripple and Larsen (2000) suggest that elk browsinghas had a negative effect on the recruitment of aspen intothe overstory in Yellowstone National Park, most probablybecause elk populations and associated browsing intensityincreased significantly after gray wolves were eliminatedfrom the park in the early 1900s.

In the Midwest and in the deciduous and boreal forestsof the eastern U.S. and Canada, overbrowsing by white-tailed deer and, increasingly, moose, is causing changes inplant species compositions in many ecosystems (Waller andAlverson 1997; DeCalesta 1994; McShea and Rappole1997; Connor et al. 2000). The absence of effectivepredation by native carnivores, particularly wolves, isprobably a major—if not the principal—cause of thischange (Soulé et al. submitted). The impacts of white-taileddeer in the Midwest and the forests of the eastern U.S. arewell known. Among the effects frequently observed are:

• Local extirpation of many herbs and tree seedlings,including important commercial species like oak,white pine, and northern white cedar (Waller andAlverson 1997; McShea et al. 1997).

• The transformation of understories from mixedherbaceous species to park-like understories dominatedby grasses and ferns (Waller and Alverson 1997).

• D e c l i ne s in the abu n da nc e and di v e r s i t y of fo r e s t s o n gb i r d s (D e Ca l e s t a 199 4 ; Mc S h e a an d Ra p po l e 19 9 7) .

In short, recent evidence from studies in a diverserange of natural communities indicates that large predatorsare probably instrumental in maintaining the integrity ofecosystems. Given these landscape effects, it seems prudentto restore natural predation wherever possible (Soulé et al.submitted).

Two top carnivores, the eastern cougar and easterntimber wolf, have been extirpated from Maine. Althoughthese two carnivores prey extensively on ungulates, both areprey generalists. Their absence in Maine’s forests could beexpected to have far-reaching effects on everything fromdeer/moose/vegetation dynamics (e.g., Ripple and Larsen2000) to coyote and fox abundance. Thus, large, wide-ranging predators play a major role in the proposed MaineWildlands Network.7

7 Although the wolverine (Gulo gulo) is perhaps the widest-ranging ofthe large predators, and although a functioning Maine WildlandsNetwork should help restore this ultimate wilderness indicator, it is notspecifically addressed in this document for three reasons: 1) Littleinformation is available on the original range or habitat needs of thewolverine; 2) being a scavenger and a predator of smaller animalsgenerally, the wolverine may not play the top-down regulatory role thatwolf and cougar play; 3) global climate change may make a future Mainelandscape inhospitable to wolverine and other boreal species.

Core, Linkage, Compatible-Use Area ModelThe Maine Wildlands Network Design uses an

extension of the biosphere reserve concept (UNESCO 1974)as refined by Noss (1983, 1987b, 1992, 1995), Noss andHarris (1986), Noss and Cooperrider (1994), and Noss etal. (1997) to define its protected area system. As adaptedfrom the biosphere reserve concept, a wildlands networkuses three basic categories of protected lands: core wildareas, wildlife linkages (or corridors), and compatible-uselands (or buffer zones) (Figure IV.1). In this networkdesign, we employ two of these categories: cores andlinkages. We also identify “study areas” that could later bedesignated core or compatible-use lands. At this time wedo not propose compatible-use designations, although wehope that gentler forestry practices on commercialtimberlands will enable these extensive holdings to servethat purpose. These categories and their use in our designare discussed below.

Figure IV.1 Components of a Wildlands Network.

Core Wild AreasCore areas are expansive wildlands in which ecological

processes are allowed to function as naturally as possible.They are wilderness areas—self-willed lands where Nature,not human agency, directs the ebb and flow of life. Somemanagement of cores may be necessary in the short term, tocompensate for historical alteration of natural processes(e.g., disruption of natural fire and flood regimes,introduction of exotic species) until a time when suchprocesses resume their natural function. Core areas are thebackbone of a wildlands network, and are designed toprotect those landscape features that are nearlyirreplaceable, or to serve an irreplaceable function in thelarger ecosystem (e.g., to protect populations of keystonespecies) (Noss 1995; Noss et al. 2001; Noss et al. 2002).


Core areas are not “human exclusion” zones, but allowfor human uses that are compatible with the protection andfunction of the core (similar to how some national parks aremanaged). Such activities might include hiking, non-destructive/invasive research, hunting, fishing, andcamping. Motorized on- or off-road uses would not becompatible with maintaining the protection and functionof core areas (see Webb and Wilshire 1983 for a summaryof the ecological effects of off-road vehicles; Havlick 1998).Cores could be established and maintained over the long-term on private land, but there must be some guarantee,such as a “forever wild” easement that runs with theproperty, that the land will be permanently protected forecological integrity.

Core areas were identified in large part based on theresults of a spatially explicit GIS analysis. We used thereserve selection software SITES 1.0 (Andelman et al.1999) to systematically identify a set of areas within thestate that would meet planning targets related to MaineWildlands Network Design’s goals and objectives. TheSITES program and our application of it to the MaineWildlands Network Design are described in Section VI.

Wildlife LinkagesWildlife linkages are designed to facilitate movement

of animals (migration, breeding, foraging), plantpropagules (seeds, pollen, pollinators), and essential abioticresources such as water and nutrients, and permit large-scale and long-term ecological processes to continueoperating within fragmented ecosystems. Further, linkagesprovide connectivity between protected areas to helpfacilitate the successful migration of both plants andanimals as climatic conditions change over time.Functional connectivity is necessary to prevent or mitigatedeleterious population effects associated withfragmentation, and to ensure the viability of wide-rangingspecies that require extensive habitat for foraging, seasonalmovement, and other needs (Noss 1983; Harris 1984; Nossand Harris 1986; Soulé 1987; see Dobson et al. 1999 for adetailed discussion of linkage design and function).

Noss et al. (1997) suggest four specific guidelinesarising from the connectivity principle:

• All else being equal, wide swaths of suitable habitatare better than narrow corridors.

• Corridors longer than normal dispersal distances for atarget species should be sufficiently wide or haveenough “stepping stone” habitat patches to provide forresident individual home ranges.

• Animals usually follow a path of least resistance whenmoving through a landscape.

• Planners should base connectivity designs on the needsof species most sensitive to fragmentation.

The Maine Wildlands Network follows theseguidelines whenever possible. In some instances, especiallyin areas with high road and human density, linkages followriparian corridors or ridgelines. Three types of linkages areproposed:

Terrestrial LinkagesTerrestrial linkages have been designed to support the

movement of wide-ranging species and species sensitive todisturbance and human exploitation. They are typically“overland” corridors—not intentionally designed aroundwater features. The locations of these linkages are based onhabitat quality or physiological features (e.g., ridgelines),or are in areas that minimize exposure to disruptions suchas roads and human population centers). Terrestriallinkages should be managed to permit ecological processesto function as naturally as possible. As their primarypurpose is to allow the movement of species that aresensitive to human disturbance, their protected statusshould be similar to that of proposed core areas. Ideally, thedesign of linkages should also be based on evidence ofactual use by wildlife. However, such information currentlyis unavailable for many areas in Maine.

Aquatic LinkagesAquatic linkages are intended to protect biodiversity

associated with aquatic ecosystems, as well as to providehabitat for terrestrial species that use riparian zones formovement or dispersal. Further, protected linkagesincorporating entire subwatersheds will help protect theintegrity of the streams running through them, and willserve a vital role in moving water and dissolved nutrientsbetween habitats (Dobson et al. 1999). Maine possesses anextensive river system that serves as valuable habitat formyriad species. In other regions, such riparian habitatshave been shown to have considerable value as naturallinkages (Naiman et al. 1993). Aquatic linkages for MaineWildlands Network Design were identified primarilyusing water features (e.g., potential salmon habitat,subwatersheds of large rivers) as opposed to terrestrialfeatures. These linkages were delineated by selectingsubwatersheds that occur in areas of very low humandensity (≤5 people/km2).

Although the management of aquatic linkages willneed to be determined on a case-by-case basis, theoverriding goal should be to maintain ecological integrity,which will be essential for providing functionalconnectivity between core areas. The protected status oflinkages should be sufficiently high to ensure theattainment of this objective.


Aquatic BuffersAquatic buffers consist of subwatersheds of major

rivers that pass through areas of higher human density (>5people/km2). Although these subwatersheds are likely to beas ecologically important as, if not more important than,those in other areas with lower human populations, it isunrealistic to assume that significant restoration of naturalhabitats will be feasible—even in the medium to long term(e.g.100 years)—given current patterns of development andhuman habitation. Thus, we propose that the managementof aquatic buffers be case specific and dependent on anumber of factors, including human access and the value ofcompeting uses. Nonetheless, emphasis should be placedon promoting practices that ultimately contribute to theeffectiveness of these buffers as travel corridors and/or asprotective buffers for aquatic habitats. Implementation inthese areas will likely need to focus extensively on privateland stewardship.

Compatible-Use LandsCompatible-use lands can help protect core areas and

critical linkages from the ecological impacts of humanactivities while allowing for limited human economic uses.Compatible-use areas will allow human uses compatiblewith the protection of cores, such as hunting and fishing,low-density housing, ecological agriculture and forestry,and light tourism. These areas would be ideal places toapply positive tax and market-based incentives forconservation, such as carbon banking. These lands may beowned and managed through a wide variety of publicand/or private programs, but specific managementdecisions should be based on their purpose relative to thecores or linkages they protect.

This proposed Maine Wildlands Network does notinclude compatible-use lands, because the identification ofsuch areas requires information about specific threats toproposed core areas. Because there are large gaps inknowledge about the habitat quality and speciescompositions of Maine’s natural communities, the abilityto assess the need for compatible-use areas around proposedcores is beyond the scope of this document. The WildlandsProject will conduct studies aimed at defining and locatingcompatible-use lands for Maine and the greater northernAppalachians in the context of the larger regionalwildlands network design project.

Study AreasThe forests of northern and western Maine offer

many opportunities for the establishment of large,interconnected core areas. Indeed, forest coverage is soextensive that a major challenge is identifying which areasshould be given the highest priority. As discussed ingreater detail in Section VI, we use a computerized siteselection algorithm to help us identify the highest priorityareas. These are designated as proposed core areas andlinkage zones. The same algorithm allows us to identifyother areas that should be included in the network, butwhich require further study to determine if they should beconsidered potential core, compatible use, or linkage areas.These are designated study areas.

Three-track Wildlands Network DesignMethodology

The Maine Wildlands Network Design planningprocess uses a three-track methodology (Noss et al. 1997)to design and evaluate the proposed wildlands network.These three tracks—special element mapping, focal speciesplanning, and representation analysis—are rarely combinedin practice, yet all three tracks are critical for effectiveconservation area selection and network evaluation.

1. Special Element MappingSpecial elements are sites that are deemed important

for inclusion in a proposed protected area, and havetypically included such elements as rare speciesoccurrences, roadless areas, endangered or criticalecosystems, and critical watersheds. We use the term“special element” to refer to relatively discrete occurrencesor habitats that represent areas of concentrated conservationvalue (Noss et al. 1999, 2001). Maine Wildlands Networkdesign incorporates the following special elements:

• Large wetland complexes;• Areas with high variability in slope;• Large roadless areas (e.g., areas > 5 km from nearest

class 1-3 road).

Special element mapping often includes rare,threatened, and endangered (RTE) species. Good RTE dataexist but, because most of Maine is private land, it isdifficult to obtain due to confidentiality agreementsbetween the state government and private landowners.Therefore we have not incorporated these data. It ispossible, however, that such data will be available in someform for the regional wildlands network design nowunderway.


2. Focal Species PlanningFocal species planning complements special elements

and representational approaches to wildlands networkdesign. Special elements and representation help identifywhich areas to include in the network, whereas focal speciesare used primarily to help address how large the networkcomponents should be and how they should be configured(Miller et al. 1998). A carefully selected set of focal species,effectively representing a broad range of life requisites, canbe seen as surrogates for the protection of many otherspecies.

Focal species are important as “ends” as well as“means” (Soulé and Noss 1998). That is, in addition totheir use in planning protected areas, the simple presenceof some focal species in a landscape will likely contribute toits ecological integrity and wildness. For example, Souléand Noss (1998) suggest that: “Once large predators arerestored, many if not most of the other keystone and‘habitat-creating’ species (e.g., beavers), ‘keystoneecosystems’ (deMaynadier and Hunter 1997), and naturalregimes of disturbance and other processes will recover ontheir own.” For this design we profiled nine species—wolf,lynx, cougar, marten, otter, goshawk, red-shouldered hawk,loon, and salmon. We focus on the habitat requirements oftwo large, wide-ranging carnivores—wolf and lynx—andthe subwatersheds of salmon-bearing rivers, in good partbecause the data for these three species is the most rigorous.See Section V and Appendix 3 for details of focal speciesselection and justification. A much more extensive focalspecies analysis will be carried out for the larger regionalwildlands network.

3. Representation AnalysisMany protected area assessments, including this one,

use “inclusion” models in an attempt to represent thecomplete spectrum of certain physical or landscape features(e.g., natural communities, physiography, vegetation) inthe final set of protected areas. One assumption of thisapproach is that the feature represented will be a “coarsefilter” serving to ensure the representation of many speciesfor which few data exist. The following features are assessedin Maine Wildlands Network design:

• Maine ecoregions (15 biophysical regions found withinthe state);

• Major watersheds;• Terrestrial vegetation and land-cover;• Aquatic vegetation.

Assumptions and LimitationsThe team that developed the proposed Maine

Wildlands Network was guided by several assumptions.Despite our efforts to use up-to-date data, approaches, andtechniques, the proposal does not address all issues relatedto conservation planning in Maine. Here we list theassumptions that guided our work. We hope that byidentifying these, reviewers and users will be better able toevaluate and understand the wildlands network design.

Assumptions• The Maine Wildlands Network applies the science

of conservation biology to land use planning andmanagement and is heavily informed by biologicaldata, current ecological theory, and expertknowledge. It is science-based in that it recognizesthat keystone species, in particular largecarnivores, are often necessary for ecosystemintegrity, and that reserves will need to be largeand interconnected to adequately protect shy,sensitive, and wide-ranging species (Soulé andNoss 1998; Soulé and Terborgh 1999; Soulé et al.submitted; Noss 1992).

• The Maine Wildlands Network does not analyzenor directly address the needs of all native speciesnor even all sensitive, threatened, and endangeredspecies. By focusing on (1) the needs of severalprimary focal species (wolf, lynx, and salmon);(2)special elements, and (3) ecosystem representation,we hypothesize that the network, if implemented,will protect nearly all of the native species,including extirpated species, in the northern two-thirds of the state.

• The Nature Conservancy’s effort to prioritize largematrix-forming forest systems and ecological landu n i t s t h r o u g h o u t t h e n o r t h e r nAppalachian/Acadian ecoregion identifies majornatural communities and special elementoccurrences, and strongly complements theanalysis conducted here. TNC’s analysis will beuseful in our upcoming regional wildlandsnetwork design.

Limitations• The proposed Maine Wildlands Network is

restricted to the state of Maine. We recognize thatelements and criteria used in its design (e.g.,species distributions and restoration of large scaleecological processes) may be more meaningfulwithin a larger, regional context. The WildlandsProject has initiated a regional wildlands network


design process that will stretch from the Tug HillPlateau and Adirondacks in north-central NewYork, north into Ontario and east to Vermont,New Hampshire, Maine, New Brunswick andNova Scotia.

• Nevertheless, Maine’s size (relative to othernortheastern states), central location within thenorthern Appalachian/Acadian ecoregion, andlarge expanses of unpopulated territory justify theattention that this product represents. Perhapsmost importantly, this design represents a test ofthe analytical techniques that will be applied at amuch larger scale.

• The Maine Wildlands Network does notincorporate Natural Heritage data identifyingclusters of rare, threatened, or endangered (RTE)species. These data, though requested, were notavailable. We will consider the incorporation ofNatural Heritage data in the upcoming largerregional wildlands network design.

• Focal species inputs used in the Maine WildlandsNetwork Design were limited to predicted wolfand lynx habitat and subwatersheds of salmon-bearing rivers. We felt that the models and dataavailable for other focal species, including cougar,marten, river otter, goshawk, red-shoulderedhawk, and loon were not sufficiently rigorous toinclude in the analysis. The data for wolf and lynxare strong, however, and given that these two arekeystone species (see Section V), we feel that theyprovide adequate inputs, especially whencombined with the accurate representation andspecial elements data.

• The Maine Wildlands Network is not based onanalyses of area needed to maintain regionallyviable populations of the lynx and wolf, either inthe short or long term. Such population viabilityanalyses will be developed for the upcomingregional wildlands network design. In the case ofwolf, lynx and marten, dynamic, individual-basedmodels will provide information on extinction riskand viability. More traditional static habitatmodels that incorporate minimum areaconsiderations will be also be created for a numberof other focal species as part of the regional design.

• The southern coastal counties of Maine are tooheavily populated and fragmented by major roadsto support any part of the Maine WildlandsNetwork, even under rosy restoration scenarios.

Nevertheless, this coastal region has a highoccurrence of threatened and endangeredvertebrates (Krohn et al. 1998) and facessignificant development pressures, as discussed inSection II, and has few GAP 1 or 2 lands; so thisarea is arguably a prime candidate for conservation,albeit at a different scale than that envisioned forthe northern and western portions of the state.Other conservation projects currently underway inMaine are focusing on this region. The WildlandsProject intends to monitor their progress as well asassist in their design and implementation whenpossible.

• Coastal species and marine habitats are notexplicitly addressed in the Maine WildlandsNetwork design, largely because less is knownabout coastal and marine species. In addition, thecoastal regions of Maine typically have higherhuman and road densities than other regions. Wewill continue to communicate with initiatives thatfocus on coastal regions (e.g., USFWS Gulf ofMaine Project), in order to incorporate newinformation into regional planning activities.

• Because there are still very large gaps inknowledge about the ecological condition of manyimportant areas in Maine, the ability to designcompatible use areas around proposed cores isbeyond the current scope of this project. Futurestudies aimed at defining and locating these areasare recommended.

• We recognize that a monitoring and evaluationprogram is critical to both determining whetherprogress is being made on achieving the goals of aconservation plan, and evaluating and revising thewildlands network design and conservation plan. Amonitoring and evaluation program will bedesigned and implemented at the regional level.

• The Maine Wildlands Network does not contain adetailed analysis of the social and economicimpacts of implementing this network. Suchanalyses will be a component of the regionalnortheast wildlands network design. Power’s(2001) recent study of the potential economicimpacts of the proposed North Woods NationalPark is a solid step in this direction.

• To the extent possible, Maine or Northeast-specificinformation was used, but where such informationwas unavailable, results from other areas wereconsidered better than no information at all.


The wildlands network design approach presented aboverepresents a synthesis of the major advances in conservationbiology over the last ten years. There is a strong consensusthat effective landscape-scale conservation will depend onlarge, strictly protected core wild areas that include a fullcomplement of keystone species—especially largecarnivores—and in which natural ecosystem processes areallowed to proceed unimpeded. These large cores must befunctionally linked to facilitate movement of animals andessential abiotic processes. In designing core-linkagenetworks, many practitioners have focused on just a few

(major) elements, such as rare species occurrences, roadlessareas, endangered or critical ecosystems, and criticalwatersheds. The Maine Wildlands Network takes these“special elements” and adds focal species, such as wolf andlynx, and a representational analysis, to produce a muchmore comprehensive model. The design also overcomes thechallenges of integrating such a large amount ofinformation by taking advantage of recent advances ingeographic information systems and spatially explicitanalysis techniques. The result is a core-linkage designmethodology that is transparent, analytical, and robust.


V. FOCAL SPECIES PLANNINGThe emphasis that this plan, and the Wildlands

Project in general, places on focal species stems frompersuasive evidence that large predators and other“keystone” species (see definition below) play anextraordinarily important role in ecosystem functioning. AsSoulé et al. (submitted) observe, “where the density of akeystone species falls below some threshold, the diversity ofspecies in an ecosystem virtually always decreases,triggering ecological chain reactions ending with degradedor simplified ecosystems (Estes and Palmisano 1974;Terborgh et al. 1999; Crooks and Soulé 1999; Jackson etal. 2001; Terborgh et al. 2001). It is important, therefore,to identify those endangered or regionally extirpatedspecies that interact strongly with others so that they canbe restored to levels of ecological effectiveness (Soulé et al.submitted). Among such keystones are large predators suchas wolves, jaguars, and sea otters (Estes and Palmisano1974; McLaren and Peterson 1994; Terborgh et al. 1999;Ripple and Larsen 2000). As discussed in Section IV, inorder to ensure the long-term existence and functioning ofpopulations of focal species, especially large predators, it isnecessary to protect and restore big, interconnected wildareas. And because large predators require extensive spaceand connectivity, the modeling of their habitatrequirements on the landscape is a key tool in wildlandsnetwork design (Soulé and Noss 1998). Miller et al.(1998:82) elaborate:

Focal species are organisms used in planning andmanaging reserves because their requirements forsurvival represent factors important to maintainingecologically healthy conditions. Ultimately,questions of ecological patterns and processescannot be answered without reference to thespecies that live in that landscape (Lambeck 1997).Representation and special elements themes pointto which areas should be considered in a reserve,but focal species analysis identifies additionalhigh-value habitats and addresses the questions:“How much area is needed? What is the quality ofhabitat?” and “In what configuration should wedesign components of a reserve network?”

Focal Species Categories for the MaineWildlands NetworkFocal species selected for the Maine Wildlands Networkbelong to one or more of the following categories:

• Keystone: Keystone species are defined as relativelyrare (low density), and having particularly strong,ramifying interactions (Paine 1969; Mills et al.1992; Power 1992; Power et al. 1996), withimpacts that are disproportionate to theirpopulation densities (Mills et al. 1992, Power etal. 1996) and not wholly duplicated by otherspecies (Kotliar 2000).

• Umbrella: Species that generally cover large andecologically diverse areas in their daily or seasonalmovements (Miller et al. 1998) such thatprotecting their habitat should provide adequatespace and resources for myriad species that havemore modest area requirements but with whichthey are sympatric (share the same range). Thecarnivores that figure prominently in WildlandsProject designs fall into the area-limited and, insome cases, the dispersal-limited categories ofumbrella focal species suggested by Lambeck(1997). The most sensitive species in thesecategories are assumed to serve as umbrellas forother species with less demanding spatialrequirements. Other focal species that may serveumbrella functions include resource-limitedspecies, sensitive to the availability of resources,and process-limited species, sensitive to thefrequency, intensity, extent, or timing of naturalprocesses such as disturbance (Lambeck 1997).

• Habitat Quality Indicator: Species that requirenatural habitat of high ecological integrity andthat may provide an “early warning system”because of their sensitivity to ecological changes(Miller et al. 1998). Prey species are also includedin this category as they indicate habitat quality forthe predators that rely on them.

• Foundation: A species that, like keystone species,enriches ecosystem function in a unique andsignificant manner, but that occurs at much higherdensities. Examples from various parts of the worldinclude Atlantic salmon (Salmo salar), bison (Bisonbison), prairie dogs (Cynomys spp., Miller et al.1994), cottonwood and aspen trees (Populus spp.),American chestnut (Castanea dentata), cod (Gadusspp.), krill (Euphausia spp.), bees, and mound-building termites. Soulé et al. (submitted) note that“by definition, species that are typically abundantor dominant, such as fig trees, salmon, coral and


social insects including termites and ants, thoughoften critical interactors, are not classified askeystone species, even though the effects [onecosystem functioning] are similar when they aregreatly diminished in abundance.”

Focal Species SelectionThe Maine Wildlands Network’s focal species were

selected using the following two-phase process:

Phase I: Literature Review and Expert InputA preliminary list of focal species was developed using

information from the scientific literature. This list was thenrefined by biologists, naturalists, and others withknowledge of Maine’s biodiversity.

Phase II: Survey of Regional ExpertsThe literature review and expert input were

complemented by a framework used to identify focalspecies based on a survey of regional experts (Beazley1998). This framework aims to identify areas of consensusamong experts regarding whether a given species meetscriteria designed to characterize the species’ “value” as afocal species. The survey process ultimately should identifymultiple species deserving special management attentionbecause of their value to wildlands network designplanning and ecosystem integrity.

The categories and criteria used in this survey include:

Keystone/Functionally Important Species• Top predator, large carnivore or mesocarnivore;• Important prey species, or species that provide

important resources for other wildlife;• Species that transform landscapes or waterscapes

[in the sense described by Soulé et al. (submitted)];

Umbrella Species• Space-demanding/wide-ranging;• Migratory species;• Requires specialized or defined habitat, or sensitive

to edge effects.

Habitat Quality Indicator Species• Limited by dispersal ability;• Limited by resource availability;• Limited by ecological processes such as fire, flood,

grazing;• Sensitive to stresses such as acid precipitation,

pollution;• Sensitive to disturbance by humans, roads, or

exploitation;

Vulnerable Species• Small population in Maine;• Population declining in Maine;• Small geographic range/distribution in Maine;• Decline in range/distribution in Maine.

Information Availability• Extensive information available (presence/absence

data)• Intensive information available (species-habitat

relationship data)• Ongoing study or monitoring effort for species

The last two categories (Vulnerability, InformationAvailability) were included to help explain the results, andwere not used in the scoring or results summarized below.

Of the 27 surveys sent to individuals who agreed toconsider participating, 11 (40.7%) were returned. Anumber of participants responded to only one taxonomicsection or completed the matrix for only a few species.Because of the small response sample size for most species,the results of this survey should be interpreted cautiously.Nevertheless , there were some areas o fconsensus—especially regarding mammals.

The results of this survey (summarized in Table V.1)identify species that possess characteristics that shouldmake them useful as focal species. A number of thesespecies had already been selected as Maine WildlandsNetwork focal species through literature review and expertopinion.


Table V.1 Species scoring highest in the Maine Focal Species Survey.*

Score Mammals Birds Amphibians/ Reptiles Freshwater Fish

Canada lynx(Lynx canadensis)

Bald eagle(Haliaeetusleucocephalus)

Blandings turtle(Emydoidea blandingii)

Atlantic salmon(Salmo salar)

River otter(Lutra canadensis)

Common loon(Gavia immer)

Common box turtle(Terrapene carolina)

Golden eagle(Aquila chrysaetos)

Spotted turtle (Clemmys guttata)

Hig

her

Peregrine falcon(Falco peregrinus)

Wood turtle(Clemmys insculpta)

Eastern cougar(Puma concolorcouguar)

Osprey(Pandion haliaetus)

Brook trout(Salvelinusfontinalis)

Eastern timber wolf(Canis lycaon)

Med

ium

-Hig

h

Woodland caribou(Rangifer taranduscaribou)

Low

-M

ediu

m Merlin(Falco columbarius)

Beaver (Castor canadensis)

Cooper’s hawk(Accipiter cooperii)

American black bear(Ursus americanus)

Northern goshawk(Accipiter gentilis)

American marten(Martes americana)

Northern harrier(Circus cyaneus)

White-tailed deer(Odocoilius virginianus)

Sharp-shinned hawk(Accipiter striatus)

Low

er

Upland sandpiper(Bartramia longicauda)

*Levels correspond to total scores. Species within a taxon (e.g., mammals) that shared the same scoreshare a level. See Long and MacKay 2001.


Other species ranked highly in the survey, but werenot chosen as focal species for the Maine WildlandsNetwork—primarily because their habitat needs greatlyoverlap with species already selected as focal species.

We use vertebrates exclusively because they tend tohave large area requirements. Theoretically, areas protectedfor vertebrates would tend to encompass the specifichabitats needed for many invertebrates and other taxa(Noss 1991; Simberloff 1998; Miller et al. 1998; but seeAndleman and Fagan 2000 for a critique of the umbrellaspecies assumption).

Summary of Profiled SpeciesUsing the process discussed above, we identified the

species in Table V.2 as the focal species for inclusion in theMaine Wildlands Network design. Focal species planningrequires species-specific information concerning habitatneeds, life history characteristics, area requirements,sensitivities and threats, reasons for inclusion as a focalspecies, and recommendations for conserving or recoveringthe species. Appendix 3 presents profiles for each MaineWildlands Network focal species and summarizes brieflythe species’ historical and current status, ecology and

habitat needs, focal value, and recommendations for itsconservation or recovery. The profiles are not meant toserve as a comprehensive literature review. As noted inSection IV, there is a good deal of variation in the qualityof the static habitat suitability models for the focal specieschosen for the design. Therefore, we have chosen to includeonly the best habitat models, namely those for wolf andlynx, plus subwatersheds of salmon-bearing streams, in theselection process described in Section VI. Nevertheless, therecovery and protection of all identified focal speciesrepresents a central goal of the Maine Wildlands Network.Other focal species, such as additional habitat qualityindicators, will become more significant to theconservation planning process as the network isimplemented and monitored at the regional and locallevels. Planning at the local level should take into accounttargeted community types and habitat features linked tothe requirements of focal species, such as canopy closure,patch size, presence of woody debris, or presence of vernalpools. The species not modeled in this analysis will beincluded in the more extensive regional wildlands networkdesign now underway.

Table V.2 Focal species profiled in the Maine Wildlands Network Vision.

Focal Species Keystone Umbrella HabitatQuality

Indicator

Foundation

Mammals Eastern timber wolf* X X Canada lynx* X Eastern cougar X** X American marten X X River otter X X Birds Northern goshawk X X Red-shouldered hawk X X Common loon X Fish Atlantic salmon* X X X * Predicted habitat included as a data input in the conservation area selectionprocess. ** Indicates that the species may play a keystone role, either alone or in concertwith other species (see descriptions in Appendix 3).


VI. CONSERVATION AREASELECTION PROCESS

The process of wildlands network designfundamentally addresses two questions: “where should thenetwork components (especially cores and linkages)specifically be located?” and, “how large should thenetwork be?” In this section, we describe our use of theconservation area selection software SITES 1.0 (Andelmanet al. 1999) to systematically identify a set of areas withinthe state that would meet planning targets related to theMaine Wildlands Network’s goals and objectives. InSection VII, this initial set of areas, along with a number ofother input layers, are used to identify a proposedwildlands network consisting of core areas and linkages.

The SITES Selection AlgorithmConservation assessments and wildlands network

designs typically have depended on manual mapping todelineate network components, and on somewhatsubjective scoring procedures to compare and prioritizethese components (see Noss et al. 2001 for examples ofthese procedures). We used a more systematic and efficientsite selection procedure to integrate the large number ofmap layers and the large size and diverse types of data setsin the wildlands network design process. We used thereserve selection software SITES (v1.0) to assemble aninitial set of sites in Maine that best represents a selectedset of elements at chosen target levels. We then adjustedthe resulting portfolio based on expert knowledge andother data layers.

Using a selection algorithm, SITES attempts tominimize portfolio “cost” while maximizing therepresentation of multiple conservation elements in anefficient set of sites, where each site is composed ofmultiple planning units. This set of objectives constitutesthe “Objective Cost Function” defined as:

Total Portfolio Cost = (cost of selected sites) +(penalty cost for not meeting the stated targetlevel for each element) + (cost of spatial dispersionof the selected sites as measured by the totalboundary length of the portfolio).

More formally:

In other words, the algorithm seeks to minimize TotalPortfolio Cost by selecting the set of planning units thatcovers as many elements as possible, as “cheaply” aspossible, in as geographically compact a set of sites aspossible.

In the sections below, we describe the various inputsused for a SITES analysis. We ran the SITES model withnumerous combinations of input levels, varying each inputto assess the outcome based on project goals. For example,we varied the target levels for different elements (e.g., focalspecies, critical habitats, etc.). We also ran the model withexisting protected areas locked in and locked out, therebyallowing us to assess how these areas influenced the model.The ultimate objective was to “find the set of sites that metstated goals for all target groups in an efficient manner,while also meeting the general criteria of reserve design(e.g., connectivity, minimal fragmentation)” (Noss et al.2001:19). The inputs used in the final model are shown inFigure VI.1 and explained below.

Planning UnitsSITES strives to meet target goals by selecting

combinations of “planning units,” with each planning unitcontaining a proportion of the total area or number ofoccurrences of one or more targets. We used a hexagonalshaped planning unit for the Maine Wildlands Networkdesign analysis. A hexagonal shape was selected over othershapes or entities (e.g. square cells, watersheds) because theunit size remains constant (planning units that vary widelyin size can present problems for the SITES algorithm),approximates a circle, which has a low edge to area ratio,and provides a relatively smooth output (as compared withsimilar sized square cells). We chose a hexagonal planningunit of 1000 ha (Figure VI.3) because this size was largerthan the resolution of the coarsest input data set, andprovided output that had sufficient resolution for thepurposes of this study. To facilitate the use of data onconserved lands in the SITES model (see ExistingConservation Lands under Data Layers), we intersectedvarious conserved lands layers with the planning unit layer.The result is that some planning units are smaller than1,000 hectares in size, and irregularly shaped. A total of12,047 planning units were used for our analysis, 57%of which were 1,000 ha, and 69% of which weregreater than 500 ha.


Site CostA “cost” for each planning unit is included in the total

cost function. Cost can be a financial value, the weighted-sum of area and suitability for conservation goals (e.g.,Stoms 1999), or some other value. We assigned each unit acost based on its size—1,000 (hectares) in many cases, withthe exception of some units along state borders, and whereconserved lands were intersected with the planning unitlayer. With a nearly uniform unit cost, the model is drivenby its input elements, and attempts to minimize total area.

Penalty CostThe SITES algorithm will try harder to meet the

assigned targets for elements with higher “penalty” valuesthan those with lower values. Thus, if one target elementwas very important, and had to be represented in the finalset of sites, one might assign a penalty cost of 1,000, asopposed to the default of 1.0. In our model, all elementswere given the same penalty cost factor of 1.0 (i.e., noelement had a higher intrinsic value of being included inthe final portfolio).

Boundary LengthSITES allows the user to control the amount of

“clumping” of sites in the output portfolio. This isaccomplished by controlling the weight given tominimizing the total length of the portfolio boundary (i.e.,the sum of the circumferences of all clumps of planningunits in the portfolio). A boundary length modifier of 0results in no influence over clumping, whereas increasingthe modifier value gives relatively greater importance toboundary costs and results in greater clumping (a highboundary length modifier would create the extreme of asingle clump of planning units in the shape of a circle).Using a very small boundary modifier of 0.05 resulted in afinal portfolio composed of fewer single, outlying planningunits, but still reflected much of the output generated byan “unconstrained” run. As discussed in Section IV,minimizing the edge to area ratio helps retain theecological integrity of protected areas by decreasing theamount of edge effect and allowing fewer intrusions.

TargetsThe targets chosen for the Maine Wildlands Network

Design included areas greater than or equal to fivekilometers from the nearest major road (Class 1-3; TableVI.1), large wetland complexes, high-ranking slopeheterogeneity, salmon-bearing subwatersheds, andpredicted habitat for lynx and wolf. These are discussed inSection IV and below.

Table VI.1. Road classes in Maine.

Class Description1 Interstate2 Primary3 Secondary4 Improved5 Unimproved6 Trail

Model OutputThe SITES algorithm uses a process termed “simulated

annealing.” Through many iterations (thousands tomillions) of the model, this technique gradually “hones in”on a set of planning units that best meet the targetconservation goals, while minimizing cost. We ran themodel 10 times, each with one million annealingiterations. SITES selected the run that best met the targetgoals, with the least cost, as the “Best” run for that set often runs. SITES also creates a summary file, called SumRuns, which indicates the number of times (out of ten) aplanning unit was included in the final set of planningunits. Planning units that had high Sum Runs values (7-10) were used to help delineate core areas. Planning unitsselected five or six times out of ten were used to help definestudy areas. Sum Runs results were also used to helpdelineate linkages for the Maine Wildlands Network.


Data LayersThe following data layers were used as inputs to the

SITES analysis:

Predicted Habitat for Focal SpeciesThe requirements of focal species are useful for

determining design parameters such as core area locationand size and linkage location and width. Focal species arealso chosen based on their value as keystones or theirp o t e n t i a l u m b r e l l a v a l u e f o r t h e c o n s e r v a t i o n o f o t h e r

species (see Section V and Appendix 3). Salmonsubwatersheds and predicted habitat maps for wolf andlynx were used as inputs to the SITES analysis (see TableVI.2, Figure VI.1). Habitat maps for the other identifiedfocal species are not included in this analysis, as accurate,spatially explicit data modeled at the appropriate scale arenot currently available.

Table VI.2 Focal species data used in SITES analysis

Layer Description SourceEastern timber wolf Predicted core and dispersal wolf habitat Harrison and Chapin 1998

Canada lynx Predicted habitat Krohn et al. 1998 (Maine GAP)

Atlantic salmon Subwatersheds of eight rivers containingendangered runs of Atlantic salmon

Derived from Maine Office of GIS data


Figure VI.1 Focal species habitat included as inputs in SITES 1.0 analysis. (Sources: see table VI.2)


Large Wetland ComplexesThe ecological values of wetlands have been well

documented. Krohn et al. (1998) suggest that “because ofthe high number of vertebrates (and probably also plantsand invertebrates, although we have no direct measures) inwetland habitats, the importance of wetland and shorelandregulations to conservation in Maine can not beoverstated.” Further, as a result of human settlementpatterns in lowlands and along river valleys, it is likely(although poorly documented) that many of Maine’swetlands have been lost and that remaining wetlands are atrisk (Krohn et al. 1998). This is especially true for southernMaine, where human population densities are much higher.Thus, we include wetland complexes (including all wetlandtypes except mudflats, sand shore, gravel shore, rock shore,open water or salt water) that were greater than 150hectares in size as a special element input to SITES (FigureVI.4).

Areas with High Variability in SlopeElevation has a significant effect on both temperature

and precipitation (Krohn et al. 1999), and thus affects thedistribution of both plants (McMahon 1990) and animals(Boone 1996). Therefore, areas where slope is highlyvariable or heterogeneous are likely to support relativelyhigher species and natural community richness (per unitarea) than more homogeneous areas (Burnett et al. 1998;Nichols et al. 1998). Slope heterogeneity was calculatedusing a “variety” neighborhood function,8 and areas withhigh slope heterogeneity were included as a special elementinput in SITES (Figure VI.4).

Existing Conservation LandsPublic conservation lands comprise only about 5.6% of

Maine, compared with 18%, 18%, and 13% in New York,New Hampshire, and Vermont respectively (McGroryKlyza 2001). If other regions are any indication, many ofthese areas were likely conserved for non-biological reasons,such as scenery, recreation, or because they lacked potentialfor resource extraction (Noss and Cooperider 1994; Noss etal. 1999). Moreover, only a small portion of Maine’s publiclands have GAP 1 or 2 protection status. In addition topublic lands, there are hundreds of thousands of hectares ofprivate lands, mostly commercial timberlands, whoseresidential development rights have been transferred to thestate or conservation groups. These lands are generallygiven GAP 3 conservation status.

Although most “conservation” land in Maine does nothave a high level of protection, many of these lands arelarge and sparsely populated, and may harbor important 8 This function assesses, for any given cell, the number of neighboringcells with different slope values than its own value. In this way, thefunction helps to identify cells located in areas of high topographiccomplexity.

habitat for focal species, even though some restoration andimproved management will likely be needed. Thus, despitetheir relatively low protection status, the potential of theseareas to “anchor” or enhance future protected areas justifytheir inclusion in the SITES analysis.

SITES allows user-identified planning units to belocked into or out of a final SITES output. Those unitslocked into the process are included in the final selectionregardless of how they affect the objective cost function.Although these units are not considered “elements” to berepresented at a specific target level, they are nonethelessused by SITES in identifying a best set of units. We lockedin planning units containing a high proportion of large orwell-protected conservation lands (Figure I.2). These landswere defined as land in federal, state, municipal, or non-profit ownership or easement that are classified as havingeither GAP level 1 or 2 protection (Table I.3, Figure I.2),or are private parcels, very large in size (>5000 ha) andmanaged with at least GAP level 3 protection. The area oflocked-in units equaled 7.7% of the state (Figure VI.3).

Developed LandsAs mentioned above, SITES allows planning units to

be locked out of the final output. Because this design isprimarily concerned with identifying a network ofwildlands, or restorable lands, we locked out planningunits with an excess of 3% of their area classified asdeveloped (urban/industrial, dense residential, sparseresidential, or highways/runways; Figure VI.3). The area oflocked-out units equaled 8.8% of the state.

Terrestrial and Aquatic Vegetation ClassesRepresentation analysis provides the “coarse filter” to

ensure the representation of the many species for which fewdata exist. As part of the Maine GAP study (Krohn et al.1998), 37 terrestrial and aquatic vegetation and land covertypes were mapped. For the purposes of this analysis, wecollapsed those 37 types into 11 broad classes, as shown inTable VI.3. The 11 types are displayed in Figure VI.5.


Figu

re V

I.3 S

ITES

Pla

nnin

g un

its s

how

ing

units

lock

edin

and

out

.Fi

gure

VI.2

Sch

emat

ic o

f SIT

ES a

naly

sis


Figure VI.4 Special elements included as inputs in SITES 1.0 analysis.


Table VI.3 Relationship of the 37 GAP-ME Land Use Land Cover Types (Krohn et al. 1998) to the collapsed the land coverclasses used in this analysis.

Land Cover Class GAP Land Use Land CoverTypeAbandoned fieldBlueberry fieldGrasslands

Agriculture/Grassland

Crops/GroundSparse residentialDense residentialUrban/Industrial

Developed Land (results notcalculated)

Highways/RunwaysClearcutEarly regenerationLight partial cut

Cut or Regenerating Forest

Heavy partial cutDeciduous Forest Deciduous forest

Deciduous/coniferous forestConiferous/deciduous forest

Mixed Forest

Late regenerationConiferous forest Coniferous forest

Deciduous forested wetlandsConiferous forested wetlandsDead forestDeciduous scrub-shrubConiferous scrub-shrub

Forested Wetland

Dead scrub-shrubFresh aquatic bedFresh emergentPeatland

Freshwater Wetland

Wet meadowSalt aquatic bedSaltwater WetlandSalt emergent

Mudflat MudflatSand shoreGravel shore

Shore

Rock shoreShallow waterWater (not included in results)Open waterAlpine tundraAlpineExposed rock/Talus


Figure VI.5 Land cover classes used in SITES analysis (collapsed classification based on data fromMaine GAP analysis).


Data Sources

Data layers used for Maine Wildlands Network DesignSITES analysis were primarily in the form of digital GIS

coverages and were obtained from a variety of sources(Table VI.4).

Table VI.4 Digital data layers used in the SITES analysis.

Physical Features Scale / Resolution SourceElevation (Digital Elevation Model) 90 m grid The Nature

ConservancySlope 90 m grid Derived from

elevationWetlands 1:100,000 ME GAPCultural Features Scale / Resolution SourceRoads 1:100,000 Maine Office of GISPublic Lands and Private Conservation Lands Multiple The Nature

Conservancy, AMCBiological Features Scale / Resolution SourceAtlantic Salmon Habitat 1:24,000 Maine Office of GISPredicted Eastern Timber Wolf Habitat 1 km grid Harrison and

Chapin 1999Predicted Canada Lynx Habitat 90 m grid ME GAPLand Use Land Cover Classes 30 m grid Derived from ME

GAP


Element Target Levels for SITES AnalysisSITES identifies the number and configuration of

planning units that most efficiently and compactlyrepresent the identified elements at target levels specifiedby the investigator. Thus, although SITES will help rankareas to meet target goals, these goals must be set ahead oftime. To a large extent, the number of elements and theirtarget levels determine the size of the Best and Sum Runsoutputs.

There have been few studies that attempt to addressthe question of how much representation is enough toprotect biodiversity in perpetuity. To complicate matters,the answer is likely region and species (or process) specific.In a recent assessment for the Greater YellowstoneEcosystem and Utah-Wyoming Rocky MountainsEcoregion, Noss et al. (2002) used as a planning goal 25%representation of each vegetation and physical habitat type(see Anderson et al. 1999 for similar guidelines). Soulé andSanjayan (1998) question whether specific conservationtargets are helpful in protected areas planning, and cautionagainst the setting of interim targets, or of final targetsthat are unrealistically (for biodiversity’s sake) low.Margules et al. (1988) conclude, “[T]he belief thatbiological diversity is ‘reasonably secure’ or ‘as well takencare of as possible’ with the dedication of one or a few wellchosen reserves in an ecological domain is unfounded. Thereality is that a very large number of reserves seems to benecessary to secure biological diversity.”

For the Maine Wildlands Network design, we chosetarget levels (Table VI.5) that best reflected our goals ofrestoration and the recovery of focal species, especially thewide-ranging carnivores, and native terrestrial andfreshwater ecosystems. Thus, predicted habitat for thosespecies that are wide-ranging, or whose habitat in Maine isrestricted or represents a major percentage of the totalremaining habitat in the Northeast, were given therelatively high target levels of 70% for subwatersheds ofsalmon-bearing rivers and 50% for predicted wolf and lynxhabitat. Particularly important natural community typessuch as large wetlands, and areas greater than 5 km fromthe nearest Class 1-3 road, were also assigned relativelyhigh target levels of 50%. Major aquatic vegetation typesreceived a lower target of 25%. Major watersheds,ecoregions, and areas with high slope heterogeneityreceived targets of 20% each. Major terrestrial landuse/land cover types received targets of 15%.

SITES Selection ResultsWe had SITES report the Sum Runs results, which

identify all planning units that were in the selected set atleast once in the ten model runs (Figure VI.6). We used aboundary modifier of 0.05, with planning units containinga high percentage of conservation lands (see above) lockedinto the final selection, and with planning units containingmore than 3% of their area as developed land locked out ofthe run.

Table VI.5. Element target values used for the SITES analysis.

Element Representation Target(%)

Salmon Bearing Subwatersheds 70 Large Wetland Complexes 50 Predicted Lynx Habitat 50 Predicted Wolf Habitat 50 Areas >5km from nearest class 1-3 road 50 Aquatic Vegetation Classes 25 Terrestrial Vegetation Classes 15 Major Watersheds 20 Maine Ecoregions 20 High Slope Heterogeneity 20


Figure VI.6 Sum Runs results from SITES Analysis.


Using the Sum Runs output, we created four alternativesbased on the planning units selected most frequently.These are:1. Planning units selected five times or more out of ten

runs (5-10);2. Planning units selected six times or more (6-10);3. Planning units selected seven times or more (7-10);

and4. Planning units selected eight times or more (8-10).

These alternatives are shown in Figures VI.7 and VI.8.Alternative one covers the largest area of the state,alternative four the smallest. Together they form thebaseline for further design of the Wildlands Network.More specifically, planning units selected seven or moretimes serve as the basis for defining proposed core areas ofthe Maine Wildlands Network, while planning unitsselected five times or more are designated as study areas.This allows for a degree of prioritization within theproposed network. The rationale for the use of thealternatives to establish core and study areas, and otherdetails of our conservation planning methodology arediscussed more fully in Sections VII and VIII.


Figu

re V

I.7 S

um R

uns

alte

rnat

ives

one

and

tw

o: p

lann

ing

units

sel

ecte

d fiv

e or

mor

e tim

es o

ut o

f 10

(5-1

0) a

nd s

ix o

r m

ore

times

out

of t

en (

6-10

).


.

Figu

re V

I.8 S

um R

uns

alte

rnat

ives

thr

ee a

nd fo

ur: p

lann

ing

units

sel

ecte

d se

ven

or m

ore

times

out

ten

(7-1

0)an

d ei

ght

or m

ore

times

out

of t

en (

8-10

).


VII. WILDLANDS NETWORK DESIGNEcological reserve design is as much an art as a science,

given the typical scenario of incomplete data and highlevels of uncertainty (Noss 1995; Dobson et al. 1999).Nonetheless, we based our wildlands network design

strongly on the results of the SITES analysis discussed inSection VI, plus several other important data layers (TableVII.1), in a building block approach (see Trombulak 1996for a review of this approach).

Table VII.1 Digital data layers used in the Maine Wildlands Network Design process.

Physical Features Scale / Resolution SourceSubwatersheds (drainage divides; GreatPond level)

1:24,000 Maine Office of GIS

Elevation (Digital Elevation Model) 90 m grid The NatureConservancy

Cultural Features Scale / Resolution SourceHuman Population Centers 1:1,000,000 ESRI Digital Chart of

the WorldRoads 1:100,000 Maine Office of GISPublic Lands and Private ConservationLands

Multiple The NatureConservancy, SweetWater Trust

Other Scale / Resolution SourceSITES Best Portfolio 1,000 ha hexagons DerivedSITES Sum Runs Layer 1,000 ha hexagons DerivedOther, non-digital, data used included information on physiographic features (e.g.,mountainous areas) from the Maine Atlas and Gazetteer (Delorme 1996).


Core Wild Area DelineationThe forests of northern and western Maine offer many

opportunities for the establishment of large, interconnectedcore areas. Indeed, forest coverage is so extensive that amajor challenge is identifying which areas should be giventhe highest priority. SITES, through its Sum Runs outputs,offers a way of approaching this prioritization issue. Thatis, planning units chosen most frequently are probably thebest candidates for core areas. We reviewed the alternativespresented in Section VI and decided to use the one inwhich planning units were selected seven or more times (7-10) as the basis for delineating proposed network coreareas.9 We chose this alternative because we felt that the 8-10 output did not satisfy enough of our conservation goalsand the 5-10 and 6-10 outputs did not provide enoughprioritization for those engaged in day-to-day conservationwork. Nevertheless, the cores identified via the 7-10output do not constitute the entire proposed network. Theareas identified in the 5-10 and 6-10 outputs should alsobe considered part of the network, though subject tofurther study to determine if they should be consideredpotential core or compatible-use areas.

Wildlife Linkage DelineationThe Maine Wildlands Network proposes three

different types of linkages to provide connectivity betweenidentified core areas and study areas: terrestrial linkages,aquatic linkages, and aquatic buffers (see Section IV for asummary of proposed linkage types).

Terrestrial LinkagesTechniques for determining the appropriate width for

functional landscape-scale linkages are relatively untested(Dobson et al. 1999). Further, few studies have successfullyevaluated whether specific linkages have providedfunctional connectivity (see Dobson et al. 1999; Beier andNoss 1998). It is widely recognized, however, that effectivelinkages must be based upon the requirements of thespecies they are designed to benefit (Dobson et al. 1999).For these reasons, we chose to delineate terrestrial linkagesin a general way only, as lines representing broad linkagezones. These lines represent the general region that appearsto present the best possibility for connectivity betweencores. Linkages were identified using the SITES Sum Runsoutputs, and the Conservation Land, Roads, HumanPopulation Centers, and Land Use Land Cover map layersdescribed in this section and elsewhere in the document.

9 All the Sum Runs alternatives (5-10, 6-10, 7-10, 8-10) includeexisting GAP 1-2 and Large GAP 3 lands because those lands werelocked into the SITES model.

Aquatic LinkagesGiven the importance of riparian and wetland habitat,

we included as proposed aquatic linkages thosesubwatersheds of large rivers, and rivers bearing native runsof Atlantic salmon (Figure VII.1), occurring in regionswith human populations densities less than or equal to5/km2. These linkages will complement terrestrial linkagesby providing habitat for species that tend to use rivervalleys and associated cover types for dispersal, andproviding core habitat for riparian and aquatic species.10

Further, protected linkages incorporating entiresubwatersheds will help protect the integrity of the streamsrunning through them.

Aquatic BuffersIn regions with human population densities greater

than 5/km2, subwatersheds of large rivers were included asproposed aquatic buffers. These buffers are intended toprovide some level of protection for species associated withaquatic ecosystems, as well as to provide essential habitatfor terrestrial species that may use riparian areas formovement or dispersal.

10 As noted in section IV, we recommend that terrestrial linkages receivea protected status similar to that of cores, in which ecological processesare allowed to function as naturally as possible.


Figure VII.1 Subwatersheds of major rivers and rivers containing runs of native Atlantic salmon.


VIII. NETWORK DESIGN SUMMARYAND EVALUATION

Network Summary and Spatial StatisticsUsing the approach and methods described in Sections IV-VII we designed a proposed wildlands network for Maine(Figure VIII.1). This design includes approximately 28,725km2 of core wild areas, both within the network itself andas isolated reserves elsewhere in the state, 1,175 km2 ofaquatic linkages, and 4,671 km2 of aquatic buffers. Coreareas were delineated using the Sum Runs 7-10 alternative.We also propose an additional 17,301 km2 that should bestudied for inclusion in the network as core areas,compatible-use lands, or linkages (Table VIII.1). Thesestudy areas were delineated using the Sum Runs 5-10alternative. The proposed core areas and aquatic linkagesshould be managed with significant (GAP Level 1 or 2)protection. See Table VIII.2 for an illustration of thegeneral relationship between wildlands networkcomponents and GAP status levels. The Maine Wildlands

Network Conservation Plan does not propose that thisentire network be managed solely for its biodiversityvalues, nor necessarily in a completely “wild” state. Forexample, aquatic buffers may need to be managed formultiple uses, as these areas generally contain relativelyhigh human and road densities. Conservation in thesebuffer areas will likely focus on private land stewardshipand local conservation efforts (e.g., watershed groups,conservation commissions). Further, some areas identifiedas potential core or linkages may need to be refined in thelarger regional wildlands network design as moreinformation becomes available. We do not expect, however,that additional information will suggest reducing the sizeof cores or linkages. More likely, additional informationwill show the need to make cores and linkages even larger.

Table VIII.1 Area and percentage of state represented by proposed core wild areas, aquatic linkages, aquatic buffers,and study areas in the Maine Wildlands Network together with existing conservation lands.

Network Element Area (km2) Area (acres) Proposed Network Core Wild Areas and Isolated Reserves 28,725 7,098,163

Existing GAP 1-2 Areas 1,880 464,618 Existing Large GAP 3 Areas 4,514 1,115,545

Proposed Aquatic Linkages 1,175 290,378

Proposed Aquatic Buffers 4,671 1,154,298

Proposed Study Areas 17,301 4,275,145


Table VIII.2 General relationship of Wildlands Network Components to GAP status levels.1

GAP Status Level Wildlands Network Component Description

Status 1(“Wild Lands”)

Core Wild Areas (and where possible,Wildlife Linkages, especially terrestrialand aquatic linkages)

A parcel totally and permanentlyprotected from conversion of naturalland cover and with a managementplan in operation to maintain land in anatural state. Natural processes areal lowed to proceed withoutinterference or are mimicked throughmanagement practices.

Status 2(“Low Use”)

Core Wild Areas, Wildlife Linkages(terrestrial linkage, aquatic linkage),sometimes Compatible-use Lands

A parcel totally protected fromconversion of natural land cover andwith a management plan in operationto maintain a primarily natural state,but where uses or suppression ofnatural processes may degrade thequal i ty o f ex i s t ing natura lcommunities.

Status 3(“Medium Use”)

Compatible-use Lands, WildlifeLinkages (aquatic buffer)

A parcel protected from conversion ofnatural cover for more than 50% ofarea, but subject to extractive usessuch as timber harvest or mining.

Status 4(“High Use”)

Outside of Network

A parcel with more than 50% of areaplanned or in use for agriculture or as“open space” for active recreationpurposes (e.g., ballfields, golf courses).Natural processes are altered orreplaced by human use andmanagement of land.

1These categories are similar to those used in the GAP Mapping and Categorizing Land StewardshipHandbook, by Crist et al. 2000, available at www.gap.uidaho.edu/handbook/Stewardship/default.htm.


Figure VIII.1 Proposed Maine Wildlands Network .


Evaluation of Conservation Goals Achieved by the MaineWildlands Network

As discussed in Section VI (Table VI.5), the MaineWildlands Network seeks to achieve a number of focalspecies, representation, and special element conservationgoals. We evaluated how well the network designpresented in Figure VIII.1 achieves these conservationgoals. We then compared those results to the 5-10, 6-10,and 8-10 sum runs alternatives presented in Section VI, aswell as existing GAP 1-2 lands and large GAP 3 lands.More specifically, we quantified the percentage of eachconservation element captured within the proposed corearea of the network, as derived from the 7-10 sum runsalternative, and compared that with existing conservationland (GAP 1-2 and GAP 1-2 and large GAP 3), with aminimal set of core areas (Alternative 8-10) and with the

proposed network plus the study areas (Alternatives 5-10and 6-10).

The actual level of representation necessary to ensurethe protection and persistence of any given element (forbiotic elements) depends on many variables (e.g., theoverall area represented by the element, the number ofoccurrences of the element in the state, connectivitybetween the elements, the natural history characteristics ofbiotic elements). Noss and Cooperrider (1994) observe that,in general, “science cannot tell us precisely how manytimes or in what sized reserves each species or ecosystemtype must be represented to be viable.” We propose,therefore, that representation values be used to identifyelements that may be relatively under-represented withinthe proposed network, and not to speculate on exactly whatlevel would be “enough” representation within thenetwork.


Figu

re V

III.2

Tw

o el

emen

ts a

sses

sed

for

repr

esen

tatio

n w

ithin

the

pro

pose

d M

aine

Wild

land

s N

etw

ork:

maj

or w

ater

shed

s an

d el

evat

ion

clas

ses.


Figure VIII.3 Biophysical regions (Maine ecoregions) assessed for representation within the proposedMaine Wildlands Network.


Focal Species and Special ElementsThree focal species (wolf, lynx, and salmon) and three

special elements (large wetlands, roadless areas, and slopeheterogeneity) were incorporated in the SITES model.Figure VIII.4 displays the percentage of focal species andspecial elements conservation goals included underdifferent protection alternatives. We did not calculate slopeheterogeneity results.

As Figure VIII.4 demonstrates, existing GAP 1-2lands provide little protection for any of these focal species

or special elements. When GAP 3 lands are added in, theresults improve considerably, except for salmonsubwatersheds. For example, only about 13,600 hectares oflarge wetlands are included in GAP 1-2 lands (TableVIII.3), just 3.4% of all large wetlands. Adding in GAP 3lands brings the total to about 88,000 hectares or 21.7% ofall large wetlands. Under the proposed WildlandsNetwork, about 150,000 hectares of large wetlands wouldbe protected within core areas, about 36% of the total.Only when study areas are included does the total riseabove the 50% target, to 58.6% of all large wetlands.

Figure VIII.4. Percentage of focal species and special element conservation goals included in differentprotection alternatives.


Table VIII.3. Focal species and special element conservation goals included in different protectionalternatives, in hectares.

ElementExisting GAP

1-2(Ha.)

Existing GAP1-2 + Large

GAP 3(Ha.)

Sum Runs5-10(Ha.)

Sum Runs6-10(Ha.)

Sum Runs7-10(Ha.)

Sum Runs8-10(Ha.)

Large Wetlands 13,649 88,244 238,239 188,773 148,580 115,576 Roadless Areas 464,261 1,576,651 4,601,665 3,684,640 2,872,017 2,192,682 Salmon Subwatersheds 1,589 1,589 130,119 106,769 87,549 61,486 Wolf Habitat 124,401 499,507 2,997,796 2,517,651 2,025,854 1,604,491 Lynx Habitat 31,543 377,131 1,431,983 1,270,914 1,083,229 916,884

In the case of lynx and wolf habitat, existing GAP 1, 2,and 3 lands do not meet the conservation targets. Underthe proposed Wildlands Network, however, both targetswould be satisfied, with 51.7% and 67.4% of wolf andlynx habitat represented, respectively.

Under no protection alternative do we meet the salmonsubwatershed target of 70%. Current GAP 1, 2, and 3lands provide almost no protection—less than one percentof the subwatersheds, although this situation may bechanging with recent conservation efforts in the MachiasRiver watershed. Even under the most expansive protectionalternative, with all study areas included (Sum Runs 5-10),only about 65% of salmon subwatersheds, about 130,000hectares, receive some degree of protection.

Vegetation and Land CoverWe evaluated vegetation and land-cover representation

using data from the Maine GAP Analysis Project (Krohn etal. 1998). The 37 land use land cover types mapped by theMaine GAP project were collapsed into 11 major types, asdiscussed in Section VI (see Table VI.3 and Figure VI.5).Representation of each vegetation or land cover type underdifferent protection alternatives is summarized in FiguresVIII.5. The goal of the Maine Wildlands Network Designis to protect at least 15% of each terrestrial vegetation typeand 25% of each aquatic vegetation type. Open water anddeveloped land were not included in these types.


Figure VIII.5 Percentage of the land use land cover types included under different protection alternatives.

Within the proposed Wildlands Network, core wildareas (Sum Runs 7-10) would protect more than 30% of allmajor forest types. For example, nearly 1.2 million hectaresof mixed forest would be protected (Table VIII.4). Theserelatively high values are to be expected, as many of thelarger proposed cores occur in areas where land is primarilyowned by timber companies and has been cut for decades.Predominantly high elevation vegetation land cover types(alpine) were well represented at nearly 80%. Even currentGAP 1-2 lands provide good protection for alpine areas.However, this is the only major land cover type that is wellrepresented under current protection regimes. The addition

of large GAP 3 lands increases representation somewhat,but only to about 10% for the major land cover types.Proposed wildlands network core areas would protect about26% of shore types, but only 10% or less ofagriculture/grassland, saltwater wetland, and mudflatwetland types. Finally, the network design processgenerally avoided human dominated land cover types bylocking out planning units with an excess of 3% of theirarea classified as developed (urban/industrial, denseresidential, sparse residential, or highways/runways; FigureVI.2).


Table VIII.4. Land use land cover types included under different protection alternatives, in hectares.

Land-use/Land-cover Type Existing GAP1-2 (Ha.)

Existing GAP 1-2+ Large GAP 3

(Ha.)

SumRuns 5-10 (Ha.)

Sum Runs6-10 (Ha.)

Sum Runs7-10 (Ha.)

Sum Runs8-10 (Ha.)

Mudflat Wetland 133 133 2,646 2,089 1,871 1,232 Saltwater Wetland 693 693 3,526 3,002 2,536 1,918 Agriculture/Grassland 2,325 2,659 148,973 96,520 62,074 37,797 Shore 1,048 1,167 4,264 3,910 3,496 2,738 Mixed Forest

68,827 258,5811,897,26

8 1,516,756 1,175,754 892,922 Freshwater Wetland 5,620 12,689 75,388 59,607 45,644 34,644 Deciduous Forest 28,224 89,425 768,684 607,248 462,239 344,897 Forested Wetland 12,677 47,558 313,515 250,095 194,677 148,961 Coniferous Forest 35,791 103,579 481,756 403,755 327,047 267,306 Cut and RegeneratingForest 11,323 92,569 626,884 518,137 418,944 329,460 Alpine 5,155 5,164 5,409 5,324 5,186 5,159

Biophysical RegionsBiophysical regions (or ecoregions) are relatively

homogeneous areas identified using climate and landformvariables (Bailey 1995; Bailey et al. 1994). Bailey (1995)delineated ecoregions for the entire U.S., with Mainefalling within portions of three different ecoregions. Forour analysis, we use a modified classification of theseecoregions (McMahon 1990) that contains 15 distinctbiophysical regions based on climate and topography.Because these biophysical regions are related to bioticfactors, it is important that each region be represented in aconservation network. Although two revisions of theseoriginal biophysical regions have since been conducted(Keys et al. 1995; Krohn et al. 1999), these layers were notavailable from the Maine Office of GIS.

The representation of biophysical regions within coreareas of the Maine Wildlands Network Design (FigureVIII.6, Sum Runs 7-10), is summarized below:

• Core areas of the Maine Wildlands Network Designrepresent the Western Mountains, Eastern Lowlands,St. John Uplands, Central Mountains, Aroostook Hills,and Boundary Plateau biophysical regions at greaterthan 45%, well exceeding the 20% target. Theseregions are characterized by low human populationdensity and large private and industrial timberlands.

• The East Coastal Region is well represented, at about25%, exceeding the target level. This translates into a

little less than 74,000 hectares of core area (TableVIII.5). This region is sparsely populated in places andencompasses the major salmon subwatersheds.However, it is separated from the rest of the networkin northern and western Maine by major transportationcorridors. It will be important to consider connectivityto New Brunswick when conducting the broaderregional network design.

• The Midcoast, Aroostook Lowlands, Western Foothills,Southwest Interior, and Eastern Interior biophysicalregions were captured less well by the core areas ofproposed network, at between 10% and 15%. Theseareas are on the fringe of more developed regions, and,in the case of the Western Foothills, host highnumbers of tourists, cottagers, hunters and fishers.They are thus intermediately remote, containinghigher road densities than the most remote areas.

• The remaining three biophysical regions, SouthCoastal, Penobscot Bay, and Central Interior are poorlyrepresented in the core areas of the network at 8% orless. These areas generally contain high population androad densities, resulting in relatively lower habitatsecurity scores in our model. Further, the lack of largewetlands, few large public or private conservationlands, and a small total area of watersheds containingAtlantic salmon-bearing rivers in these regions resultedin relatively low scores in the SITES analysis.


Figure VIII.6 Percentage of the ecoregions of Maine included under different protection alternatives (e.g.,core areas of the proposed wildlands network include 11% of the Western Foothills bioregion.)


Table VIII.5. Ecoregions of Maine included under different protection alternatives, in hectares.

Maine Ecoregions Existing GAP1-2 (Ha.)

Existing GAP 1-2 + Large GAP

3 (Ha.)

Sum Runs5-10 (Ha.)

Sum Runs6-10 (Ha.)

Sum Runs7-10 (Ha.)

Sum Runs8-10 (Ha.)

South Coastal Region 794 794 23,738 16,835 13,513 10,187 Penobscot Bay Region 3,202 3,202 41,918 30,048 14,984 7,748 Central Interior 4,978 4,978 178,279 93,035 57,765 36,294 Midcoast Region 872 872 39,989 26,734 21,537 12,738 Aroostook Lowlands 250 250 112,859 84,053 48,937 27,943 Western Foothills 2,108 2,108 195,827 116,152 67,372 33,423 East Coastal Region 22,137 22,137 105,799 93,755 73,998 51,193 Southwest Interior 1,641 1,641 166,158 116,155 56,696 31,648 Eastern Interior 4,162 14,066 353,903 204,263 126,027 72,926 Western Mountains 18,448 64,132 759,262 627,962 490,612 358,806 Saint John Uplands 5,177 51,093 487,933 396,211 300,846 213,326 Eastern Lowlands 9,927 9,927 253,160 213,632 173,000 126,874 Aroostook Hills 20,445 184,879 907,458 787,662 674,912 574,826 Central Mountains 84,063 84,063 588,250 514,495 423,300 333,835 Boundary Plateau 9,786 194,021 387,109 364,115 328,922 301,260

Major WatershedsThe ecological integrity of watersheds ultimately

affects the integrity of streams flowing through them.Representing significant portions of Maine’s majorwatersheds in a wildlands network will help assure that thesurface waters of these watersheds maintain their integritywhile providing habitat and associated resources for thespecies that inhabit them.

Representation of major watersheds in the proposednetwork displays a pattern similar to that of biophysicalregions. Eight of the northern and western most watershedsare represented at greater than 40% in the proposed coreareas of the network (Figure VIII.7, Sum Runs 7-10).Because the watersheds vary so greatly in size, thesepercentages translate into very different areas, ranging from

over 400,000 hectares of the St. John River to 224,000hectares of the E. Branch of the Penobscot to about120,000 hectares of the Dead River (Table VIII.6).

In addition, four of the noncoastal watersheds in theeast (Cathence Stream, Dennys, East Machias, and PleasantRiver) were captured at greater than 40% as well. Of the11 watersheds that are represented at less than 20%, theconservation target, eight have significant portions of theirarea in the southern part of the state or along the coast.

Under existing GAP 1-2 protections, less than 20% ofany watershed is protected. Under the GAP 1-2 and LargeGAP 3 regime, the St. John River watershed is about 40%conserved, and three others, East Branch of the Penobscot,the Upper Androscoggin, and the Allagash River havemore than 20% of their area under some form ofconservation.

.


Figure VIII.7 Percentage of major watersheds included under various protection alternatives (e.g., core areas of proposedWildlands Network encompass 85.5% of the Upper Androscoggin watershed).


Table VIII.6. Major watersheds included under various protection alternatives, in hectares.

WatershedsExistingGAP 1-2

(Ha.)

Existing GAP 1-2 + Large GAP

3 (Ha.)

Sum Runs5-10 (Ha.)

Sum Runs6-10 (Ha.)

Sum Runs7-10 (Ha.)

Sum Runs8-10 (Ha.)

Presumpscot River 1,248 1,248 41,469 26,856 16,304 12,544 Piscataqua River 3 3 10,590 7,792 4,827 1,637 Meduxnekeag River 0 0 28,091 22,524 12,476 3,930 York-Mousam Rivers 158 158 18,710 8,073 4,929 1,778 Penobscot River 5,928 10,350 147,143 92,703 66,128 45,919 Mattawamkeg River 1,702 1,702 108,041 55,564 23,606 7,149 Kennebec, Lower 6,054 6,054 257,796 173,870 128,467 84,727 Coastal 9,095 9,095 153,251 108,758 75,102 45,014 Androscoggin, Lower 6,009 6,432 156,679 93,369 61,611 42,868 Chandler River 0 0 7,391 7,391 5,775 3,158 St. Croix River 3,207 3,207 126,182 74,018 44,489 24,495 Machias River 127 127 61,999 53,946 41,796 27,931 Union River 4 5,485 72,515 49,974 29,815 18,210 Tunk Stream 177 177 7,061 7,061 4,676 2,452 Sheepscot River 79 79 35,160 23,422 11,626 4,816 Piscataquis River 7,944 7,944 233,707 181,057 123,278 96,510 Saco River 4,057 4,057 142,064 109,288 63,940 42,941 Penebscot, W. Branch 34,320 52,762 369,451 305,184 239,649 179,590 East Machias River 113 113 57,880 45,944 38,531 28,798 Aroostook River 3,829 93,599 419,223 356,508 289,155 241,646 Pleasant River 2,283 2,283 23,018 21,643 20,321 18,859 Upper Kennebec 1,399 1,399 311,609 232,894 142,443 74,173 Fish River 463 5,587 239,194 184,603 138,456 85,388 Narraguagus River 68 68 54,851 45,029 38,986 28,742 Dead River 1,421 1,432 197,877 163,115 120,204 66,363 Ducktrap River 0 0 4,746 3,398 1,777 777 St. John River 9,803 223,506 488,940 446,329 403,515 364,813 Dennys River 1,365 1,365 23,862 21,096 18,057 9,229 Penebscot, East Branch 50,685 67,447 267,245 253,688 224,794 196,909

Androscoggin, Upper 4,026 49,276 195,986 191,711 181,620 166,888 Allagash River 12,172 62,958 292,401 275,503 257,332 229,856 Cathance Stream 12 12 9,101 9,101 7,124 6,079

ElevationsAlthough we did not establish conservation targets for

elevation per se, elevation nevertheless plays an importantrole in determining the type of natural communityoccurring at a given site. Thus, it is important to representall elevations in a wildlands network designed to protect allnatural community types. Additionally, networks that

contain protected areas and linkages that span elevationalgradients may better protect biodiversity in the face ofupcoming climate change (Noss 1993, 2001; Bennett1999).

We partitioned elevation into five classes (1-200meters, 201-500 meters, 501-700 meters, 701-1000meters, and greater than 1000 meters), following theclassification used by the Maine GAP Analysis Project(Krohn et al. 1998). The proposed network represented the


four classes containing the highest elevations very well,with values of over 53% (Figure VIII.8, Sum Runs 7-10).The 1-200 meter class was represented at the lower level of18.5% (682,000 hectares, Table VIII.8), primarily becausemost of Maine’s settled areas containing relatively highhuman and road densities occur in this elevation zone.

Current conservation regimes (GAP 1-2 and Lg. GAP 3)cover about 39% of elevations above 1000 meters,dropping to about 20% for the 701-1000 meter category,11% for 501-700 meters, 14.2% for 201-500 meters, andonly 1.5% (65,000 hectares) for the 1-200 meter class.

Figure VIII.8 Representation of elevation categories under various protection alternatives (e.g., core areas of the proposedwildlands network include 15.6% of all lands having elevations of between 1 and 200 meters).

0.0

20.0

40.0

60.0

80.0

100.0

120.0

1-200 201-500 501-700 701-1000 >1000

Elevation Class (meters)

Sum Runs 5-10 Sum Runs 6-10Sum Runs 7-10 Sum Runs 8-10Existing GAP 1-2 and Existing Lg. GAP 3 Existing GAP 1-2

Table VIII.7. Representation of elevation categories under various protection alternatives, in hectares.

Elevation(meters)

ExistingGAP 1-2 (Ha.)

Existing GAP 1-2 +Large GAP 3 (Ha.)

Sum Runs5-10 (Ha.)

Sum Runs6-10 (Ha.)

Sum Runs7-10 (Ha.)

Sum Runs8-10 (Ha.)

1-200 51,988 65,182 1,439,190 1,002,075 681,801 438,875 201-500 94,803 492,865 2,678,679 2,267,207 1,855,853 1,486,742 501-700 22,117 46,881 352,235 300,087 237,979 184,460 701-1000 14,628 27,763 114,229 100,734 83,870 71,500 >1000 3,949 4,616 9,879 9,234 8,422 8,168


ConclusionThis evaluation of focal species, special elements, land

use and land cover, biophysical regions, major watersheds,and elevations demonstrates that existing GAP 1-2 landsare inadequate to protect focal species, ecoregions,watersheds, and other important conservation features inMaine. Even the addition of existing GAP 3 lands is notenough to protect either biodiversity—as expressed in thecoarse filters of land use land cover types, state ecoregions,and large wetlands—or the wide-ranging focal species thatrequire broad expanses of connected wildlands. On theother hand, the proposed core wild areas of the MaineWildlands Network, when combined with study areas,meet nearly all the conservation goals set out in Section VI,from over 50% of wolf and lynx habitat, over 50% of largewetlands and roadless areas, to over 50% of every foresttype. The proposed core and study areas do not meet thetargets for salmon subwatersheds and for several

land use land cover types, particularly mudflat andsaltwater wetlands and agriculture/grassland types. Anumber of Maine ecoregions and watersheds are also notwell represented in the network. Those elements not wellrepresented are primarily in the southeastern and coastalregions of the state. These areas are also some of the mostheavily populated and densely roaded areas of the state.Our wildlands network design methodology, in seeking toidentify, and link, large core areas, gives lower priority toprecisely these populated areas. Nevertheless, there aresignificant conservation needs in these areas and we look toother organizations to identify and protect them. As theWildlands Project expands its planning and design effortsto a regional scale, we will revisit the representation of focalspecies habitat, large wetland complexes, land use landcover types, and so on, with an eye to balancing the mix ofthese features across the much broader landscape.


IX. CONSERVATION ACTIONOverview

The preceding sections of this document present theobjectives and rationale for wildlands restoration andrecovery in Maine, together with the science of wildlandsnetwork design. The next step is to discuss how thisnetwork of conserved lands can become reality. Therecommendations presented here are intended to guide andinspire those working on conservation issues in Maine.

The wildlands network presented in this documentshould be considered a first step in what must be a long-term and iterative process. To make the network afunctioning system, detailed conservation planning willneed to occur at all levels, from the individual parcel ofland to the landscape and beyond. Planning and action willbe incremental, proceeding over decades, and will be drivenby grassroots participation and a collective commitment toensuring a future for wild nature.

It is important to stress that this is a time oftremendous conservation opportunity in Maine, in goodpart because large amounts of undeveloped andunpopulated commercial timberland are constantlychanging hands and are available for conservation ifsufficient resources can be brought to bear. But there arealso significant threats, some new and some old, includingextensive pesticide and herbicide use, unsustainable rates oftimber harvest, land conversion, and road construction.Nevertheless, we have the planning tools at our disposal, asdescribed in this document and elsewhere, that can help usestablish landscape-scale conservation strategies andpriorities based on many conservation goals. Though wemust think in the long-term, we must accomplish much inthe short-term.

The realization of an ambitious proposal such as theMaine Wildlands Network seems formidable. The gapbetween the situation we face today and the wildlandsnetwork we hope for tomorrow is vast indeed, especially inMaine, where most lands are privately owned andwildlands are sorely underrepresented. Or is it? Perhaps ashift in perspective—a new way of looking at a very oldlandscape—can help us bridge this gap sooner than later.This plan was created with just such an ideal in mind.

The task of creating a wildlands network can best beapproached by breaking its elements down intomanageable pieces. To help visualize this process, themetaphor of a jigsaw puzzle is useful, as described in theSky Islands Wildlands Network (Foreman et al. 2000):

The completed reserve design … is the picture onthe cover of the jigsaw puzzle box. Inside the boxare all the different puzzle pieces that, when fittedtogether, will make the complete picture ([thewildlands] network). We will not put the wholepuzzle together in one fell swoop asconservationists did for, say, the Alaska Lands Act.Rather, different cooperating groups will placeseparate pieces down on the table from time totime.

How do we place the pieces on the table? Here auseful metaphor is the toolbox. Conservationistshave a toolbox containing many tools (lobbying,litigation, organizing, public relations, workingwith government agencies, fundraising, writingnew legislation, writing management plans,monitoring, applying science, doing scientificresearch, working with private landowners, landpurchase, ecological restoration, etc.). Differentconservationists have expertise in using differenttools. Certain tools are appropriate to reach certaingoals, others for other goals. There may be somenew tools in the box with which no one is yetexpert.

An important role of the Wildlands Project is toproduce the big “picture on the cover on the jigsaw puzzlebox.” The individual pieces of the Maine WildlandsNetwork puzzle will be produced by many otherindividuals and organizations (e.g., state agencies, landtrusts, non-profit conservation groups, landowners)working within their specific regions and within their areasof expertise (e.g., clean water, wilderness, endangeredspecies, stewardship, land acquisition, organic anddiversified farming). Further, timelines for differentprojects will inherently vary. Ultimately, the MaineWildlands Network can be thought of as an assemblage ofmany smaller projects and initiatives, workingsynergistically to achieve the big picture. Many existingconservation efforts in Maine (see Appendix 2) will nodoubt contribute to the long-term establishment of awildlands network across and beyond the state.

A critical first step toward this vision is thedissemination of this report, and vigorous dialog on Maineconservation priorities among the conservation community,local residents, and policy-makers.


Conservation Action StrategySuccessful establishment of a Maine Wildlands

Network will require the work of “networks of peopleprotecting networks of land (Soulé 1995).” Therecommendations below should be seen as a set of priorityactions that the conservation community could implementover the next three to five years to advance the goals andobjectives described in Section III. This plan is organizedinto eight broad components that build on one another andthat should be carried out in a coordinated, multi-prongedfashion. The components are arranged more or less in orderof operational importance. That is, from the point of viewof carrying out effective conservation action, the highestpriority is to establish a well-organized statewide networkof individuals and organizations that share a commonvision inspired by the Maine Wildlands Network, perhapsmodeled after the Yellowstone to Yukon (Y2Y) network inthe northern Rockies of the US and Canada. Next, it willbe important to establish an outreach and educationprogram focused on decision-makers, government agencies,land trusts, and local citizens. The third key parallel trackis to advocate for the creation and protection of large coreareas, guided by the Maine Wildlands Network Design.

This last element is arguably the most important partof the action plan, because Maine has so little protectedwilderness, yet restoration and species reintroductions willrequire large core wild areas to succeed in the long run.

However, without an established network or organizations,and knowledge of and outreach to decision-makers, it willbe difficult to advance the full restoration agenda. In themedium and long term, once a suitable social, economic,and political climate has been established (Simberloff et al.1999), then the real work of restoration can begin,including restoration of freshwater ecosystems, focalspecies, and natural ecological and evolutionary processes.There is nevertheless a good deal of work on the restorationfront that should occur in the short to medium term,including natural resource inventories to improve ourknowledge of the population and distribution of focalspecies and the establishment of an outreach strategy tocommunicate focal species recommendations to wildlifeagencies and others. Each component is described ingreater detail below and includes a set of recommendedactivities. Given that this action plan is designed toprovide strategic direction that will be valid for severalyears, the components and their activities have purposelybeen left somewhat general. The organizational landscapein Maine, and elsewhere in the northeastern US andsoutheastern Canada, is in such flux—and contains so manyactors—that anything more specific would probably beoutdated as soon as this plan is printed.

Table IX.1 Conservation Action Components and Operational Priority

OperationalPriority

Conservation Action Component

Establish mechanisms to coordinate the implementation and monitoring ofa Wildlands Network in MaineEstablish outreach and education programs focused on decision-makers,government agencies, and local citizens

Shor

t-te

rmPr

iorit

y

Advocate for the creation and protection of large core wild areas inaccordance with the Maine Wildlands Network DesignSupport the conservation of large blocks of forest subject to biodiversitycompatible forestry and linked to ecologically sound and economicallyviable economic developmentPursue policy and research initiatives on public conservation funding,“mega-easement” projects, alternative incentives for conservation, and theeffects of air pollutionPromote the restoration and protection of focal species in the MaineWildlands NetworkAdvocate for the restoration and maintenance of natural ecological andevolutionary processes, including wildfire, insect outbreaks, predator/preydynamics, natural succession, and flood regimes in all components (cores,linkages, study areas) of the Maine Wildlands Network

Med

ium

and

long

ter

m p

riorit

ies

Promote the restoration and protection of freshwater ecosystems, includingsalmon-bearing streams and rivers


There is not a one-to-one correspondence betweengoals and action plan components. Somecomponents—network coordination, environmentaleducation, and policy/research initiatives—apply to all thegoals and so should stand on their own from an operationalpoint of view. One goal, eliminating exotics, should takeplace in both terrestrial and aquatic ecosystems, but theapproaches are different, so the means to carry out the goalare integrated into the components that pertain to thoseparticular ecosystems.

1. Establish mechanisms to coordinate theestablishment and monitoring of a WildlandsNetwork in Maine, update the network designas needed, and ensure collaboration withconservation efforts in other parts of thenortheastern US and southeastern Canada.

For the Maine Wildlands Network to serve as aneffective conservation-planning strategy, it is essential tocreate functional mechanisms for coordination,collaboration and catalytic action. Such mechanisms mighttake the form of a formalized coordination unit, or perhapsa network, established by one or more collaboratingorganizations. In addition, processes for stakeholderinvolvement and buy-in should be included within thiscomponent. Finally, a core working group of non-governmental organizations and governmental agenciesshould be identified. Ideally this core working groupwould have a small-grants or action fund to be managedwithin the state and aimed at supporting grassrootsinitiatives, community outreach, time-sensitive researchand other small-scale projects that help accomplish MaineWildlands goals.

This coordination mechanism could be modeled on thesuccessful Yellowstone to Yukon (Y2Y) ConservationInitiative. This initiative is a “joint Canadian-U.S. networkof over 270 organizations, institutions, foundations, andconservation-minded individuals who have recognized thevalue of working together to restore and maintain theunique natural heritage of the Yellowstone to Yukonregion and the quality of life it offers.” (Seewww.rockies.ca/y2y/ for more information).

Key Activities:• Establish a statewide coalition to advance the

establishment of the Maine Wildlands Networkand monitor its progress. This coalition should be asbroad-based as possible and should seek to monitor andrespond to changes in land ownership, forestrypractices, economic trends, infrastructure development,suburban growth, and funding opportunities. Thecoalition can also serve as a forum for the discussion ofvarious issues related to the Maine Wildlands

Network, including conservation easements andstatewide reserve planning. It is important that theorganizations that make up this coalition are aware ofthe range of ongoing initiatives and programs thatrelate to their own activities, as well as where and howtheir actions, and those of other organizations, cansupport the implementation of the Maine WildlandsNetwork. To advance implementation, eachcooperating organization should have considerablefamiliarity with conservation science, including:- The goals of a wildlands network

(protecting/restoring native biodiversity andhabitat, establishing and protecting habitatconnectivity, restoring ecological processes, etc.);

- Principles of conservation biology;- Relevant vocabulary (core areas, compatible use

areas, connectivity, rewilding, etc.).

• Establish a clearinghouse to collect and distributeinformation on a timely basis regarding the MaineWildlands Network. This clearinghouse couldoperate under the auspices of the network coalition andshould have the capacity to integrate statewide andregional data in a GIS and distribute that informationrapidly to coalition members.

• Promote statewide regional conservation planningbased on the principles of conservation biology.There are a number of conservation planning effortsunderway in Maine that complement the MaineWildlands Network Vision. Furthermore, a largeamount of land, including potential conservation land,is changing hands every year. The state of Maine andother entities are purchasing or establishing easementson some of this land. A formal, statewide project thatinvolves government, the private sector, and non-profitorganizations should be established to coordinateefforts and improve the conservation impact of landpurchases, acquisitions, and transfers.

• Work with national and local land trusts toconduct mapping of significant wildlife habitatand natural communities, prepare local wildlifehabitat and linkage maps, and form multi-organization coalitions to acquire and protect keyblocks of land. Use the Maine Wildlands Network asa means of deciding which areas should have thehighest conservation priority. Also, given the recentemphasis on “working forest” deals in the conservationcommunity, this process is also an opportunity tohighlight the need for greater acquisition of coreconservation areas, whether in fee or through “foreverwild” conservation easements, especially on the part ofthe large regional and national conservation groups.


• Similarly, work with town Conservation Commi-ssions and Planning Commissions to conduct localmapping and form multi-town coalitions toencourage planning at the landscape level.

• Establish a small grants fund to be administeredby the coalition promoting the Maine WildlandsNetwork Vision. This “action fund” would beavailable to support grassroots initiatives, communityoutreach, time-sensitive research and other similarsmall-scale projects.

2. Establish outreach and education programsfocused on decision-makers, governmentagencies, and local people.

The o utr e a ch a nd c oo r dina t io n st r a te gy di s cus s ed i nc o mpo nent 1 wi ll r ea c h o nl y a li m i te d se t of c ol l a bo r a ti ngo r ga ni za t i ons . Br o ade r publi c suppor t wi l l be ne e ded f or e f fe c t iv e im pl em e nta t i on o f t he pr opo s ed c ons e rv a t io n pl a n o v e rt he l o ng t er m . The re f o re , the publ ic m us t gr a s p t he c o nc e pts a nd r e as o ning be hi nd t he goa l s a nd o bj ec t i ve s of t he pla n. The ys houl d a l s o unde r s ta nd t ha t a func ti o ning wi l dla nds ne two r kdo es not c ons t it ut e a “l a nd gr ab” tha t i s go i ng t o di s pl a c e t hem f r om t he i r l a nd a nd t r adi t io na l wa ys o f l i fe .

To build public support for the Maine WildlandsNetwork, an outreach campaign should be designed basedon a set of carefully articulated objectives. One keyobjective of this campaign should be to improve publicunderstanding of the concepts behind wildlands networkdesign (the regulatory role of top carnivores in anecosystem, connectivity, the importance of large core areas,etc.), and how those concepts are applied to Maine.

It is also important to develop formal environmentalcurricula incorporating the concepts of conservation biolo-gy, to complement public outreach. The writing ofenvironmental education curricula is a highly involved anddelicate process, and the product must be objective enoughto allow the students or audience to make their owndecisions regarding the merits of any given idea. A numberof conservation organizations have designed educationalcurricula that are relevant to some of the topics mentionedabove. The Wildlands Project is also producing materialsthat put the science of conservation biology into user-friendly terms.

Key Activities:• Establish clear goals for a communications and

education strategy to advance the MaineWildlands Network. This could be carried out underthe auspices of the coalition spearheading theestablishment of the Maine Wildlands Network.

• Con duc t pol l i ng re s e a r c h to de t e r m i ne the di f f e r e nt pot e nt i a l audi e nc e s for a com m uni c a t i on s st r a t e gy . A

gr o up wi t h e x pe r i e nc e i n s u c h r e s e a r c h, s uc h a s t he Bi o di v e r s i t y Pr o j e c t , s ho ul d c a r r y o ut t hi s wo r k.

• Create and deliver messages, using traditional andnon-traditional means.

• Regularly poll or otherwise monitor attitudes andperceptions regarding the messages. Such moni-toring will help fine-tune the messages and the meansof delivering them.

• Analyze the feasibility of developing andinstituting environmental education curricula inpublic schools and other venues. These curriculashould focus on the science of conservation biology.Ideally, a local group with experience in environmentaleducation can be recruited to assist in this process.

3. Advocate for the creation and protection oflarge core areas in accordance with the MaineWildlands Network Design.

The 1996 Biological Diversity in Maine report concludes,“the opportunity to avoid an acute biodiversity crisis inMaine is before us. If we do not initiate biodiversity-maintenance strategies now, we will be faced with a loss inbiological diversity that will be more difficult to address inthe future” (Gawler et al. 1996).

More recently, the Maine Forest Biodiversity Project,in its Ecological Reserves System Inventory (McMahon 1998),assessed the potential for conservation areas on existingpublic lands, and established 1,893 acres as the median sizefor lands that met their design criteria—a very low figure.Nevertheless, only 25% of suggested potential conservationareas in Maine met the scientific advisory committee'sminimum acreage. Further, only 23% (16 of the 69potential conservation areas) would be "selfcontained"—protect the entire ecosystem within theboundaries—and only 46% of ecosystem types arerepresented by at least one conservation area (McMahon1998).

Currently, between slightly more than two percent ofMaine’s forestland is strictly protected (McGrory Klyza2001; Maine Forest Service 2001). Most of this reservedforest is in Baxter State Park, though other, smaller,conservation areas have been established on some stateforests and on lands owned by The Nature Conservancy.

It is clear that Maine will require a much largeramount of land in conservation areas to meet even theminimum standards of protecting representative ecosystemtypes, let alone protecting connected ecosystems on a largescale. The Wildlands Network presented in this documentprovides a way of prioritizing those areas that should beincluded in an interconnected system of conservation areas.


Key Activities:• Encourage full funding for government programs

such as the Land and Water Conservation Fund,Forest Legacy program, Conservation andReinvestment Act, and National WetlandsProgram, which support the purchase of newconservation lands. Maine alone is seeking roughly $30million from the Forest Legacy program for fiscal year2002. Funders should be encouraged to put moremoney into core area land acquisition, perhaps via awilderness fund (see below), rather than into purchaseof development rights on commercial forestland.

• Establish a fund dedicated to the purchase of areassuitable for core wild areas. Of late, most federal,state, and private conservation resources have been usedto purchase development rights on managedtimberlands rather than acquire land for outrightprotection. To complement those efforts, a separatewilderness fund should be established for the purchase,from willing sellers, of land to be established as strictlyprotected conservation areas. Support for this fundcould come from existing some of the existing sourcesmentioned above. New sources of funding should alsobe vigorously pursued.

• Assess the conservation needs of existing andproposed large core areas. Because so many of theexisting and proposed large core areas have beendegraded by human activities, it may be necessary tocarry out temporary management interventions to getthese areas to the point where natural disturbancepatterns and processes have reasserted themselves onthe landscape.

Carry out field studies to verify proposed coreboundaries, and forest and habitat integrity withinproposed core areas, including remaining old-growthforest stands and large roadless areas. The map layersused to develop the proposed wildlands network in thisdocument are derived largely from remotely senseddata. Additional vegetation and fauna studies should,therefore, be conducted in the areas of highest wildlandvalue to confirm their status and to fine-tune thedesign of the network.

• Identify and prioritize roads the removal of whichwould restore large areas to roadless status.Additional analysis and field studies should beconducted to identify roads whose removal would mostbenefit the wildlands network. Economic and culturalconsiderations should be significant factors in thisanalysis.

• Advocate for the designation of new wildernessareas on existing state lands, and for theacquisition of new state lands. In January 2001,70,000 acres of Maine’s Public Reserved Lands were setaside as ecological reserves (only 14% of the areaidentified as potential benchmark reserve sites by arecent scientific assessment [McMahon 1998]). TheMaine Legislature, however, stipulated that no morethan 15% of Maine’s Public Reserved Lands, and nomore than 6% of Reserved Lands that is capable ofsustaining forestry operations, be designated asecological reserves. Further, almost 70% of the landsidentified as core areas are on slopes, in wetlands or athigh elevations. These areas do not represent many ofMaine’s ecosystem types, and their limited size clearlyrender them insufficient for protecting Maine’sbiodiversity. State land management should attempt tobetter balance the needs of economics, recreation, and ecological integrity. On the positive side, new stateland acquisitions may not be bound by the limitationsimposed by the legislature. To the extent possible, newstate land purchases should be designated ecologicalcore areas.

• On the federal lands of the White MountainNational Forest, advocate for the implementationof the recent Roadless Area Conservation Rule,which prohibits road-building and logging inroadless areas of national forests.

• Study the feasibility of establishing a MaineWoods National Park and Preserve. Led by theregional conservation organization RESTORE: TheNorth Woods, a movement to create the proposed 3.2million-acre national park has gained strong regionaland national momentum. Over 100,000 citizens fromacross the country have signed petitions supporting astudy of the park concept; 100 nonprofit organizationshave joined the Maine Woods National Park Coalition;over 350 Business Partners have endorsed a park study;and polls have consistently shown that more than 60%of Maine voters favor the park. Studies also predict thatthe park would have a positive net effect on jobs andthe economy (Power 2001).

The National Park Service should be encouraged toundertake a full feasibility study of the park proposal.All stakeholders would have an opportunity toparticipate, and the study would examine whether theproposed park region encompasses resources of nationalsignificance. If the findings of the study are positiveand there is public and political support for the park,Congress could pass a law authorizing creation of thepark and a purchase boundary. Lands could then be


acquired by the National Park Service throughpurchase from interested sellers or by donation.

• Encourage land trusts to purchase more land forcore protected areas. Because the vast majority ofMaine’s land is held in private ownership, the role ofland trusts is vital to the long-term implementation ofthe Maine Wildlands Network. Dozens of land trustsare working to protect land in Maine for a variety ofuses, ranging from open space and recreational fields totimber lots and nature preserves. Given the scarcity ofland dedicated to core reserves, however, existing landtrusts, especially large regional and national groups,should be encouraged to increase their efforts toacquire large tracts of land for the purpose ofprotection and restoration of ecological integrity. Anew regional entity, the Northeast Wilderness Trust,may also be able to play a role in acquiring andmanaging core wild areas.

4. Support the conservation of large blocks offorest subject to biodiversity compatible forestryand tied to ecologically sound and economicallyviable development.

Strictly protected, core wild areas are the most essentialelements of a natural landscape with ecological integrity.These cores will need to be surrounded and linked by largetracts of land managed in ways that complement thefunctions of the cores. Many private commercial forestlandscan play this crucial buffer and connectivity role. Underthis scenario, however, forestry practices will have tochange considerably, if for no other reason than toguarantee sustained yield of timber over the long term.Guidelines developed by the Maine Low Impact ForestryProject, among others, offer further insight on how torefine forestry to meet ecological, economic, and socialgoals—so-called biodiversity compatible forestry11

(Thompson and Sorenson 2000).Although these commercial forestlands span millions

of acres, employment in forest products across Maine hasfallen significantly—largely due to mechanization—and isprojected to continue to fall. Also, the harvest of manycommercial tree species in recent years has exceeded netforest growth by a considerable margin. To stay withinsustained yield, Maine forest harvests will have to bereduced. This could further accelerate the decline in forestproducts employment in Maine (Power 2001).

The changing nature of the forest products industry inMaine actually provides an opportunity to diversify the

11 Biodiversity compatible forestry is defined as the practice of managing“land for biological diversity and ecological integrity, while stillharvesting timber (Thompson and Sorenson 2000).”

rural economy, increase employment, raise wages, and conserve vast tracts of northern Maine.

A recent Northern Forest Alliance publication echoesthis approach, calling for a three-pronged economicdevelopment strategy based on 1) amenity-baseddevelopment; 2) value-added forest productsmanufacturing; and 3) recreation and tourism based on theregion’s cultural, heritage and ecological values. Under thisstrategy, resource extraction would shift from the “pastpattern of resource depletion to ecological forestry, greatlyenhancing the sustainability of Northern Forestcommunities” (Northern Forest Alliance 2002).

Key Activities:• Encourage the permanent conservation of large

tracts of forestlands dedicated to biodiversitycompatible forestry. One of the most successfulmeans to achieve this end is to establish partnershipsbetween land trusts and responsible, sustainableforestry companies. These types of relationships (e.g.,the recent Hancock Lumber/Maine NatureConservancy partnership) can lead to cooperativeprojects aimed at protecting sizeable blocks offorestland in southern Maine from sprawl, and thusprovide some level of connectivity and habitatprotection. In northern and western Maine, suchprojects can also play a central role in the formation ofa functioning wildlands network, but only if forestrycompanies agree to adopt biodiversity compatibleforestry practices. It is important to emphasize thatthese projects should be considered for forestlandsoutside proposed core wild areas, such as wildlifelinkagesand compatible-use lands.

• Build awareness regarding techniques to eliminateor control exotic species on commercialforestlands. Where exotic species are a threat, takeprecautions to prevent their invasion and spread. Suchprecautions might include keeping openings small andleaving uncut buffers between exotic-infested forestand uninfested forest that is to be cut (Thompson andSorenson 2000). Avoid the use of invasive exotics ofany sort in commercial forestry operations.

The management of forests and farms in an ecologicallysound manner must be linked to local needs and capacities.The following steps can contribute to the diversification ofthe rural economy:

• Re se a r ch pot e nti a l va l ue a dde d pr oduc t s. S pe c i fi c ar e a so f r e s ea r c h i ncl ude wo od r ec o v er y pr o j ec t s , us e o f under - a ppr e c ia t e d s pec i e s, a nd new nic he m a r ke t s t ha t c i rc um -v e nt i nt e r na t i ona l c o m pe t i ti o n i n co m m odi t y pr oduc ts .


Investigate new wood product markets in thefollowing three areas:

Institutional buying – Pursue the strategy of workingwith large state institutions (e.g., universities andcolleges, state government, hospitals) to procuresustainably harvested and manufactured forest productsfrom local businesses.

Certification – Forest Stewardship Council (FSC)certification has encountered some barriers in Maine,including a mismatch of certified species with thosedemanded by secondary manufacturers. Help is neededto fill in the missing links of the supply and demandchain, primarily by working with small secondarymanufacturers to begin making their products withcertified wood, encouraging those manufacturers toundergo chain-of-custody certification, and helpingthem find markets for their certified products.

I n f o r ma t i o n b r o k e r i n g – Es t a bl i s h br o ke r s f o r m a nuf a c t ur e r s t o f o r m f l e x i bl e m a nu f a c t ur i ng ne t wo r ks i n o r de r t o m e e t l a r ge o r d e r s t h a t i nd i v i dua l m a nu f a c t ur e r s wo ul d ha v e t o t ur n do wn due t o i ns u f f i c i e nt c a p a c i t y .

• Encourage entrepreneurship and innovation inMaine’s forest products industry, particularlyamong small businesses. It is important to increasethe number of young conservation-minded peopleentering the industry. An example of such an effort isthe Maine Wood Products Association’s collaborationwith other groups in the Northern Forest region on aproject to create materials that highlight the diverseopportunities in the industry and make forest-relatedwork more attractive to young people.

• St u dy t he pot e n t i a l f or non - t i m be r f or e s t pr o duc t s ,i nc l udi ng t our i s m , t o be c om e a l a r ge r e c onom i c s e c t or . I t i s e s s e nt i a l t o d i v e r s i f y t he e c o no m i e s o f f o r e s t e d c o m m un i t i e s i nt o e c o l o gi c a l l y - s us t a i na bl e l i v e l i ho o ds ba s e d o n t he us e a nd no n- us e v a l u e s o f t he f o r e s t . P o s s i bl e ne w v e nt ur e s i nc l ude t o ur i s m - r e l a t e dbus i ne s s e s ( e .g ., bi r d wa t c hi ng, hi ki ng , c r o s s - c o un t r y s ki i ng) , c r a f t c e nt e r s , wi l d e di b l e s , a nd o r n a m e nt a l s .

5. Pursue policy and research initiatives onpublic conservation funding, “mega-easement”projects, alternative incentives for conservation,and the effects of air pollution on aquatic andforested ecosystems.

Several science and policy issues directly affecting theimplementation of the Maine Wildlands Network deservemore analysis. One of these is the “mega-easement”phenomenon discussed in Section II.

In the scientific realm, the threat posed by airpollution on interior forest species needs further study.Scientists from state and federal agencies and majoruniversities have sampled rain (and snow), and havediscovered that New England’s precipitation containsmercury in concentrations that exceed EPA’s safe level formercury in lakes and streams. Mercury contamination ofrain has been documented in Maine’s parks and ruralcommunities, including Acadia National Park, wheremercury levels in rain are up to four times as high as theEPA’s aquatic life standard for mercury in surface water.That standard is also applied to wildlife (National WildlifeFederation 2000).

Although the state of Maine recognizes that there is asevere problem with invasive aquatic species, it isimportant to broaden policy efforts to the federal level, andto lobby for comprehensive measures to control bothterrestrial and aquatic exotics.

Key Activities:• Carry out a cost-benefit analysis of “mega-

easements.” Given the controversies surrounding thePingree and West Branch projects, it would beworthwhile to conduct a careful cost-benefit analysis ofthe conservation impacts of these deals, consideringespecially the conservation benefits obtained withpublic funds. This analysis should consider ecological,economic, forest management, and social issues.

• Research alternative economic incentives forprotecting wildlands. More and more research isdemonstrating that natural ecosystems provide a hugearray of goods and services, but few of these serviceshave value in the marketplace. One promisingpossibility, however, is carbon banking. Theapplicability and feasibility of establishing incentivesfor this practice in Maine should be investigatedfurther. The possibility of establishing “conservationperformance payments” should also be researched. Thisconcept, wherein individuals, communities, or otherentities are paid directly for conservation actions, suchas not logging a piece of forest to which they have fullrights, is gaining increasing currency in the developingworld (Ferraro 2001; Ferraro and Simpson 2001). Itsapplicability to Maine should be explored further.

• S ol i d i f y und e r s t a nd i ng of the ef f e c t s of at m os ph e r i c p ol l u t a n t s on for e s t int e gr i t y , in t e r i o r fo r e s t spe c i e s ( e s pe c i a l l y foc a l sp e c i e s ) , and fr e s h w a t e r e c o s y s t e m s . G i v e n r e c e n t f i nd i ng s o n hi gh l e v e l s o f m e r c u r y i n p r e c i p i t a t i o n i n M a i n e , f u r t he r s t u dy o f i t s e f f e c t s o n f r e s hwa t e r a n d f o r e s t e d e c o s y s t e m s , i n c l ud i n g i nt e r i o r f o r e s t s pe c i e s , i s wa r r a n t e d . Fu r t h e r s t ud y o f t h e e c o s y s t e m e f f e c t s o f a c i d p r e c i p i t a t i o n i s a l s o n e e de d.


• Lobby the U.S. Congress, U.S. Departments ofAgriculture and Commerce, and appropriateMaine state agencies, to improve measures todetect, eliminate, and control exotic species.

• Promote legislation that provides screening and adecision making process on exotics imported forpets, science, agriculture, etc.

• Promote legislation that prevents the planting/release of known invasive exotics on state andfederal lands.

• Lobby the US Department of Agriculture to adopteffective controls on imports of foreign wood andestablish a monitoring system to detect intro-ductions of forest pests. A ban on imports ofunprocessed wood should be considered.

• Build awareness of members of committees inCongress on the threat of exotic pests and weedssuch that the Animal Plant Health InspectionService (APHIS) places a higher priority ondetection and control, and receives additionalfunds to carry out its duties.

6. Promote the restoration and protection offocal species in the Maine Wildlands Network.

The needs of the focal species described in sections IVand V, and Appendix 3, of this document drive the designof the proposed Maine Wildlands Network, and theirrestoration and protection is key to realizing a viablenetwork over the long term. The restoration of some focalspecies will probably have to wait until social and politicalconditions improve—and until there are more areas withsufficient habitat security. But restoration and protection ofmany other species can proceed apace. Below we list a fewkey, broad activities that should be carried out in supportof focal species recovery. The needs of the different speciesvary so significantly, however, that we do not attempt tolist them exhaustively here. Please refer to the individualfocal species descriptions for recommended actions.

Key Activities:• Carry out comprehensive natural resource

inventories to improve our knowledge of thepopulation and distribution of focal species inMaine.

• Carry out fieldwork to assess areas important forwildlife movement between identified cores, suchas feeding and breeding areas, stopover points andmovement and dispersal zones (including areas

with minimal human population and roaddensity).

• Similarly, assess watershed linkages betweenidentified cores and between terrestrial andcoastal ecosystems.

• Develop recommendations and guidelinesoriented toward federal and state agencies, otherconservation groups, large landholders and thebroader public, for the recovery of focal species.

• Design and implement an outreach strategy tocommunicate focal species recommendations.

• Conduct a systematic analysis of the current stateof wildlife management in Maine. This includesreviewing the effects of coyote snaring on lynx andother focal species, trapping pressures on marten andotter, and current state practices (e.g., hatcheries,aquaculture licensing) that should be changed toenable Atlantic salmon recovery. See the focal speciesprofiles for additional recommended analyses.

7. Advocate for the restoration and maintenanceof natural ecological and evolutionary processes,including wildfire, insect outbreaks, predator/prey dynamics, natural succession, and floodregimes in all components(cores, study areas,linkages) of the Maine Wildlands Network.

Key Activities:• Encourage land trusts and others to incorporate

protections of biological integrity into theirconservation easements. Although land trustsconserve at some level thousands of acres annuallyacross the country, they often do not integrateecological principles into their easements. If land trustsare to play a role in the conservation of biodiversity, asopposed to simply protecting open space, they mustdevelop and incorporate language into conservationeasements that provides maximum protection forbiological integrity. No easements should in any wayprohibit or limit biological protection (e.g., perpetuallogging easements that mandate logging in thefuture). Forever wild conservation easements should beused whenever possible.

More specifically, land trusts should incorporatethe principles of conservation biology into easementsand examine the larger landscape context of individualparcels. Easements should take into account whetherthe area to be protected has value as a core wild area or


a wildlife linkage, or might be more appropriate as acompatible-use area, and be written accordingly. Forexample, easements for core forests may restrict vehicleaccess so as to protect sensitive species. The MaineForest Biodiversity Project has produced a guide(Flatebo et al. 1999) that takes a good step forward inthis regard. The Wildlands Project, the Society forConservation Biology, and other groups cancomplement this work by developing a specific set ofguidelines aimed at restoring and maintainingecological integrity. Even in compatible-use lands,ecological considerations should be included in theeasement (e.g., agricultural easements should providemaximum buffers along streams and restrict pesticideuse; forestry easements should establish minimumstandards for coarse woody debris, standing dead trees,and so on). The Maine Low Impact Forestry Projectand Maine Forest Biodiversity Project provide excellentguides for this purpose.

Ultimately, the challenge for land trusts is todesign conservation easements, which are inherentlyperpetual, that can accommodate new scientificfindings and adjust to the naturally changinglandscape. Some conservation easements may includestages of protection. For example, an area intended toprovide “Forever Wild” core habitat with minimallong-term management may in the short-term requireactive restoration, such as the removal of roads orexotic species.

• Encourage the state of Maine and the federalgovernment to manage public lands to favor therestoration and maintenance of natural ecologicaland evolutionary processes. Most public lands inMaine—nearly 700,000 acres—lie outside of officiallydesignated core areas. Traditionally, many of theselands have been managed chiefly for the production oftimber and game species. With the evolution ofconservation science, the technical underpinnings arenow available to expand management objectives toinclude:

- Enhancement and restoration of natural resourceconditions (e.g., promote older, uneven-aged foreststands, restore riparian buffers, and enhancewildlife habitats);

- Protection of wild nature at site specific, ecosystemand landscape levels; and

- Integration of disparate public land holdings intobroader regional reserve networks.

Technical working groups involving a cross-section of public and private entities should be formed

for the purpose of devising improved managementplans on state and federal lands.

8. Promote the restoration and protection offreshwater ecosystems, including salmon-bearing streams.

Freshwater ecosystems are an essential complement tothe terrestrial ecosystems that form the backbone of theMaine Wildlands Network. Riparian areas, for example,play a key role in establishing connectivity between coreareas. Freshwater ecosystems are threatened by thecontinued presence of dams, invasions of exotic species,escaped fish from the aquaculture industry, the withdrawalof water to irrigate blueberry fields, and the use ofpesticides and herbicides. Certain forest managementpractices may also damage freshwater ecosystems. Tocounter these threats, the following actions should bepursued:

Key Activities:• Seek congressional support for and designation of

additional National Wild and Scenic Rivers inMaine. Thus far in Maine, this federal designation,designed to preserve rivers “in their free-flowingcondition to protect the water quality of such riversand to fulfill other vital national conservationpurposes,” has only been bestowed upon 92.5 miles ofthe Allagash River in the north-central part of thestate.

• Identify dams whose costs (ecological and other)outweigh their benefits. The removal of the Edwardsdam on the Kennebec River in 1999 demonstratesconclusively the environmental benefits of damremoval, including restored fisheries, improved waterquality, and new economic opportunities (NaturalResources Council of Maine 2000). There areundoubtedly many dams in the state that could beremoved with little or no lasting economic impact;removal may even generate positive economic benefits.Using the Wildlands Network Design as a guide, acomprehensive study should be conducted to identifydams most in need of removal. Dams slated for FederalEnergy Regulatory Commission license renewal shouldbe especially carefully studied.

• Carry out research to improve understanding ofthe ecological values of headwater streams andother small watercourses, and the effects of forestmanagement on the integrity of thesewatercourses. Thousands of miles of small streamscourse through the forests of Maine. In fact, theBoundary Mountains of western Maine may contain


one of the highest densities of headwaters streams onearth. Because there are so many more small streamsthan big ones, and because heavy wood harvestingmachinery frequently encounters such streams, it isimportant to determine how harvesting should takeplace around these watercourses. The Manomet Centerfor Conservation Science is carrying out research as partof its Headwater Streams Project, with the goal ofdetermining how wide a forest buffer should be toprotect the broad array of ecological values provided bythese streams (Hagen 2001).

• Identify and assess the most ecologicallysignificant portions of all major rivers andwatersheds in the proposed Wildlands Network,especially headwaters, required for maintainingthe viability of native aquatic species (e.g., Atlanticsalmon populations). As with terrestrial cores andlinkages, field studies should be conducted in andaround key freshwater ecosystems to confirm theirstatus and to fine-tune the Wildlands NetworkDesign.

• Analyze the options available to permanentlyprotect the most important watercourses andwatersheds in the Wildlands Network. The state ofMaine, the federal government and other major actorsare working vigorously to conserve the state’swatercourses, especially major salmon rivers. In somecases, land around these rivers has been acquiredoutright. In other cases, such as the Machias River, thestate plans to complement acquisition with thepurchase of conservation easements on additional land.The purchase of such easements is controversial,however. As with the purchase of easements on largetracts of forestland not threatened by imminentdevelopment, a careful analysis of the efficacy ofeasements for salmon habitat conservation should beconducted.

• Seek better information on water withdrawals,and investigate the feasibility of significantlyreducing withdrawals, or otherwise mitigating theeffects of this practice. Maine’s current waterwithdrawal database is inadequate to assess, reduce,and find solutions to water use demands. This lack ofdata hinders the development of sustainable water useapproaches. While water use information is required ofmany water users who need permits under Maine law,it is haphazard at best. Many major water users are notrequired to report their water use at all. Indeed, waterwithdrawals that are subject to Land Use RegulationCommission permits require water use informationwhile nothing at all is required of similarly situated

water users in Department of EnvironmentalProtection jurisdiction. The state of Maine shouldrequire the reporting of major water withdrawals toprovide the state and the public with information thatis critical to the development of sound approaches forsustaining one of Maine’s most valuable publicresources—fresh water (Day 2001). Sound data onwithdrawals will also contribute to the development ofproposals to mitigate the effects of water withdrawals,such as the use of off-stream storage ponds that wouldstore water from high flow periods and provide it foruse during low flow periods.

• Develop and disseminate recommendations for acomprehensive management program to controland prevent the spread of exotic plant and animalspecies in freshwater ecosystems.The state of Maineclearly recognizes that invasive aquatic species are aserious problem. To this end, it has convened aninvasive aquatic species workgroup, composed ofrepresentatives from state agencies and privateconservation groups. In early 2001 that grouprecommended the development of a comprehensivestate management plan for the prevention and controlof non-indigenous aquatic nuisance species. They alsorecommended that an Invasive Species Committee,composed of state agencies, federal agencies, sportingorganizations, boating interests and environmentalorganizations, oversee the development of the plan.Invasive aquatic species plans are based on thefollowing three goals (Maine Invasive Aquatic SpeciesWork Group 2001):

- Preventing new introductions of non-indigenousinvasive aquatic species into the waters of the state.

- Limiting the spread of established populations ofnon-indigenous invasive aquatic species intouninfested waters of the state.

- Abating harmful ecological, economic, social, andpublic health impacts resulting from infestation ofnon-indigenous invasive aquatic species.

The workgroup further recommends that the stateof Maine improve its education and public outreachprograms on invasive aquatic species, modify andimprove the enforcement of state laws related toinvasive aquatic species, improve coordination amongstate agencies, and establish a fully-funded position tocoordinate efforts state-wide.


ConclusionThe eight components of the action plan presented

above provide a set of recommended “lines of work” thatcould be carried out by the conservation community, notonly in Maine, but also regionally, to achieve the largergoals and objectives of the Maine Wildlands Network.Short-term action plans for each component should bedeveloped to establish the specific mix of participatingorganizations, responsibilities, timelines, budgets and soon. Some form of state-wide coordination capacity, asdescribed in Component 1, will greatly facilitate thedevelopment and execution of these action plans. Thiscoordinating entity will also make it easier to integrateMaine’s efforts with those of neighboring states andprovinces. The specific conservation work spelled out inComponents 2 through 8 represent a significant expansionon current efforts now underway. However, it is clear thatmany Maine groups are comfortable with landscape-scaleplanning and management and we hope that this plan willprovide useful guidance and focus as we embark on a new,grander phase of conservation in Maine.


X. CONCLUSIONVoices from the past provide invaluable insight to the

future. During his 1853 journey to Maine, Henry DavidThoreau (1864) reflected on the fate of forests:

The civilized man not only clears the landpermanently to a great extent, and cultivates openfields, but he tames and cultivates to a certainextent the forest itself. By his mere presence, hechanges the nature of the trees as no other creaturedoes…It has lost its wild, damp, and shaggy look,the countless fallen and decaying trees are gone,and consequently that thick coat of moss whichlived on them is gone too.

Thoreau goes on to share his awe and concern for theNorth Woods that had so captured his imagination; for aMaine that still maintained a wild quality that his homestate had lost:

The greater part of New Brunswick, the northernhalf of Maine, and the adjacent parts ofCanada…are still covered with an almostunbroken pine forest. But Maine, perhaps, willsoon be where Massachusetts is. A good part of herterritory is already as bare and commonplace asmuch of our neighborhood, and her villagesgenerally are not so well shaded as ours.

Fortunately, Thoreau’s foreboding has yet to be fullyrealized, although much of the forest that he saw has beenreplaced by a much younger, less complex forest.Nevertheless, Maine presents both an exceptionalopportunity and a powerful mandate for protecting andrestoring native species and habitat. Nowhere else in thenortheastern United States is there so much forested landleft undeveloped, so much potential for wildlandsconservation. The challenge now is to chart the way aheadon a broad, landscape scale, and the Maine WildlandsNetwork Vision seeks to do just that.

But we need to begin acting now if we wish to makethe network a reality. Current trends suggest that thesituation for biodiversity and wildlands in Maine will onlybecome more urgent as land transactions continue at theirfurious pace. Since October 1998, an astonishing 22percent of Maine’s lands has changed hands. Ownershippatterns are changing too; many Maine-based paper andlumber companies have left, replaced by internationalcorporations, pension funds, family trusts, and wealthyindividuals (Curtis 2001). These changes in ownership offernew opportunities for conservation: many of the newowners have different objectives for their land than the

traditional paper and lumber companies and have longer-term planning horizons, both of which factors may aidconservation. New emphases on green certification andvalue-added forest products also signal a potentiallypositive shift in resource management toward moreecologically sound forestry. Nevertheless, even areas subjectto state-of-the-art management are no substitute for corewild areas with a high degree of permanent protection. Thelarge amount of land on the market offers tremendousopportunity in this regard: the chance has never been betterto set aside large tracts of land for the purpose of protectionand restoration of ecological integrity, whether throughoutright acquisition, purchase of “forever wild”conservation easements, or some other mechanism.

Despite promising examples of ecologically soundresource management, large portions of the state continueto be threatened by on-going land-use practices. Accordingto Gawler et al. (1996), “land-use trends point toincreasing fragmentation from development in thesouthern part of the state and increasing fragmentation andforest simplification from harvest activities, such asshortened rotations, in the northern part of the state.” Theauthors conclude: “prudence dictates that we begin todevelop biodiversity conservation measures now, given thedata at hand.” This conclusion is supported by recentfindings in the field of conservation biology indicating thatlarge, well-connected areas capable of hosting a fullcomplement of species—including large carnivores andothers that have been absent from the landscape fordecades—will be needed to ensure biodiversityconservation over the long-term.

As discussed in the network design summary (SectionVIII), the proposed Maine Wildlands Networkrecommends approximately 28,725 km2 of core wild areas,1,175 km2 of aquatic linkages, and 4,671 km2 of aquaticbuffers (Table VIII.1). By comparison, the Maine ForestBiodiversity Project (MFBP) in 1998 published a reportidentifying potential benchmark reserve sites encompassing2,019 km2 (McMahon 1998). The MFBP report cautionsthat, even if all of their qualifying ecoreserves wereprotected, they alone would not maintain—much lessrestore—biodiversity. In January 2001, Maine acted on therecommendations of the MFBP and mandated theestablishment of a series of ecological reserves on existingPublic Reserved Lands. Despite the cautionary words of thefinal MFBP report, only 70,000 acres (roughly 0.3%) of thestate’s land was protected. Further, the State Legislaturestipulated that no more than 15% of Maine’s PublicReserved Lands, and no more than 6% of such land that iscapable of sustaining forestry operations, be designated asecological reserves.


These actions highlight the fact that the quantity andquality of public lands in Maine are inadequate to meet theneeds of biodiversity, despite the rarity of many of Maine’snatural communities and species, and the threat posed bymore forest fragmentation and destruction. More publiclands with higher levels of protection are clearly needed.The potential of private conservation lands (managedprimarily for biodiversity) and wildlands philanthropy(Hitt 2001; Carpenter 2001) to contribute to thedevelopment of a wildlands network in Maine must also beexplored.

Most recent private conservation efforts in Maine,however, aimed at maintaining “working forests,” generallyfail to adequately protect biodiversity and wildlands.Although these transactions seek to meet the demands forpublic tourism, maintain the appearance of natural beauty,and manage for sustained production, they have notsufficiently incorporated preservation of biodiversity andecological integrity into the equation. Gawler et al. (1996)stress that:

Evidence indicates that the working forest can bemanaged to maintain many aspects of biologicaldiversity in an economically viable framework, butthat certain aspects—in particular, representative,unmanaged ecosystems and some habitatspecialists—will be incompatible with forestry asit is generally practiced.

We therefore strongly recommend that theconservation community turn its attention to acquiring andfully protecting large areas of contiguous forested habitatto ensure a future that preserves Maine’s native species andhabitat, natural processes, and wilderness values. Theseintegral strands of ecological integrity cannot be separated.Aldo Leopold (1949) offers us this wisdom from anothertime and place:

A representative sample of these [wilderness] areascan, and should, be kept. Many are of negligible ornegative value for economic use. It will becontended, of course, that no deliberate planningto this end is necessary; that adequate areas willsurvive anyhow. All recent history belies so

comforting an assumption. Even if wild spots dosurvive, what of their fauna? The woodlandcaribou, the several races of mountain sheep, thepure form of woods buffalo, the barren groundgrizzly, the freshwater seals, and the whales areeven now threatened. Of what use are wild areasdestitute of their distinctive faunas?

The acquisition and protection of large wilderness areascertainly represents the highest near-term priority of theMaine Wildlands Network. But this only takes us part ofthe way to achieving the ultimate goal of restoring andprotecting the ecological integrity of Maine. Over the longterm, we must embark on an ambitious program ofrestoration, based on the principles laid out in Soulé andNoss (1998) and Simberloff et al. (1999), among others.Only then will we again see the full array of species,ecosystems, and ecological processes on the landscape.

There is concern that the transformation of northern,eastern, and western Maine from a landscape dominated bycommercial timberland to one where wildernesspredominates would harm the rural economy and culture.As Power (2001) and others demonstrate, however,commercial forest management and the traditional pulpand paper industries are playing a smaller and smaller rolein the rural economy every year. In fact, a rural economybased on the amenities offered by a network of conservedareas may offer the best development option for ruralMaine. Ultimately, we concur with Noss et al. (1999) thata conservative approach, based on biological and ecologicaldata, should set the “sideboards” within whichsocioeconomic options are evaluated, and that “…thisapproach is in line with the historical observation thathuman cultures are much more adaptable to rapidenvironmental change than many non-human species.”

In closing, we wish to express our hope that the MaineWildlands Network Design will serve as a blueprint for therestoration and protection of biodiversity and wildlands inMaine. We also hope that the Maine Wildlands Networkwill become part of a larger wildlands network linkingNew England, New York and eastern Canada, which, inturn, will connect with other regions to restore criticalecological linkages throughout North America.


REFERENCES

Alexander, L. L. 1985. Trouble with loons. Living Bird Quarterly 4: 10-13.

Allen, T. 2001. Press Release from U.S. Representative Tom Allen. 17 October 2001.

Altherr, T. L. 1994. The Catamount in Vermont Folklore and Culture, 1760-1900. In Proceedings of the Eastern Cougar Conference,1994, ed. J.W. Tischendorf and S.J. Ropski. Fort Collins, CO: American Ecological Research Institute.

American Rivers. 2001. Ten reasons why dams damage rivers. Retrieved January 23, 2001, fromhttp://www.amrivers.org/damremoval/tenreasons.htm .

Andelman, S. J. and W. F. Fagan. 2000. Umbrellas and flagships: Efficient conservation surrogates or expensive mistakes?Proceedings of the National Academy of Sciences. Vol 97(11): 5954-5959.

Andelman, S., I. Ball, F. Davis and D. Stoms. 1999. SITES V 1.0: An analytical toolbox for designing ecoregional conservationportfolios. A manual prepared for the Nature Conservancy.

Anderson, A. E., D. C. Bowden and D. M. Kattner. 1992. The puma on the Uncompahgre Plateau, Colorado. ColoradoDivision of Wildlife Technical Publication, 40: 1-116.

Anderson, M., P. Comer, D. Grossman, C. Groves, K. Poiani, M. Reid, R. Schneider, B. Vickery and A. Weakley. 1999.Guidelines for representing ecological communities in ecoregional conservation plans. Boston, MA: The NatureConservancy.

Apfelbaum, S. I. and P. Seelbach. 1983. Nest tree, habitat selection and productivity of seven North American raptor speciesbased on the Cornell University Nest Record Card Program. Raptor Research 17: 97-113.

Atkinson, I. A. E. 1990. Ecological restoration on islands: prerequisites for success. In: Ecological restoration of New Zealandislands, ed. D. R. Towns, C. H. Daugherty, and I. A. E Atkinson. Wellington: New Zealand Department ofConservation.

Bailey, R. G. 1995. Description of the ecoregions of the United States. Miscellaneous Publication No. 1391, Washington,D.C.: USDA Forest Service.

Bailey, R. G., P. E. Avers, T. King and W. H. McNab (eds.). 1994. Ecoregions and subregions of the United States. WashingtonD.C.: USDA Forest Service.

Banfield, A. W. F. 1974. The Mammals of Canada. Toronto, Canada: University of Toronto Press.

Bangs, E. E. and S. H. Fritts. 1993. Reintroduction of gray wolves to Yellowstone National Park and central Idaho. EndangeredSpecies Technical Bulletin. 18(3): 1, 18-20.

Barry, G. R., T. P. Rooney, S. J. Ventura and D. M. Waller. 2001. Evaluation of biodiversity value based on wildness: a studyof the western northwoods, upper great lakes, USA. Natural Areas Journal 21: 229-242.

Beazley, K. 1998. A focal-species approach to biodiversity management in Nova Scotia. Ph.D. Dissertation, DalhousieUniversity, Halifax, Nova Scotia, Canada.

Bednarz, J. C. 1979. I. Productivity, nest sites, and habitat of red-shouldered and red-tailed hawks in Iowa. II. Status, habitatutilization, and management of red-shouldered hawks in Iowa. Ames, IA: M.S. thesis, Iowa State University.

Bednarz, J. C. and J. J. Dinsmore. 1982. Nest-sites and habitat of red-shouldered and red-tailed hawks in Iowa. Wilson Bulletin94: 31-45.

Beier, P. 1993. Determining minimum habitat areas and habitat corridors for cougars. Conservation Biology 7(1): 94-108.

_____. 1996. Metapopulation models, tenacious tracking, and cougar conservation. In Metapopulations and Wildlife Conservation,ed. D. R. McCullough. Washington, D.C.: Island Press.

Beier, P. and R. F. Noss. 1998. Do habitat corridors provide connectivity? Conservation Biology 12(6): 1241-1252.


Beier, P., D. Choate and R. H. Barrett. 1995. Movement patterns of mountain lions during different behaviors. Journal ofMammalogy 76(4): 1056-1070.

Beland, K. 1984. Strategic plan for management of Atlantic salmon in the state of Maine. Bangor, Maine: Atlantic Sea RunSalmon Commission.

Bennett, A. F. 1999. Linkages in the landscape: the role of corridors and connectivity in wildlife conservation. Cambridge, UnitedKingdom: World Conservation Union Publications.

Bennett, D. B. 1988. Maine’s Natural Heritage. Camden, ME: Down East Books.

_____. 1996. The Forgotten Nature of New England. Camden, ME: Down East Books.

Bissonette, J. A., D. J. Harrison, C. D. Hargis and T. G. Chapin. 1997. The influence of spatial scale and scale-sensitiveproperties on habitat selection by American marten. In Wildlife and Landscape Ecology, ed. J. A. Bissonette. New York,NY: Springer.

Boal, C. W. and R. W. Mannan. 1994. Northern goshawk diets in ponderosa pine forests on the Kaibab Plateau. Studies inAvian Biology 16: 97-102.

Bolgiano, C., T. D. Lester, D. W. Linzey and D. S. Maehr. In press. Field evidence of cougars in eastern North America. InProceedings of the 6th Mountain Lion Workshop, San Antonio, Texas, December 2000, ed. L.A. Harveson. Texas Parks andWildlife Department.

Boone, R. B. 1996. An assessment of terrestrial vertebrate diversity in Maine. Ph.D. Dissertation, Orono, ME: University ofMaine.

Bosakowski, T. and R. Speiser. 1994. Macrohabitat selection by nesting northern goshawks: implications for managing easternforests. Studies in Avian Biology 16: 46-49.

Brand, C. J. and L. B. Keith. 1979. Lynx demography during a snowshoe hare decline in Alberta. Journal of WildlifeManagement 43: 827-849.

Brown, W. H. 1971. Winter population trends in the red-shouldered hawk. American Birds 25(5): 813-817.

Bryant, A. A. 1986. Influence of selective logging on red-shouldered hawks, Buteo lineatus, in Waterloo region, Ontario, 1953-1978. Canadian Field-Naturalist 100: 520-525.

Buskirk, S. W. and L. F. Ruggiero. 1994. American Marten. In The scientific basis for conserving forest carnivores: Americanmarten, fisher, lynx, and wolverine in the western United States, ed. L. F. Ruggiero, K. B. Aubry, S. W. Buskirk, L. J.Lyon and W. J. Zielinski. General Technical Report RM-254. Ft. Collins, CO: USDA Forest Service Rocky MountainForest and Range Experiment Station.

Burnett, M. R., P. V. August, J. H. Brown and K. T. Killingbeck. 1998. The influence of geomorphological heterogeneity onbiodiversity I. A patch-scale perspective. Conservation Biology 12(2): 363-370.

Buskirk, S. W., S. C. Forrest, M. G. Raphael and H. J. Harlow. 1989. Winter resting site ecology of marten in the centralRocky Mountains. Journal of Wildlife Management 52: 191-196.

Canadian Parks And Wilderness Society, Ottawa Valley Chapter (CPAWS). 2002. Algonquin Wolves: A Population at Risk.Retrieved April 20, 2002, from <http://www.cpaws-ov.org/AlgonquinWolves.htm>

Carpenter, S. R. and J. F. Kitchell. 1993. The trophic cascade in lakes. Cambridge: Cambridge University Press.

Carpenter, M. 2001. Park proposal stays afloat, an acre at a time. Christian Science Monitor. 20 September 2001.

Carroll, C., R. F. Noss and P. C. Paquet. 2000. Carnivores as focal species for conservation planning in the Rocky Mountainregion. Unpublished report to World Wildlife Fund Canada. 61 pp.

Carroll, C., R. F. Noss, N. H. Shumaker and P. C. Paquet. 2001. Is the return of wolf, wolverine, and grizzly bear to Oregonand California biologically feasible? In Large Mammal Restoration: Ecological and Sociological Implications, ed. D. Maehr, R.F. Noss, J. Larkin. Washington, D.C.: Island Press.

Chapin, T. G., D. J. Harrison and D. D. Katnik. 1998. Influence of landscape pattern on habitat use by American marten in anindustrial forest. Conservation Biology 12(6): 1327-1337.


Chapin, T. G., D. J. Harrison and D. M. Phillips. 1997. Seasonal habitat selection by marten in an untrapped forest preserve.Journal of Wildlife Management 61: 707-717.

Chubbs, T. E., L. B. Keith, S. Mahoney and M. McGrath. 1993. Responses of woodland caribou (Rangifer tarandus caribou) toclear-cutting in east-central Newfoundland. Canadian Journal of Zoology 71: 487-493.

Clark, T. W., E. Anderson, C. Douglas and M. Strickland. 1987. Martes americana. Mammalian Species Number 289. AmericanSociety of Mammalogists.

Connor, K. J., W. B. Ballard, T. Dilworth, S. Mahoney and D. Anions. 2000. Changes in structure of a boreal forestcommunity following intense herbivory by moose. Alces 36: 111-132.

Corn, J. G. and M. G. Raphael. 1992. Habitat characteristics at Marten subnivean access sites. Journal of Wildlife Management 56:442-448.

Crooks, K. R. and M. E. Soulé. 1999. Mesopredator release and avifaunal extinctions in a fragmented system. Nature 400: 563-566.

Curtis, Wayne. 2001. Easement does it: Maine woods emerge as ground zero for a grand land conservation experiment. GristMagazine. 8 January 2001. Earth Day Network. Retrieved November 21, 2001,from< www.gristmagazine.com/grist/maindish/curtis010801.stm >.

Davis, M. B. 1996. Old growth in the east: A survey. Richmond, VT: The Cenozoic Society, Inc.

Day, L. R. 2001. Testimony of Laura Rose Day, Watershed Project Director, Natural Resources Council of Maine, in support ofL.D. 1488, An Act to Require Public Information Regarding Water Use By Major Water Users. Natural ResourcesCommittee, Maine State Legislature. April 4, 2001.

DeCalesta, D. 1994. Impact of deer on interior forest songbirds in northwestern Pennsylvania. Journal of Wildlife Management58: 711-718.

DeLorme. 1996. Maine atlas and gazetteer. Freeport, ME: Delorme.

deMaynadier, P. and M. Hunter, Jr. 1997. The role of keystone ecosystems in landscapes. In Ecosystem Management: Applicationsfor Sustainable Forests and Wildlife Resources, eds. M.S. Boyce and A. Haney. New Haven and London: Yale UniversityPress.

Dobson, A., K. Ralls, M. Foster, M. E. Soulé, D. Simberloff, D. Doak, J. A. Estes, L. S. Mills, D. Mattson, R. Dirzo, H. Arita,S. Ryan, E. A. Norse, R. F. Noss and D. Johns. 1999. Corridors: Reconnecting fragmented landscapes. In ContinentalConservation: Design and Management Principles for Long-term, Regional Conservation Networks, eds. M. E. Soulé and J.Terborgh. Covelo, CA and Washington, D.C.: Island Press.

Downing, R. L. 1996a. Investigation to determine the status of the cougar in the southern Appalachians. In Proceedings of theEastern Cougar Conference, 1994, ed. J. W. Tischendorf and S. J. Ropski. Fort Collins, CO: American Ecological ResearchInstitute.

_____. 1996b. The cougar in the East. In Proceedings of the Eastern Cougar Conference, 1994. ed. J. W. Tischendorf and S. J.Ropski. Fort Collins, CO: American Ecological Research Institute.

Droege, S. and J. R. Sauer. 1990. The North American breeding bird survey, annual summary, 1989. U.S. Fish WildlifeService, Biological Report. 90(8).

Dubuc, L. J., W. B. Krohn and R. B. Owen, Jr. 1990. Predicting occurrence of river otters by habitat on Mount Desert Island,Maine. Journal of Wildlife Management 54: 594-599.

Dunstone, N. 1993. The Mink. London, United Kingdom: T and A D Poyser Ltd.

Ebbers, B. C. 1989. Relationships between red- shouldered hawk reproduction and the environment in northern Michigan.Unpublished report submitted to Michigan Department of Natural Resources.

Ehrlich, P. R., D. S. Dobkin and D. Wheye. 1992. Birds in Jeopardy: the Imperiled and Extinct Birds of the United States andCanada, Including Hawaii and Puerto Rico. Stanford, CA: Stanford University Press.

Eldridge, N. 1991. The Miner's Canary: Unraveling the Mysteries of Extinction. New York: Prentice Hall Press.


Environmental Protection Agency (EPA). 2001. Environmental Effects of Acid Rain. Boston, Massachusetts: EPA NewEngland. EPA Acid Rain Program. Retrieved November 30, 2001, from< http://www.epa.gov/region01/eco/acidrain/enveffects.html >.

Erdman, T. C., D. F. Brinker, J. P. Jacobs, J. Wilde and T. O. Meyer. 1998. Productivity, population trend, and status ofnorthern goshawks, Accipiter gentilis atricapillus, in northeastern Wisconsin. Canadian Field-Naturalist 112:17-27.

Estes, J. A. and J. F. Palmisano. 1974. Sea otters: their role in structuring nearshore communities. Science 185:1058-1060.

Estes, J. A., N. S. Smith and J. F. Palmisano. 1978. Sea otter predation and community organization in the western AleutianIslands, Alaska. Ecology 59: 822-833.

Estes, J.A., M. T. Tinker, T. M. Williams and D. F. Doak. 1998. Killer whale predation on sea otters linking oceanic andnearshore ecosystems. Science 282: 474-475.

Evers, D. C. 1992. A guide to Michigan's endangered wildlife. Ann Arbor, MI: University of Michigan Press.

Falk, J. 1973. Regional inequality and rural development: The Maine paper industry. Senior Honors Thesis, HarvardUniversity, Cambridge, MA.

Ferraro, P. J. 2001. Global Habitat Protection; limitations on development interventions and a role for conservationperformance payments. Conservation Biology 15(4): 990-1000.

Ferraro, P. J. and R. D. Simpson. 2001. The cost-effectiveness of conservation payments. Resources for the Future Discussion PaperNo. 00-31, July 2000 (Revised, April 2001). Washington, D.C.: Resources for the Future.

Fitzgerald, J.P., C.A. Meaney and D.M. Armstrong. 1994. Mammals of Colorado. Denver Museum of Natural History andUniversity Press of Colorado.

Flatebo, G., C. Foss and S. Pelletier. 1999. Biodiversity in the Forests of Maine:Guidelines for land management. Orono, ME:University of Maine Cooperative Extension.

Foley, R. E., S. J. Jackling, R. J. Sloan and M. K. Brown. 1988. Organochlorine and mercury residues in wild mink and otter:comparison with fish. Environmental Toxicology and Chemistry 7: 363-374.

Forbes, G. 2001. University of New Brunswick. Fredricton, New Brunswick. Personal communication with Conrad Reining,October 2001.

Foreman, D., K. Daly, B. Dugelby, R. Hanson, R. Howard, J. Humphrey, L. Klyza Linck, R. List and K. Vacariu. 2000. SkyIslands Wildlands Network Conservation Plan. Tucson, AZ: The Wildlands Project.

Foster, D. 2001. Wild Earth Interview. Wild Earth 11(1): 33-39.

Free World Data Foundation. 1999. Dam data for the U.S. Boston, MA: Free Software Foundation, Inc.

Fritts, S. H. 1983. Record dispersal by a wolf from Minnesota. Journal of Mammalogy 64: 166-167.

Fuller, T. K. 1989. Population dynamics of wolves in north-central Minnesota. Wildlife Monographs 105: 1-41.

Fuller, T. K., W. E. Berg, G. L. Radde, M. S. Lenarz and G. B. Joselyn. 1992. A history and current estimate of wolfdistribution. Wildlife Society Bulletin 20: 42-55.

Gawler, S.C. 1991. Natural Landscapes of Maine: A Classification of Ecosystems and Natural Communities. Augusta, ME:Maine Natural Areas Program, Department of Conservation.

Gawler, S.C. 2001. Natural Landscapes of Maine: A Classification of Vegetated Natural Communities and Ecosystems.Augusta, ME: Maine Natural Areas Program, Department of Conservation.

Gawler, S. C., J. J. Albright, P. D. Vickery and F. Smith. 1996. Biological diversity in Maine: An assessment of status andtrends in the terrestrial and freshwater landscape. Report prepared for the Maine Forest Biodiversity Project. Augusta,ME: Maine Natural Areas Program, Department of Conservation.

Godin, A. J. 1983. Wild Mammals of New England. Chester, CT: The Globe Pequot Press.

Goldburg, R. J., M. S. Elliot and R. L. Naylor. 2001. Marine Aquaculture in the United States. Arlington, Virginia: Pew OceansCommission.


Graham, R. T., R. T. Reynolds, M. H. Reiser, R. L. Bassett and D. A. Boyce. 1994. Sustaining forest habitat for the northerngoshawk: a question of scale. Studies in Avian Biology 16: 12-17.

Griffith, D. M. and C. L. Alerich. 1996. Forest statistics for Maine, 1995. Resource Bulletin NE-135. Radnor, PA: USDAForest Service, Northeastern Forest Experiment Station.

Grzybowski, J. A. and S. W. Eaton. 1976. Prey items of Goshawks in southwestern New York. Wilson Bulletin 88: 669-670.

Gurd, D. B., T. D. Nudds and D. H. Rivard. 2001. Conservation of mammals in eastern North American wildlife reserves: howsmall is too small? Conservation Biology 15(5): 1355-1363.

Hagen, J. 2001. MOSAIC update: Mid-summer news from Manomet, Maine. Brunswick, Maine: Manomet Forest Ecology andConservation Program. 10 July 2001.

Hakola, J. W. 1981. Legacy of a Lifetime: The Story of Baxter State Park. Woolwich, ME: TBW Books.

Hall, E. R. 1981. The Mammals of North America. Second edition. New York, NY: John Wiley and Sons.

Hands, H. M., R. D. Drobney and M. R. Ryan. 1989. Status of the Red-shouldered Hawk in the northcentral United States.Missouri Cooperative Fish and Wildlife Research Unit Report.

Hansen, L. P. and T. P. Quinn. 1998. The marine phase of the Atlantic salmon (Salmo salar) life cycle, with comparisons toPacific salmon. Canadian Journal of Fisheries and Aquatic Sciences. 55(Suppl. 1): 104-118.

Hargis, C. D., J. A. Bissonette and D. L. Turner. 1999. The influence of forest fragmentation and landscape pattern onAmerican martens. Journal of Applied Ecology 36: 157-172.

Harris, L. D. 1984. The Fragmented Forest: Island Biogeography Theory and the Preservation of Biotic Diversity. Chicago, IL:University of Chicago Press.

Harrison, D. J. and T. G. Chapin. 1997. An assessment of potential habitat for eastern timber wolves in the northeast UnitedStates and connectivity with occupied habitat in southeastern Canada. Wildlife Conservation Society Working Paper No.7.

_____. 1998. Extent and connectivity of habitat for wolves in eastern North America. Wildlife Society Bulletin 26(4): 767-775.

Havlick, D. 1998. Tooth and claw: Ecological effects of roads on predators. Missoula, MT: The Road-RIPorter (July/August):8-11.

Hayward, G. D. and R. E. Escano. 1989. Goshawk nest-site characteristics in western Montana and northern Idaho. Condor 91:476-479.

Henke, S. E. and F. C. Bryant. 1999. Effects of coyote removal on the faunal community in Western Texas. Journal of WildlifeManagement 63(4): 1066-1081.

Hitt, J. 2001. One nation, under Ted. Outside XXVI (12).

Hodgman, T. P., D. J. Harrison, D. D. Katnik and K. D. Elowe. 1994. Survival in an intensively trapped marten population inMaine. Journal of Wildlife Management 58: 593-600.

Irland, L. C. 1996. Land, timber, and recreation in Maine’s northwoods. Maine Agricultural and Forest Experiment Station,Misc. Pub. 730, Orono, ME: University of Maine.

_____. 1999. The Northeast’s changing forest. Hanover, NH: University Press of New England.

Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, B. J. Bourque, R. H. Bradbury, R. Cooke, J.Erlandson, J. A. Estes, T. P. Hughes, S. Kidwell, C. B. Lange, H. S. Lenihan, J. M. Pandolfi, C. H. Peterson, R. S.Steneck, M. J. Tegner and R. R. Warner. 2001. Historical overfishing and the recent collapse of coastal ecosystems.Science 293:629-638.

Johnsgard, P. A. 1987. Diving Birds of North America. Lincoln, NB: University of Nebraska Press.

Josselyn, J. 1672. New England’s Rarities Discovered. Reprint. Chester, CT: Applewood Books, n.d.

Kahl, S. 1996. A review of the effects of forest practices on water quality in Maine. Report to the Maine Department ofEnvironmental Protection. Augusta, Maine: Bureau of Land and Water Quality Control. October 1996.


Kennedy, P. L. 1997. The Northern Goshawk: Is there evidence of a population decline? Journal Raptor Research 31(2): 95-106.

Keys Jr., J., C. Carpenter, G. Hooks, F. Koenig, W. H. McKlab, W. Russell and M. L. Smith. 1995. Ecological units of theeastern United States, first approximation (map and booklet of map unit tables). Atlanta, GA: USDA, Forest Service.

Kimmel, J. T. and R. H. Yahner. 1994. The northern goshawk in Pennsylvania: habitat use, survey protocols, and status. Finalreport. University Park, PA: School of Forest Resources, Pennsylvania State University.

Kitchener, A. 1991. The Natural History of the Wild Cats. Ithaca, NY: Comstock Publishing Associates, Cornell UniversityPress.

Koehler, G. M. 1990. Population and habitat characteristics of lynx and snowshoe hares in north central Washington. CanadianJournal of Zoology 68: 845-851.

Koehler, G. M. and M. G. Hornocker. 1991. Seasonal resource use among mountain lions, bobcats and coyotes. Journal ofMammalogy 72: 391-396.

Kotliar, N. B. 2000. Application of the new keystone-species concept to prairie dogs: how well does it work? ConservationBiology 14(6): 1715-1721

Krohn, W. B. and R. B. Boone. 1999. Maine gap analysis: preliminary results for American marten. Maine Cooperative Fishand Wildlife Research Unit website, Retrieved February 7, 1999 from< http://wlm13.umenfa.maine.edu/progs/unit/gap/layers/vert/mamms/MAAM.html >

Krohn, W. K., R. B. Boone, S. A. Sader, J. A. Hepinstall, S. M. Shaefer and S. L. Painton. 1998. Maine gap analysis - ageographic analysis of biodiversity. Final contract report to the USGS Biological Resource Division, Gap AnalysisProgram, Moscow, ID.

Krohn, W. B., R. B. Boone and S. L. Painton. 1999. Quantitative delineation and characterization of hierarchical biophysicalregions of Maine. Northeastern Naturalist 6: 139-164.

Kruuk, H. 1995. Wild Otters: Predation and Populations. Oxford, United Kingdom: Oxford University Press.

Lambeck, R. J. 1997. Focal species: A multi-species umbrella for conservation. Conservation Biology 11(4): 849-856.

Land and Water Associates and Market Decisions. 1994. A Summary of the Commission’s Current Land Use Policies and their NetEffects After 20 Years of Development in Maine’s Unorganized Territories. Augusta, ME: Use Regulation Commission. MaineDepartment of Conservation.

Land Use Regulation Commission. 1997. Comprehensive land use plan. Augusta, ME: Maine Department of Conservation.

Lansky, A. M. 1992. Beyond the beauty strip: Saving what’s left of our forests. Gardiner, ME: Tilbury House.

_____. 1999. “Sustainable” forestry: getting there from here. An interview with Mitch Lansky. Orion Afield 3(4): 27-29.

Larivière, S. and L. R. Walton. 1998. Lontra canadensis. Mammalian Species 587: 1-8.

Leopold, A. 1949. A Sand County Almanac. New York: Oxford University Press.

_____. 1972. “The Round River—A Parable.” In Round River: From the Journals of Aldo Leopold, ed. L.B. Leopold. New York,NY: Oxford University Press.

Lewis, D. 2001. West Branch Project: An inefficient use of conservation funds. The Northern Forest Forum 8(6): 3.

Long, R. and P. MacKay. 2001. Focal species selection for Maine. Richmond, VT: The Wildlands Project. Unpublished report.13 pp.

Lorimer, C. G. 1977. The presettlement forest and natural disturbance cycle of northeastern Maine. Ecology 58: 139-148.

Maehr. D. S. 1990. The Florida panther and private lands. Conservation Biology 4(2): 167-170.

Maine Cooperative Fish and Wildlife Research Unit. 1998. Maine Gap Analysis Vertebrate Data. Part I: Distribution, HabitatRelations, and Status of Amphibians, Reptiles, and Mammals in Maine. Part II: Distribution, Habitat Relations, andStatus of Breeding Birds in Maine. Orono, ME: University of Maine, Department of Wildlife Ecology.

Maine Department of Conservation. 2001. Maine Natural Areas Program. Retrieved October, 2001 from< http://www.state.me.us/doc/nrimc/mnap/home.htm >.


Maine Department of Inland Fisheries and Wildlife (MDIFW). 1998. Wolves in Maine: a status review. February 1998.

Maine Environmental Priorities Council. 1999. An assessment of the Quality of Maine’s Environment 1998. Augusta, ME:Maine Department of Environmental Protection.

Maine Forest Service. 1999. The state of the forest and recommendations for forest sustainability standards. Final report to theJoint Standing Committee of the 119th Legislature on Agriculture, Conservation and Forestry. Augusta, ME: MaineDepartment of Conservation.

_____. 2001. The 2001 Biennial Report on the State of the Forest and Progress on Forest Sustainability Standards. Report tothe Joint Standing Committee of the 120th Legislature on Agriculture, Conservation, and Forestry. August, ME: MaineDepartment of Conservation.

Maine Invasive Aquatic Species Work Group. 2001. Invasive Aquatic Species Report. Report to the Joint LegislativeCommittees on Natural Resources and Fish & Wildlife. First Session of the 120th Legislature.

Maine Office of Geographic Information Systems. 1999. Conservation and public lands database for Maine. Augusta, ME:MOGIS.

Maine State Planning Office. 1997. The cost of sprawl. Augusta, ME: Maine State Planning Office.

Malakoff, D. 1998. Atlantic salmon spawn fight over species protection. Science 279(5352): 800.

Margules, C. R., A. O. Nicholls and R. L. Pressey. 1988. Selecting networks of reserves to maximize biological diversity.Biological Conservation 43: 63-76.

Mather, M. E., D. L. Parrish, C. L. Folt and R. M. DeGraaf. 1998. Integrating across scales: effectively applying science for thesuccessful conservation of Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences 55(Suppl. 1): 1-8.

McCollough, M. 1987. A Management and Research Plan for the Reintroduction of Woodland Caribou to Maine. Orono, ME:Maine Caribou Reintroduction Project.

_____. 1991. Final Report Maine Caribou Project 1986-1990. Orono, ME: Maine Caribou Project.

McCormick, S. D., L. P. Hansen, T. P. Quinn and R. L. Saunders. 1998. Movement, migration, and smolting of Atlanticsalmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences 55(Suppl. 1): 77-92.

McGinn, A. P. 1999. Atlantic salmon face perilous waters. World Watch 12(1): 10-11.

McGrory Klyza, C. 2001. Wildlands in the Northeastern United States. International Journal of Wilderness 7(2): 8-10.

McIntyre, J. W. 1983. Nurseries: a consideration of habitat requirement during the early chick-rearing period in commonloons. Journal of Field Ornithology 54: 247-53.

_____. 1988a. The Common Loon: Spirit of Northern Lakes. Minneapolis, MN: University of Minnesota Press.

_____. 1988b. The rise and fall of a loon lake. In 1987 Conference on Common Loon Research and Management, ed. P. I. V. Strong.Meredith, NH: North American Loon Fund.

McLaren, B. E. and R. O. Peterson. 1994. Wolves, moose and tree rings on Isle Royale. Science 266: 1555-1558.

McMahon, J. S. 1990. The biophysical regions of Maine: Patterns in the landscape and vegetation. M.S. Thesis. Orono, ME:University of Maine.

_____. 1998. An ecological reserves system inventory: Potential ecological reserves on Maine’s existing public and privateconservation lands. Rockland, ME: Maine Forest Biodiversity Project.

McShea, W. J. and J. H. Rappole. 1997. Herbivores and the ecology of forest understory birds. In The Science of Overabundance:Deer Ecology and Population Management, ed. W. J. McShea, H. B. Underwood, and J. H. Rappole. Washington, D.C.:Smithsonian Institution Press.

McShea, W. J., H. B. Underwood and J. H. Rappole. 1997. The Science of Overabundance: Deer Ecology and Population Management.Washington, D.C.: Smithsonian Institution Press.

Mech, L. D., S. H. Fritts, G. L. Radde and W. J. Paul. 1988. Wolf distribution and road density in Minnesota. Wildlife SocietyBulletin 16: 85-87.


Melquist, W. E. and A. E. Dronkert. 1987. River otter. In Wild Furbearer Management and Conservation in North America, ed. M.Novak, J.A. Baker, M. E. Obbard and B. Malloch. North Bay, Ontario: Ontario Trappers Association.

Miller, B., G. Ceballos, and R. Reading. 1994. The prairie dog and biotic diversity. Conservation Biology 8: 677–681.

Miller, B., R. Reading, J. Strittholt, C. Carroll, R. Noss, M. Soulé, O. Sanchez, J. Terborgh, D. Brightsmith, T. Cheeseman andD. Foreman. 1998. Focal species in the design of nature reserve networks. Wild Earth 8(4): 81-92.

Mills, L. S., M. E. Soulé and D. F. Doak. 1993. The history and current status of the keystone species concept. Bioscience 43:219-224.

Mladenoff, D. J., T. A. Sickley, R. G. Haight and A. P. Wydeven. 1995. A regional landscape analysis and prediction offavorable gray wolf habitat in the northern Great Lakes region. Conservation Biology 9: 279-294.

Mowat, G., B. G. Slough and S. Boutin. 1996. Lynx recruitment during a snowshoe hare population peak and decline insouthwest Yukon. Journal of Wildlife Management 60(2): 441-452.

Naiman, R. J., J. M. Melillo and J. E. Hobbie. 1986. Ecosystem alteration of boreal forest streams by beaver (Castorcanadensis). Ecology 67: 1254-1269.

Naiman, R. J., C. A. Johnston and J. C. Kelley. 1988. Alteration of North American streams by beaver. BioScience 38: 753-762.

Naiman, R. J., H. Decamps and M. Pollock. 1993. The role of riparian corridors in maintaining regional biodiversity. EcologicalApplications 3: 209-212.

National Invasive Species Council. 2000. United States Invasive Species Draft Management Plan: Preparing for the Future.Washington D.C.

National Marine Fisheries Service. 1999. Atlantic salmon: candidate species. November 17, 1999. Retrieved February 7, 2000,from < http://www.nmfs.gov/prot_res/fish/atl_salm.html>.

National Wildlife Federation. 1995. Getting the cut out: raw log exports in the northern forest region. Montpelier, VT: NortheastNatural Resource Center.

National Wildlife Federation. 2000. Clean the Rain, Clean the Lakes II: Mercury in Rain is Contaminating New England’sWaterways. Washington, D.C.: National Wildlife Federation.

Natural Resources Council of Maine. 2000. One-year anniversary of Edwards dam removal celebrated as national success story.Press release. June 29, 2000.

NatureServe Explorer. 2002. An online encyclopedia of life [web application]. Version 1.6. Arlington, Virginia, USA:NatureServe. Retrieved January 3, 2002 from < http://www.natureserve.org/explorer >.

Naylor, R. L., S. L. Williams and D. R. Strong. 2001. Aquaculture – A Gateway for Exotic Species. Science 294: 1655-1656.

New England Forestry Foundation. 2001. Conservation easement on land of six rivers limited partnership, et al. Groton, MA:New England Forestry Foundation.

Nelson, R. 1997. Heart and blood: Living with deer in America. New York: Knopf, Inc.

Nichols, W. F., K. T. Killingbeck and P. V. August. 1998. The influence of geomorphological heterogeneity on biodiversity.II. A landscape perspective. Conservation Biology 12(2): 371-379.

Nielson, J. L. 1998. Population genetics and the conservation and management of Atlantic Salmon (Salmo salar). CanadianJournal of Fisheries and Aquatic Sciences. 55(Suppl. 1): 145-152.

Northern Forest Alliance. 2002. Shaping the northern forest economy: strategies for a sustainable future. Montpelier, VT: NorthernForest Alliance.

Noss, R. F. 1983. A regional landscape approach to maintain diversity. BioScience 33(11): 700-707.

_____. 1987a. From plant communities to landscapes in conservation inventories: a look at The Nature Conservancy (USA).Biological Conservation 41: 11-37.

_____. 1987b. Protecting natural areas in fragmented landscapes. Natural Areas Journal 7: 2-13.

_____. 1991. From endangered species to biodiversity. Pages 227-246 in Balancing on the Brink of Extinction: the EndangeredSpecies Act and Lessons for the Future, ed. K.A. Kohm. Washington, D.C.: Island Press.


_____. 1992. The Wildlands Project: Land conservation strategy. Wild Earth (special issue): 10-25.

_____. 1993. Wildlife corridors. Pages 43-68 in Ecology of Greenways, ed. D. S. Smith and P. C. Hellmund. Minneapolis, MN:University of Minnesota Press.

_____. 1995. Maintaining ecological integrity in representative reserve networks. World Wildlife Fund Canada/WorldWildlife Fund-U.S. Discussion Paper. Toronto, Ontario: World Wildlife Fund Canada.

_____. 1996. The Ecological Effects of Roads. Missoula, MT: Road-Ripper's Handbook, ROAD-RIP.

_____. 2001. Beyond Kyoto: Forest management in a time of rapid climate change. Conservation Biology 15: 578-590.

Noss, R. F. and L .D. Harris. 1986. Nodes, networks, and MUMs: Preserving diversity at all scales. Environmental Management10: 299-309.

Noss, R. F. and A. Y. Cooperrider. 1994. Saving Nature’s Legacy: Protecting and Restoring Biodiversity. Washington, D.C.: IslandPress.

Noss R. F., M. A. O’Connell and D. D. Murphy. 1997. The Science of Conservation Planning: Habitat Conservation under theEndangered Species Act. Washington D.C.: Island Press.

Noss, R. F., J. R. Strittholt, K. Vance-Borland, C. Carroll and P. Frost. 1999. A conservation plan for the Klamath-Siskiyouecoregion. Natural Areas Journal 19: 392–411.

Noss, R. F., G. Wuerthner, K. Vance-Borland and C. Carroll. 2001. A biological conservation assessment for the GreaterYellowstone Ecosystem: Report to the Greater Yellowstone Coalition and The Nature Conservancy.

Noss, R. F., C. Carroll, K. Vance-Borland and G. Wuerthner. 2002. A Multi-Criteria Assessment of the Irreplaceability andVulnerability of Sites in the Greater Yellowstone Ecosystem. Upcoming in Aug 2002 issue of Conservation Biology.

O'Donoghue, M., S. Boutin, C. J. Krebs, G. Zuleta, D. L. Murray and E. J. Hofer. 1998. Functional responses of coyotes andlynx to the snowshoe hare cycle. Ecology 79(4): 1193-1210.

Okoniewski, J. C. and W. B. Stone. 1987. Some observations on the morbidity and mortality of, and environmentalcontaminants in, common loons (Gavia immer) in New York, 1972-1986. New York State Department EnvironmentalConservation, Delmar, New York. Unpublished report.

Olson, D. M., E. Dinerstein, E. D. Wikramanaya, N. D. Burgess, G. V. N. Powell, E. C. Underwood, J. A. D’amico, I. I. Toua,H. E. Strand, J. C. Morrison, C. J. Loucks, T. F. Allnutt, T. H. Ricketts, Y. Kura, J. F. Lamoreux, W. W. Wettengel, P.Hedao, and K. R. Kassem. 2001. Terrestrial ecoregions of the world: a new map of life on earth. Bioscience 51(11); 933-938.

Osborn, W. C. 1974. The Paper Plantation: Ralph Nader’s Study Group Report on the Pulp and Paper Industry in Maine. New York,NY: Grossman Publishers.

Paine, R.T. 1969. A note on trophic complexity and community stability. American Naturalist 1103: 91-93.

Palmer, R. S. 1988a. Handbook of North American Birds. Volume 5. New Haven, CT: Yale University Press.

_____. 1988b. Red-shouldered hawk, Asturina lineata. In Handbook of North American Birds. Vol. 4: Diurnal Raptors (Part 1), ed.R. S. Palmer. New Haven, CT: Yale University Press.

Paquet, P. C., J. R. Strittholt and N. L. Staus. 1999. Wolf reintroduction feasibility in the Adirondack Park. Corvalis, OR:Conservation Biology Institute.

Paquet, P. C., J. Wierzchowski and C. Callaghan. 1997. Assessing reserve designs using a predictive habitat suitability andmovement model. Oral presentation, 1997 Annual Meeting of the Society for Conservation Biology, Victoria, Canada.

Parker, G. R., J. W. Maxwell, L. D. Morton and G. E. Smith. 1983. The ecology of the lynx (Lynx canadensis) on Cape BretonIsland. Canadian Journal of Zoology 61: 770-786.

Parrish, D. L., R. J. Behnke, S. R. Gephard, S. D. McCormick and G. H. Reeves. 1998. Why aren't there more Atlantic salmon(Salmo salar)? Canadian Journal of Fisheries and Aquatic Sciences 55(Suppl. 1): 281-287.

Payer, D. C. and D. J. Harrison. 1999. Effects of forest structure on spatial distribution of American marten. Technical bulletinNo. 787. Raleigh, North Carolina: National Council of the Paper Industry for Air and Stream Improvement.


_____. 2000. Structural differences between forests regenerating following spruce budworm defoliation and clear-cutharvesting: implications for marten. Canadian Journal of Forestry Research 30: 1965-1972.

Pidot, J. 2001. Comments on Draft Conservation Easement for West Branch Project. Memo to Ralph Knoll, Director ofPlanning and Land Acquisition, Bureau of Parks and Lands and Mark DesMeules, Director, Land for Maine’s FutureProgram. 3 August 2001. Augusta, ME.

Pierce, B. M., V. C. Bleich, J. D. Wehausen and R. T. Bowyer. 1999. Migratory patterns of mountain lions: implications forsocial regulation and conservation. Journal of Mammalogy 80(3): 986-992.

Poff, N. L. and A. D. Huryn. 1998. Multi-scale determinants of secondary production in Atlantic salmon (Salmo salar) streams.Canadian Journal of Fisheries and Aquatic Sciences 55(Suppl. 1): 201-217.

Polechla, P. 1990. Action plan for North American otters. In Otters: an Action Plan for Their Conservation, ed. P. Foster-Turley, S.Macdonald, and C. Mason. Gland, Switzerland: IUCN/SSC Otter Specialist Group, IUCN.

Poole, K. G. 1994. Characteristics of an unharvested lynx population during a snowshoe hare decline. Journal of WildlifeManagement 58: 608-618.

_____. 1995. Spatial organization of a lynx population. Canadian Journal of Zoology 73: 632-641.

Poole, K. G., L. A. Wakelyn and P. N. Nicklen. 1996. Habitat selection by lynx in the Northwest Territories. Canadian Journalof Zoology 74: 845-850.

Power, M. E. 1992. Top-down or bottom-up forces in food webs: Do plants have primacy? Ecology 73:733-746.

Power, M. E., D. Tilman, J. E. Estes, M. A. Menge, W. J. Bond, L. S. Mills, G. Daily, J. C. Castilla, J. Lubchenco, and R. T.Paine. 1996. Challenges in the quest for keystones. BioScience 46: 609-620.

Power, T. M. 2001. The Economic Impact of the Proposed Maine Woods National Park and Preserve. Hallowell, Maine:RESTORE: The North Woods.

Publicover, D. Comments on January 2002 draft of Maine Wildlands Network Conservation Plan. Unpublished document.Richmond, VT: The Wildlands Project.

Ray, J. C. 2000. Mesocarnivores of northeastern North America: status and conservation issues. Wildlife Conservation SocietyWorking Paper No. 15. Bronx, NY: Wildlife Conservation Society.

_____. 2001. Wildlife Conservation Society, University of Toronto. Personal communication with Barbara Dugelby,December, 2001.

_____. 2002. Comments on January 2002 draft of Maine Wildlands Network Conservation Plan. Unpublished document.Richmond, VT: The Wildlands Project.

Reynolds, R. T., S. M. Joy and D. G. Leslie. 1994. Nest productivity, fidelity, and spacing of northern goshawks in Arizona.Studies in Avian Biology 16:106-113.

Reynolds, R. T., R. T. Graham, M. H. Reiser, R. L. Bassett, P. L. Kennedy, D. A. Boyce, Jr., G. Goodwin, R. Smith and E. L.Fisher. 1992. Management recommendations for the northern goshawk in the southwestern United States. GeneralTechnical Report RM-17. USDA Forest Service. Fort Collins, CO: Rocky Mountain Forest and Range ExperimentStation.

Rimmer, C. C. 1992. Common loon, Gavia immer. In Migratory Nongame Birds of Management Concern in the Northeast, ed. K. J.Schneider and D. M. Pence. Newton Corner, MA: U.S. Fish and Wildlife Service.

Ripple, W. J. and E. J. Larsen. 2000. Historic aspen recruitment, elk, and wolves in northern Yellowstone National Park, USA.Biological Conservation 95: 361-370.

Robbins, S. D., Jr. 1991. Wisconsin Birdlife: Population and Distribution Past and Present. Madison, WI: University of WisconsinPress.

Rohner, C. and F. I. Doyle. 1992. Food-stressed great horned owl kills adult goshawk: exceptional observation or communityprocess? Journal of Raptor Research 26: 261-263.

Seidensticker, J. C., M. G. Hornocker, W. V. Wiles and J. P. Messick. 1973. Mountain lion social organization in the IdahoPrimitive Area. Wildlife Monographs 35: 1-60.


Seton, E. T. 1929. Lives of the Game Animals, vol. 2. New York: Doubleday Doran & Co., Inc.

Simberloff, D. S. 1990. Reconstructing the ambiguous: Can island ecosystems be restored? In Ecological restoration of NewZealand Islands, ed. D. R Towns, C. H. Daughterty, and I. A. E. Atkinson. Wellington, New Zealand: New ZealandDepartment of Conservation.

Simberloff, D. S. 1998. Flagships, umbrellas, and keystones: Is single-species management passé in the landscape era? BiologicalConservation 83 247-257.

Simberloff, D. S., D. Doak, M. Groom, S. Trombulak, A. Dobson, S. Gatewood, M. E. Soulé, M. Gilpin, C. Martínez del Rioand L. Mills. 1999. Regional and continental restoration. In Continental Conservation: Scientific Foundations of RegionalReserve Networks, ed. M. E. Soulé and J. Terborgh. Washington D.C. and Covelo, CA: Island Press.

Snyder, J. E. and J. A. Bisonette. 1987. Marten use of clear-cuttings and residual forest stands in western Newfoundland.Canadian Journal of Zoology 65: 169-174.

Soulé, M. E. (ed.) 1987. Viable Populations for Conservation. Cambridge, UK: Cambridge University Press.

_____. 1995. An unflinching vision: networks of people defending networks of lands. In Nature Conservation 4: The Role ofNetworks ed. D.A. Saunders, J. L. Craig and E. M. Mattiske. Sydney: Surrey Beatty and Sons.

Soulé, M. E. and R. F. Noss. 1998. Rewilding and biodiversity: Complementary goals for continental conservation. Wild Earth(8)3: 18-28.

Soulé, M. E. and Sanjayan, M. A. 1998. Conservation targets: Do they help? Science 279: 2060-2061.

Soulé, M. E. and J. Terborgh (eds.) 1999. Continental Conservation: Scientific Foundations of Regional Reserve Networks. WashingtonD.C. and Covelo, CA: Island Press.

Soulé, M.E., J. A. Estes, J. Berger and C. Martinez del Rio. Submitted. Recovery goals for ecologically effective numbers ofendangered keystone species. (Submitted to Conservation Biology).

Squires, J. R. and R. T. Reynolds. 1997. Northern Goshawk, Accipiter gentilis. In The Birds of North America, eds. A. Poole andF. Gill. Academy of Natural Sciences, Philadelphia, and American Ornithologists' Union, Washington, D.C.

Squires, J. R. and L. F. Ruggiero. 1996. Nest-site preference of northern goshawks in southcentral Wyoming. Journal ofWildlife Management 60: 170-177.

St. Pierre, J. A. 1976. Domestic neoimperialism and the Maine paper industry. Unpublished manuscript.

_____. 1999a. Proposal for a new wilderness area resource category for the Maine BPL public lands. Unpublished paperpresented to the Maine Bureau of Parks & Lands Integrated Resource Policy Advisory Committee. Augusta, ME:RESTORE: The North Woods.

_____. 1999b. Major land ownership in the State of Maine. Augusta, Maine.

_____. In prep. The people’s forest no one knows. Augusta, Maine.

Stevens, W. K. 1999. As a species vanishes, no one can say why. New York Times. September 14, 1999.

Stocek, R. F. 1995. The cougar, Felis concolor, in the Maritime Provinces. Canadian Field-Naturalist 109(1): 19-22.

Stoms, D. A. 1999. Worked Example for the Northern Sierra Nevada Ecoregion. In: Sites: An Analytical Toolbox forEcoregional Conservation Planning, available at www.biogeog.ucsb.edu/projects/tnc/toolbox.html .

Strong, P. I. V., J. A. Bissonette and J. S. Fair. 1987. Reuse of nesting and nursery areas by common loons. Journal of WildlifeManagement 51: 123-127

Taverna, K., J. E. Halbert and D. M. Hines. 1999. Eastern cougar (Puma concolor couguar): Habitat suitability analysis for thecentral Appalachians. Unpublished report, Charlottesville, VA: Appalachian Restoration Campaign.

Terborgh, J. 1988. The big things that run the world—a sequel to E.O. Wilson. Conservation Biology 2: 402-403.

Terborgh, J., J. Estes, P. Paquet, K. Ralls, D. Boyd-Heger, B. Miller and R. Noss. 1999. Role of top carnivores in regulatingterrestrial ecosystems. In Continental Conservation: Design and Management Principles for Long-term, Regional ConservationNetworks, eds. M. E. Soulé and J. Terborgh. Covelo, CA and Washington, D.C.: Island Press.


Terborgh, J., L. Lopez, P. Nuñez V., M. Rao, G. Shahabuddin, G. Orihuela, M. Riveros, R. Ascanio, G. H. Adler, T. D.Lambert and L. Balbas. 2001. Ecological meltdown in predator-free forest fragments. Science 294: 1923-1926.

Théberge, J. B., M. T. Théberge and G. Forbes. 1996. What Algonquin Park wolf research has to instruct about recovery in thenortheastern United States. In Wolves of America Conference Proceedings. November 14-16, 1996. Albany, NY: Defenders ofWildlife.

Thompson, E. H. and E. R. Sorenson. 2000. Wetland, Woodland, Wildland: A Guide to the Natural Communities of Vermont.Hanover, NH and London, UK: University Press of New England.

Thompson, I. D. 1991. Could marten become the spotted owl of Canada? Forestry Chronicle 67: 136-140.

_____. 1994. Marten populations in uncut and logged boreal forests in Ontario. Journal of Wildlife Management 58: 272-280.

Thompson, I. D. and P. W. Colgan. 1994. Marten activity in uncut and logged boreal forests in Ontario. Journal of WildlifeManagement 58: 280-288.

Thoreau, H. D. 1864. The Maine Woods. Boston, MA: Ticknor & Fields.

Thurber, J. M., R. O. Peterson, T. R. Drummer and S. A. Thomasma. 1994. Gray wolf response to refuge boundaries and roadsin Alaska. Wildlife Society Bulletin 22: 61-68.

Titus, J. R. and L. W. VanDruff. 1981. Response of the common loon to recreational pressure in the Boundary Waters CanoeArea, northeastern Minnesota. Wildlife Monographs 79: 1-59.

Titus, K. and M. R. Fuller. 1990. Recent trends in counts of migrant hawks from northeastern North America. Journal ofWildlife Management 54: 463-470.

Titus, K. and J. A. Mosher. 1987. Selection of nest tree species by red-shouldered and broad-winged hawks in tow temperateforest regions. Journal of Field Ornithology 58: 274-283.

Toweill, D. E. and J. E. Tabor. 1982. River otter, Lutra canadensis. In Wild Mammals of North America: Biology, Management, andEconomics, eds. J. A. Chapman and G. A. Feldhamer. Baltimore, MD: Johns Hopkins University Press.

Trombulak, S. 1996. How to design an ecological reserve system. Richmond, VT: Wild Earth Special Paper #1. Richmond,VT: Cenozoic Society.

Trombulak, S. and C. A. Frissell. 2000. Review of ecological effects of roads on terrestrial and aquatic communities.Conservation Biology 14: 18-30.

Turbak, G. 1998. Seeking the missing lynx. National Wildlife 36: 18-25.

UNESCO. 1974. Task Force on Criteria and Guidelines for the Choice and Establishment of Biosphere Reserves. Man and theBiosphere Report No.22. Paris, France.

U.S. Department of Commerce. 2001. United States Census 2000. Retrieved December 6 2001, from< http://quickfacts.census.gov/qfd/states/23000.html >.

U.S. Fish and Wildlife Service. 1991. Eastern cougar (Felis concolor cougar) species account. Endangered and Threatened Speciesof the Southeastern United States. FWS Region 4.

_____. 1992. Recovery plan for the eastern timber wolf. Twin Cities, MN.

_____. 1994. Endangered and threatened wildlife and plants. CFR 17.11 & 17.12. Washington, D.C.: U.S. Department of theInterior, Fish and Wildlife Service.

_____. 1998. Listing actions. Endangered Species Bulletin 23(5): 3.

_____. 1999a. Biological report on the status of Atlantic Salmon. October 1999. Retrieved February 3, 2000, from< http://news.fws.gov/salmon/asalmon.html >.

_____. 1999b. Discovery of Canada lynx kittens proves that the wild cats reproduce in Maine. U. S. Fish and Wildlife ServicePress Release, June 28, 1999. Washington, D.C.: U. S. Fish and Wildlife Service.

_____. 2000a. Final endangered status for a distinct population segment of anadromous Atlantic salmon (Salmo salar) in theGulf of Maine. Washington, D.C.: Federal Register 65(223): 69459-69483.


_____. 2000b. Determination of threatened status for the contiguous U. S. distinct population segment of the Canada lynx andrelated rule. Washington, D.C.: Federal Register 65: 16051-16086

U. S. Department of Agriculture. 1997. Forest Statistics of the U. S. Washington, DC: U. S. Forest Service.

Van Dyke, F. G., R. H. Brocke and H. G. Shaw. 1986a. The use of road track counts as indices of mountain lion presence.Journal of Wildlife Management 50(1): 95-102.

Van Dyke, F. G., R. H. Brocke, H. G. Shaw, B. B. Ackerman, T. P. Hemker and F. G. Linzey. 1986b. Reactions of mountainlions to logging and human activity. Journal of Wildlife Management 50(1): 95-102.

Vermeer, K. 1973. Some aspects of the nesting requirements of Common Loons in Alberta. Wilson Bulletin 85: 429-435.

Vickery, J. B., R. W. Judd and S. McDonald. 1995. Maine Agriculture, 1783-1861. In: Maine: The Pine Tree State fromPrehistory to the Present, ed. R. W. Judd, E. A. Churchill and J. W. Eastman. Orono, ME: University of Maine Press.

Waller, D. M. and W. S. Alverson. 1997. The white-tailed deer: a keystone herbivore. Wildlife Society Bulletin 25: 217-226.

Walsh, D. 2001. Value of easements debated. Bangor Daily News. 20 October 2001.

Ward, P. D. 1997. The Call of Distant Mammoths: Why the Ice Age Mammals Disappeared. New York: Copernicus.

Ward, R. M. P. and C. J. Krebs. 1985. Behavioural responses of lynx to declining snowshoe hare abundance. Canadian Journalof Zoology 63: 2817-2824.

Webb, R. H. and H. G. Wilshire. (eds) 1983. Environmental Effects of Off-Road Vehicles. New York, NY: Springer-Verlag.

Whitaker, J. O. and W. J. Hamilton. 1998. Mammals of the Eastern United States. Ithaca, NY: Cornell University Press.

Wilcove, D. S., M. Bean and P. C. Lee. 1992. Fisheries management and biological diversity: Problems and opportunities.Transactions of the North American Wildlife and Natural Resources Conference 57: 373-383.

Wilcove, D. S., D. Rothstein, J. Dubow, A. Philips and E. Losos. 1998. Quantifying threats to imperiled species in the UnitedStates. BioScience 48(8): 607-615.

Wilson, E. O. 2001. Vanishing Point: On Bjorn Lomborg and extinction. Grist Magazine. 12 December 2001.

Wilson, P. J., S. Grewal, I. D. Lawford, J. N. M. Heal, A. G. Granacki, D. Pennock, J. B. Théberge, M. T. Théberge, D. R.Voigt, W. Waddell, R. E. Chambers, P. C. Paquet, G. Goulet, D. Cluff and B. N. White. 2000. DNA profiles of theeastern Canadian wolf and the red wolf provide evidence for a common evolutionary history independent of the gray wolf.Canadian Journal of Zoology 78: 2156-2166.

Wood, R. L. 1979. Management of breeding loon populations in New Hampshire. In Proceedings of the Second North AmericanConference on Common Loon Research and Management, ed. S. Sutcliffe. New York, NY: National Audubon Society.

Woodbridge, B. and P. J. Detrich. 1994. Territory occupancy and habitat patch size of northern goshawks in the southernCascades of California. Studies of Avian Biology 16: 83-87.

Wydeven, A P., T. K. Fuller, W. Weber and K. MacDonald. 1998. The potential for wolf recovery in the northeastern UnitedStates via dispersal from southeastern Canada. Wildlife Society Bulletin 1998, 26(4): 776-784.

Young, S. 2002. NSA salmon study rejects King’s stance. Bangor Daily News. 8 January 2002.

Young, S. P. and E. A. Goldman. 1946. The Puma, Mysterious American Cat. New York: Dover Publications, Inc.


APPENDIX 1 – Participants, Recommendations, and Key Points from March 1998 Maine WildlandsNetwork Workshop Participants

Emily Bateson Sweet Water Trust, MAJudy Bond Grassroots GIS, VTTatiana Brailovskaya Nereus Project, MEScott Dickerson Coastal Mountains Land Trust, MEBarbara Dugelby The Wildlands Project, TXKathleen Fitzgerald University of Vermont Field Naturalist Program, VTJeff Jaros-Su University of Maine, MECathy Johnson Maine Natural Resource Council, MEBill Krohn U.S. Fish and Wildlife Service, University of Maine, MEMitch Lansky Forestry consultant, MERobert Long Greater Laurentian Wildlands Project, VTPaula MacKay Greater Laurentian Wildlands Project, VTJanet McMahon The Nature Conservancy Maine Chapter, MESue Morse Keeping Track, VTJamie Sayen Northern Appalachian Restoration Project, NHNancy Smith Sweet Water Trust, MAMichael Soulé The Wildlands Project, COJym St Pierre RESTORE: The North Woods, MEDave Publicover Appalachian Mountain Club, NHLinda Welch U.S. Fish and Wildlife Service (Maine Office), ME

Recommendations and Key Points• Clarify goals and objectives. Determine balance between rewilding and representation. Definition of rewilding. How does

rewilding overlap with stated goals of TWP?• What is the relationship between carnivores (or other focal species) and wilderness? Should focal species be used in

designing the system? If so, how? Possibly in corridor design?• Text: consider audience and define terms (roadless, focal species, etc...)• Southern Maine needs to be addressed. Decide on approach (representation, smaller roadless areas) and informational needs.

How can we collaborate with other groups?• Wilderness removes opportunity for harassment, provides refugia. What are other ecological values of wilderness?• Design may be too opportunistic. Should be based on objectives and habitat needs.• Address recreational access.• Identify ecological insults and build design from there. Restoration should be discussed/incorporated.• Invertebrates (and other taxa) are missing from design.• Should successional stages be addressed (and evolutionary processes)?• Acknowledge and incorporate other projects, especially in Southern Maine.• Study the history of land-use in Maine to identify which issues and insults should be


APPENDIX 2 – Compatible Conservation EffortsThis section summarizes current initiatives to protect

large blocks or parcels of undeveloped land in Maine. Ittouches only on major initiatives. There are many othersmall projects, especially in southern Maine, beingspearheaded by the 70 or so local and regional land trusts.Municipal departments and state and federal agencies alsohave other conservation projects underway.

Allagash Wilderness Waterway

Created in 1966 by a statewide referendum, theAllagash Wilderness Waterway (AWW) was established to"develop the maximum wilderness character" of thewaterway. At that time, road building, logging, and twomajor hydroelectric dam projects threatened to destroy theAllagash. In 1970, the waterway was designated as aNational Wild and Scenic River, in response to anapplication from Maine Governor Kenneth Curtis. In April2000, a coalition of conservation groups, Citizens toProtect the Allagash (CPL) submitted to the Maine Bureauof Parks and Lands (BPL) a compendium of lands withinand near the Allagash Wilderness Waterway that theyrecommend for priority conservation. BPL promised toconsider the list in future negotiations for land swaps orpurchases, but it is unclear whether any discussions withlandowners for properties along the river are currentlyunderway.

In November 2000, a group of citizens andconservation groups announced that they were filing alawsuit to appeal a decision by Maine’s Land UseRegulation Commission (LURC), which would allow a newboat launch to be built near John’s Bridge betweenChurchill and Eagle Lakes on the Allagash WildernessWaterway (AWW). The Maine Department ofConservation is seeking to build the new boat launch nearJohn’s Bridge on the Waterway’s Churchill Lake despitethe fact that there are already 11 boat launches on theWaterway, including one on Churchill Lake. There are also40 boat launches on other lakes within 10 miles of theAWW(< http://www.maineenvironment.org/nwoods/allagash%20 lawsuit.htm >).

In January 2002, the Natural Resources Council ofMaine (NRCM) raised concerns about a proposal to build anew, permanent Henderson Brook Bridge, upstream fromthe current bridge. The new bridge would be a 200-foot-long, 15-foot-wide, three span bridge with two 48-foot-wide concrete wing-wall, gravel-filled, abutments, two 3-1/2-foot-wide concrete support tiers and steel girders.NRCM worries that the bridge would harm the wildernesscharacter of the Allagash Wilderness Waterway and thatpermitting the construction of a permanent, modern

structure would ignore the letter and intent of state andfederal laws to protect the Allagash Wilderness Waterwayby legitimizing a currently illegal bridge and authorizingits replacement. NRCM contends that this continues apattern by the state of legitimizing illegal activities alongthe Allagash, resulting in the degradation of its wildernesscharacter(< http://www.maineenvironment.org/nwoods/new_allagash _alert.htm >).

Appalachian Trail (AT)

In November 2000, following more than 16 years ofstalemate, the National Park Service announced anagreement to protect 3.2 miles of the Appalachian Trail(AT) across Saddleback Mountain—the longestunprotected segment of the AT. As part of this agreement,the State will acquire 600 acres on the southeast face of themountain. Further, a 100-400 foot buffer zone will beestablished along each side of the trail across the ridge ofSaddleback, and an easement pertaining to a 324-acre areanear the top of the mountain will be written to minimizeenvironmental impacts. Another 570 acres will be donatedby private landowners in exchange for $4 million fromfederal appropriations and the allocation of 5,000 acres fordevelopment.

Meanwhile, the Appalachian Trail Conference LandTrust (ATCLT), with help from the Trust for Public Land,has acquired 3,000 acres near the AT on Mt. Abraham inwestern Maine. ATCLT has an option to purchase anadditional 1,000 acres, and is launching a campaign toraise $700,000 to buy another 1,000 acres. They plan todeed the properties to the Maine Bureau of Parks andLands. ATCLT is also working with a group of individualsinterested in purchasing 7,500 acres on Mt. Abraham.

The Maine Appalachian Trail Club (MATC) hasexpressed its opposition to 28 wind turbines, each to tower390 feet above the surrounding ridge lines, that areproposed for the Redington Range and Black NubbleMountain, as close as three-quarters of a mile to theAppalachian Trail as it crosses the exposed SaddlebackRange north to Crocker Mountain. The Board of Governorsof the Appalachian Trail Conference voted unanimously tooppose the project in December 2001. The board cited thearea's "scenic beauty, remote character, rugged terrain, highmountain vistas and its undeveloped landscape" as reasonsfor its opposition.

Bald Mountain Pond/Friends of Bald Mountain Pond

Friends of Bald Mountain Pond, with support from theNorthern Forest Alliance, is pushing for state acquisition oflands surrounding Bald Mountain Pond and along the


Appalachian Trail in Bald Mountain Township. PlumCreek Corporation owns about 75% of the shoreline of thislake, described in "Maine's Finest Lakes" as “a relativelyisolated, undeveloped pond with outstanding fisheries,wildlife, scenic and shoreline character resources.”

Sappi Corporation sold the land to Plum Creek inOctober 1998. The lake provides habitat for the rareblueback trout. The remaining shoreline is publicly ownedand the AT passes by the northern shore of this lake. It isunclear whether there are currently any active negotiationstaking place in relation to this project (excerpted fromwww.powerlink.net/fen/dave.htm a n dwww.powerlink.net/fen/news.htm on 21 January 2002).

Ducktrap Coalition/Coastal Mountains Land Trust

Located midway between the rapidly developingCamden and Belfast areas, the Ducktrap River runs formore than nine miles through wetlands and forests that arestill undeveloped, quiet, and rich in wildlife and scenicvignettes. The pristine habitat for spawning and youngsalmon make it one of only eight rivers in the United Statesthat continue to support wild Atlantic salmon. To preservethis remarkable river, the Ducktrap Coalition, led byCoastal Mountains Land Trust, has launched a campaign toprotect the critical river corridor. The Coalition envisionsthat the lands beside the river will remain forever wild,permanently conserved, a legacy for future generations andfor the salmon, white pine, scarlet tanagers, black cherry,otters, and all the other wild species that depend on theriver corridor, riparian wetlands, and forested slopes. Torealize this vision, the coalition works on a strictlyvoluntary basis with river landowners to protect lands withconservation easements and parcel purchases. To date, morethan two-thirds of the river corridor has been conserved insome fashion ( http://www.coastalmountains.org/htm/duckt_coalition.ht m , accessed on 31 May 2002).

East Coast Greenways (ECG)

The goal of the East Coast Greenways (ECG) project isto create the nation's first long-distance, city-to-city,multi-modal transportation corridor for cyclists, hikers,and other non-motorized users. ECG will connect existingand planned trails that are locally owned and managed toform a continuous, safe, green route from Maine to Florida.The route will be at least 80 percent off-road, usingwaterfront esplanades, park paths, abandoned railroads,canal towpaths, and parkway corridors. It is unclear at thispoint whether the ECG will provide connectivity forwildlife.

Forest Ecology Network (FEN)

The purpose of the Forest Ecology Network (FEN),based in Augusta, Maine, is to protect the native forests of

Maine through public awareness, grassroots citizenactivism, and education. FEN promotes forest practicesthat respect the ecological and aesthetic integrity of theMaine Woods, and works to eliminate the clearcutting,overcutting, and herbicide spraying associated withindustrial forestry.

FEN believes that to protect and restore the forestenvironment while ensuring a sustainable wood supply,there is a need for a combination of ecological reserves, low-impact forestry, demand reduction/recycling, andalternative fibers for papermaking. FEN spearheaded acampaign to defeat the industry-sponsored “Compact forMaine's Forests,” which was rejected by Maine voters inNovember 1997. FEN then formed the North Woodscoalition with other environmental organizations in Mainefor the purpose of lobbying the state legislature to passmeaningful forest practices reform. The coalition agreed ona joint policy to ensure sustainable forestry. The plan wasrejected by the Maine State legislature in 1998.

FEN, in concert with other environmental groups, isalso currently campaigning for the creation of a MaineWoods National Park (excerpted from< http://www.powerlink.net/fen/default.htm >, accessed 31May 2002).

Forest Society of Maine (FSM)

The Forest Society of Maine (FSM) was established in1984 by large landowners and the Maine Coast HeritageTrust. In June 2000, the FSM officially announced itsleadership role in the Penobscot West Branch Project.Phase I would spend $4 million in federal funds (LWCFand Forest Legacy) and $4.5 million in state/private dollarsto purchase Big Spencer Mountain and a beauty strip along10 miles of the northeast shore of Moosehead Lake, pluseasements on 67,000 acres in the Upper West BranchPenobscot watershed. Phase II, which could bring the totalcost of the project to nearly $60 million, would establisheasements on the remainder of the nearly 650,000 acrespurchased in 1999 from Great Northern/Bowater byWagner Forest Management on behalf of McDonaldInvestments and other investors. FSM is working with theMaine congressional delegation, the governor’s office, and ahost of groups "to secure tens of millions of additionalfederal dollars in support of Phase II.” Although the WestBranch project is moving ahead, it has experienced a fairamount of controversy. For example, in August 2001, theassistant Attorney General criticized the draft easementlanguage, saying that the easement put too much emphasison forestry and not enough on conservation (Pidot 2001).FSM is involved in a number of other projects—primarilyeasement initiatives such as the $4 million Robbins'Nicatous project in eastern Maine. In that project FSMworked with the Maine Coast Heritage Trust, the Trust forPublic Land, and the state to complete a 20,000-acre


conservation easement and fee acquisition deal. See Landfor Maine’s Future, below, for more information on theN i c a t o u s p r o j e c t . ( E x c e r p t e d f r o mhttp://www.maineaudubon.org/resources/habitat/v17n3f2.h tml and http://www.mltn.org/trusts/FSM.HTM , bothaccessed 31 May 2002).

Friends of Baxter State Park (FOB)

Founded in 2000, Friends of Baxter State Park (FOB)is an independent citizen's group working to preserve,support and enhance the wilderness character of BaxterState Park, in the spirit of its founder, Governor PercivalBaxter. Its objectives are to articulate a philosophy ofwilderness and its importance to the human condition;encourage others to understand and embrace thisphilosophy by providing programs and educationalmaterials on Baxter State Park's unique resources, values,and wild character to park visitors, government decision-makers, the private sector, and the general public; bringpeople together to share opinions, ideas and convictions,and to work for the good of Baxter State Park; andrepresent the Park's resources and values through strong,consistent, and informed advocacy.

Friends of the Boundary Mountains/Conservation Proposal for PlumCreek Land

Friends of the Boundary Mountains, with support fromthe Northern Forest Alliance, is advancing "A ConservationProposal for Plum Creek Land" in the Boundary Mountainsregion. The proposed strategy includes easements toprevent new residential development and to maintainhiking and snowmobiling trails, as well as fee purchase ofhigh elevation and roadless areas south of an existing Statetract at Holeb. Purchases by media mogul John Malonehave added a sense of urgency. In the summer of 2000,Malone paid Plum Creek $10 million for 7,500 acres in theSpencer Lake watershed. Previously, in 1998, he hadbought 900 acres and a luxury sporting camp for $840,000in a bankruptcy sale, as well as 7,400 acres fromInternational Paper for $3,450,000.

Friends of the Down East Lakes

Friends of the Down East Lakes, with support from theNorthern Forest Alliance, has prepared A ConservationProposal for the Former Georgia-Pacific Lands in the GrandLakes Region of Maine. Abutting the former G-P lands is a2,200-acre parcel on Sysladobsis Lake, owned by HerbHaynes that was being marketed in fall 2000, by TrimbleRealty for $895,000.

Land for Maine's Future (LMF)

The Land for Maine’s Future (LMF) Program wasestablished in 1987 when Maine voters overwhelminglyapproved a $35 million bond for purchasing lands of

statewide significance for recreation and conservation. Thefund is managed by a board of eleven composed of 6 privatecitizens, appointed by the Governor and approved by theSenate, and five commissioners from the Department ofInland Fisheries and Wildlife, Department of Conservation,Department of Marine Resources, Department ofAgriculture, and the State Planning Office.

An additional $50 million ($10 million/year over fiveyears) in state bond funds were recently approved,extending the life and reach of the program. The new statemoney will be largely spent in southern Maine, withexpenditures to include water access and farmlandeasement purchases. It is also likely that some land will beacquired as beauty strips and small ecoreserves in thenorthern part of the state. LMF is “prohibited by statute toacquire land for which the primary use value has been orwill be commercially harvested or harvestable forest land.This does not prohibit the acquisition of conservationeasements on working forest lands which allow for timberproduction while securing public access and theconservation of other natural resource values(< http://www.state.me.us/spo/lmf/pdf/facts1.pdf >).” Thisprohibition clearly limits the ability of LMF to acquirelands, especially in the northern part of the state, for thepurpose of establishing core conservation areas.

In April 2000, $750,000 in funding from LMF and $3million in federal Forest Legacy dollars was paid toRobbins Lumber Company for 76 small islands and abeauty strip on 29 miles of shorefront (337 acres) onNicatous and West Lakes, and for easements on 20,295acres of managed timberlands. Other projects in 2000included the purchase of 100 acres of land adjacent toAroostook State Park. The land has been added to the statepark, bringing the park's total acreage to over 700 acres. Inearly 2000, LMF and the Lower Kennebec Regional LandTrust protected 96.5 acres, including 8.2 acres to be ownedby the City of Bath, permanently protecting three-quartersof a mile of pristine frontage on the Kennebec River andWhiskeag Creek known as Thorne Head(< http://www.state.me.us/spo/lmf/2000projects.htm >).

Little Spencer Mountain Wilderness Project

At least one conservation organization has proposedthat Mead Corporation lands in East Middlesex CanalGrant Township be acquired and returned to "wilderness."The township includes Little Spencer Mountain and mostof Spencer Pond and abuts the Big Spencer Mountain tractbeing acquired by the State as part of the West BranchProject (see above under Forest Society of Maine).

Maine Bureau of Parks & Lands (BPL)

The Maine Legislature recently passed a billauthorizing the Bureau of Parks and Lands (BPL) toestablish ecoreserves on a small portion of its lands. BPL,


which recently held hearings on its revised IntegratedResource Policy plan, has proposed setting up ecoreserveson some of the Maine Public Reserves Lands and StateParks. Rather than designating these lands as “wilderness,”BPL will probably establish a new category called“Backcountry Non-Motorized Areas.”

Maine Draft Climate Change Action Plan

Maine has recently created a draft climate changeaction plan. The plan misleadingly suggests that youngforests store carbon faster than old ones, and thereforesupports continued logging in the Northern Forest. At thesame time, there is increasing pressure on the U.S. to pushfor incentives for forest preservation, as opposed to logging,in the international global climate change protocols. Anumber of conservation organizations are working toidentify the Northern Forest as an ideal place to test aforest protection carbon credit marketplace.

Maine Low-Impact Forestry Project

The Maine Low-Impact Forestry Project (seehttp://hcpcme.org/home.html accessed 20 November2001) is a small, informal group of loggers, landowners,and foresters interested in identifying examples of excellentforestry and analyzing systematically how they work isworking towards this vision. The project is also helping toconnect landowners and practitioners of low impact forestrywith emerging markets for sustainably harvested forestproducts. The mission of the Low Impact Forestry Projectis to encourage:

• a long-term management perspective;• a view of the forest as an eco-system;• less destructive logging practices;• high value markets for products harvested using low

impact methods;• management for multiple objectives including social

and community values; and• productivity of the forest, broadly defined.

Maine Public Employees for Environmental Responsibility (PEER)

Public Employees for Environmental Responsibility(PEER) is an association of resource managers, scientists,biologists, law enforcement officials and other governmentprofessionals committed to upholding the public trustthrough responsible management of the nation’senvironment and natural resources. PEER advocatessustainable management of public resources, promotesenforcement of environmental protection laws, and seeks tobe a catalyst for supporting professional integrity andpromoting environmental ethics in government agencies.PEER maintains a Maine office in Millinocket. MainePEER recently published a booklet entitled “Losing

Paradise: The Allagash Wilderness Waterway UnderAttack” which criticizes the Maine Bureau of Parks andL a n d s ’ m a n a g e m e n t o f t h e A l l a g a s h(< www.peer.org/publications/wp_losing.html > accessed 18January 2002).

Maine Wilderness Watershed Trust (MWWT)

During the past decade, the Maine WildernessWatershed Trust (MWWT) has achieved various levels ofprotection on more than 10,000 acres in and adjacent toPierce Pond, northeast of Flagstaff Lake. They haveestablished ownership or easements on virtually all of theland in the watershed, and hope to purchase a 6,800-acreparcel from Plum Creek (to which parcels are currentlyattached Forest Legacy easements preventing developmentbut allowing industrial forestry). MWWT estimates it mayneed to raise close to $10 million for the acquisition.

Maine Woods National Park (MWNP)

RESTORE: The North Woods has been working since1994 on a proposed Maine Woods National Park andPreserve (NWNP). The proposed park and reserve wouldencompass about 3.2 million acres of land in northwesternMaine, in the general area between Moosehead Lake,Millinocket and the St. John watershed. This proposal issupported by more than 100 organizations and businesses.Further, polls have consistently shown that a majority ofMaine residents support the park concept. A number ofconservation-minded individuals have been exploringacquisition of lands within the proposed MWNP with theintention that they become part of the park in the future.In October 2001, RESTORE released an economic study ofthe proposed Maine Woods National Park. The reportindicates that, even under conservative scenarios, the parkwould be an economic asset for the state and northernMaine. Under the Maine Wildlands Network Design, aportion of the proposed park is designated as core habitat.

Manomet Center for Conservation Sciences

Manomet Center for Conservation Sciences is ascientific organization dedicated to conserving the diversityand abundance of wild species and natural resources bydeveloping systems that sustainably meet the needs of bothhumans and natural systems. Manomet works directly withcommunities, businesses, policy makers, and other partnersto build science-based environmental solutions thatintegrate economic and conservation objectives. Its homeoffice is located in Manomet, Massachusetts.

In Maine, Division of Conservation Forestry staff areworking on ways to integrate economic and environmentalgoals in forestry. Manomet strives to build links betweenscience, scientists, and the public in order to betterunderstand forestry and economic issues. The Maine officehas developed five program areas to help establish this link;


Sustainable Forestry Lecture Series , Sustainable ForestryLibrary, an Interactive Web Site , Technical Report Series ,and the Shifting Mosaic Project . (Excerpted from< http://www.manometorg/forestry/forest1.htm > 1 8January 2002.)

New England Forestry Foundation (NEFF)

In March 2001, the New England Forestry Foundation(NEFF) purchased the development rights on 762,192acres of forestland owned by Pingree Associates for$28,142, 316, or about $37/acre. This is the largestconservation easement deal in the United States to date.Although the purchase of these rights prevents residentialand other types of non-forestry development, it gives theNew England Forestry Foundation little say in how theforests are managed. Nor is public access to the landguaranteed (Curtis 2001).

Larger than the state of Rhode Island, and three and ahalf times the size of Baxter State Park, this conservationeasement encompasses some of the most spectacular naturalresources in Maine, including the Allagash Lakes and 16miles along the St. John River. The easement also includesportions of the Cupsuptic, Black, Little Black, St. John,Machias and Aroostook Rivers, as well as UpperRichardson, Kennebago, Munsungan, Loon,Caucomgomoc, Eagle, Chamberlain, Allagash, Mooseleukand Churchill Lakes.

Natural Resources Council of Maine (NRCM)

The Natural Resources Council of Maine (NRCM)operates a Maine North Woods Project, the goal of whichis to protect the ecological systems, wild and undevelopedcharacter, and sustainable timber base of Maine’s NorthWoods focusing on remote wildland regions of particularlyhigh ecological and recreational significance. Along withover 35 other groups in northern New England and NewYork, all of whom are members of the Northern ForestAlliance, NRCM has identified five wildland areas inMaine that are of highest priority for protection. They arethe Upper St. John River Valley, the Greater Baxter StatePark Area, the Western Mountains, the AndroscogginHeadwaters, and the Down East Lakes. NRCM is workingto increase public ownership of lands in the North Woods,limit development in the North Woods to those areas thatare adjacent to existing development and infrastructure,strengthen Maine’s laws and rules regarding forestmanagement to ensure that Maine’s forests are harvested ina sustainable and responsible manner, and protect thebiological diversity of the state through the creation of asystem of ecological reserves. (Excerpted from NRCM’swebsite< www.maineenvironment.org/nwoods/intro_nw.htm > on18 January 2002.)

Northern Forest Alliance (NFA)

The Northern Forest Alliance (NFA) is a regionalcoalition comprising many of the larger conservationorganizations working in the Northern Forest region. Thiscoalition is working closely with the state administrationto identify lands to protect and sources of money tofacilitate protection. Although NFA has identified 10 largeareas across the Northern Forest (including 5 such areas inMaine) that they propose should remain “wildlands,” theyhave not yet specified the characteristics of a wildland (e.g.whether snowmobiles or other ORV’s will be allowed,whether logging will continue).

The NFA has also been working closely with theFriends of Moosehead, a local committee of business ownersand residents. Together, they have presented "AConservation Proposal for Plum Creek Land" in theMoosehead region, which calls for a combination ofadditional fee purchases and no-development easements.For two years, Plum Creek has stated that it has beenevaluating which lands to keep and which to sell. In May2000, the Maine Bureau of Park & Lands said talks withPlum Creek were on hold until the company finished itsassessment.

North Woods Coalition

The North Woods Coalition, an alliance of Maine'senvironmental community, worked to pass a modest "FourPoint Plan" that would have ensured that large landownersdid not cut more trees than were growing, would havelimited clearcutting and provided for an audit of thelandowner's forestry activities. The plan was endorsed byeditorial boards and some state lawmakers but was soundlydefeated by voters in 1998.

North Woods Wilderness Trust (NWWT)

The North Woods Wilderness Trust (NWWT) is aMoosehead region land trust whose mission is to protectwild lands in the northern forest, and maintain them openand accessible to present and future generations. Theyassisted The Nature Conservancy (TNC) in the purchase of185,000 acres of woodlands along the upper St. John River(December 1999) and put together a willing seller and aconservation buyer to gain control over 5800 acres in twodifferent townships, one of which borders a large TNCho ld ing ( Ju l y 2000 ) . (Exc e rp t ed f r om< http://www.northwoodswildernesstrust.org/ > on 18January 2001.)

The Nature Conservancy (TNC)

In December 1998, The Nature Conservancy (TNC)purchased 185,000 acres for $35 million in the St. JohnRiver watershed in northwestern Maine. In July 2000,TNC swapped an undisclosed amount of that land to J.M.


Huber Corp for acreage along the St. John. TNC alsoagreed to let Huber manage its St. John lands. Hubersimultaneously donated an easement to TNC on 36,000acres of its land in the watershed around Depot Lake. InJuly 2000, for $1-1.5 million, TNC acquired partialownership of 7,000 acres owned by Dunn Timberlands,including 17 miles of land along the St. John. Those threedeals gave TNC an ownership interest in 60 miles of landalong the river. In September 2000, TNC pledged $1.5million toward the Pingree Forest Partnership’s easementsand promised to give an undisclosed amount of uplandforestland to the Pingrees. In return, the Pingrees agreed toput into place easements that will prohibit commercialforestry and new roads along 17 miles of the upper St.John. The TNC Maine Chapter has rolled the St. Johnproject into a larger $50 million "For Maine Forever"capital campaign that has collected contributions in excessof $1 million from each of 18 major donors, as well asdonations from scores of other big givers. In addition to theSt. John project, TNC’s campaign includes $15 million forseveral land projects in the coastal zone. By July 2000,more than $44 million had been raised. With active helpfrom L.L. Bean and other friends, TNC reached its $50million goal by January 2001. TNC has a long-term goalof protecting several million acres in the Northern Forestregion.

Through its eastern resource office, TNC is alsospearheading an effort to identify key “matrix forestblocks” in the Northern Appalachian/Acadian ecoregion,which includes the Adirondacks, Green Mountains, WhiteMountains, most of Maine, and large portions of southernQuebec, New Brunswick, and Nova Scotia.

Trust for Public Land (TPL)

In 1999, with help from the Trust for Public Land(TPL), Maine acquired a beauty strip along 65 miles (3,000acres) of shoreland on Moosehead and Flagstaff Lakes fromPlum Creek Timber Company. TPL is also currentlyworking on the Nicatous project.

Tumbledown Mountain is a major priority for TPL inMaine. Nestled in Maine's Western Mountains less than ahalf a day’s drive from Boston, Tumbledown and itssurroundings are threatened by growing developmentpressures and changes in land ownership. At the end ofJuly 2001, TPL announced agreements to purchaseapproximately 19,400 acres surrounding the mountain andnearby Mount Blue State Park from McDonald InvestmentCompany, TR Dillon Logging, and Hancock TimberResource Group. If sufficient public and private funds canbe raised, the Maine Bureau of Parks and Lands hopes topurchase roughly one-third of the property. TPL then plansto establish conservation easements on the remainder,leaving the land in private hands but permanentlyprohibiting development and protecting sensitive

environmental areas. In addition, TPL is negotiating withseveral other landowners in the area, with the ultimate goalof protecting approximately 30,000 acres.

Tumbledown Mountain and Mt. Blue State Park

The Tumbledown Conservation Alliance, led by theAppalachian Mountain Club (AMC) and supported by theNorthern Forest Alliance, has prepared "A Vision forProtecting Special Places Around Tumbledown Mountainand Mt. Blue State Park" in western Maine. The vision is amix of fee purchases by the state around TumbledownMountain (3,600 acres) and Forest Legacy easementsaround Mount Blue State Park (26,000 acres). AMC is alsotrying to raise some private funds for the project from REIand others. (See the description of conservation efforts bythe Trust for Public Land.)

In January 2002, the Maine Department ofConservation's Bureau of Parks and Lands announced thepurchase of 2,468 acres as an addition to the Mt. Blue StatePark. The property includes a 1,298-foot peak known asHedgehog Hill and approximately half of the Park'spopular 20-mile Multi-use Trail. The property waspurchased for $980,000 from New River Franklin Ltd., asubsidiary of McDonald Investment Company, Inc.Funding came from a combination of the Land for Maine'sFuture Program, the federal Land and Water ConservationFund, the Maine Outdoor Heritage Fund, and privatedonors. The Trust for Public Land assisted with thepurchase ( www.tpl.org accessed on 18 January 2002).

Umbagog National Wildlife Refuge

Since the Umbagog National Wildlife Refuge wasestablished in 1992 on the Maine-New Hampshire border,nearly 7,000 acres have received some form of protection.The long-term goal is to protect 29,000 acres,approximately half in fee and half in easements. The U.S.Fish & Wildlife Service expects to have a conservationinterest in 16,000 acres. The states would hold legalinterests in the other 13,000 acres.

White Mountain National Forest (WMNF)

Roughly 50,000 acres of the White Mountain NationalForest (WMNF) spills from New Hampshire into Maine. Itis unclear what will happen to Clinton-era regulationsenacted to protect roadless areas. Conservation groups areexploring other ideas for improving protection of theWMNF.


APPENDIX 3 - FOCAL SPECIES PROFILES

Eastern Timber Wolf (Canis lycaon)

Status and History

Note: The origin and taxonomy of eastern North Americanwolves have recently been called into question. While studyingwolves occupying Algonquin Provincial Park in Ontario, Wilsonet al. [2000] discovered a close genetic relationship between theeastern timber wolf [proposed name: Canis lycaon] and the redwolf [Canis rufus]. Further, the absence of gray wolf [Canislupus] genetic material in sampled eastern Canadian wolves andred wolves challenges the position that C. lycaon is a subspecies ofgray wolf and supports a counter-hypothesis that a small NorthAmerican wolf evolved independently of the gray wolf—whichitself evolved in Eurasia 1-2 million years ago. These findingshave broad implications for wolf recovery in northeastern NorthAmerica and will no doubt catalyze further research.

The eastern timber wolf (Canis lycaon), long-assumed tobe a subspecies of the gray wolf (Canis lupus), once occurredthroughout most of the eastern United States andsoutheastern Canada (U.S. Fish and Wildlife Service 1992).With the European settlement of New England camewidespread clearing of the land and fervent bounties on thewolf. In Maine, a statewide bounty was placed on thespecies in 1832 and remained in effect until 1916 (Bennett1988). By 1900, wolves were extirpated from NewEngland and New York as a result of human persecution,habitat alteration, and human-induced reductions in preyspecies. Indeed, by 1960, Minnesota hosted the onlyremaining wolf population in the coterminous U.S.(Harrison and Chapin 1997).

Today, the eastern timber wolf in the U.S. is restrictedto the Great Lakes region, where their populations arestable or increasing (Barry et al. 2001). In eastern Canada,the wolf occupies the St. Lawrence regions of Quebec andOntario. The eastern wolf’s largest protected habitat isAlgonquin Provincial Park in Ontario (CPAWS 2002). Nobreeding population of wolves is currently recognized inthe northeastern U.S., and no recovery program has beenundertaken in this region to date. The Maine Departmentof Inland Fisheries and Wildlife receives occasional reportsof wolf sightings, and two wolf-like canids of unconfirmedorigin were killed in Maine in the 1990s (MDIFW 1998).

In 1967 the U.S. Secretary of Interior listed the graywolf as Endangered in the coterminous United States, andthe eastern timber subspecies was given endangered statusunder the ESA in 1974 (see U.S. Fish and Wildlife Service1992). In 1978, the US Fish and Wildlife Service (U.S.Fish and Wildlife Service) developed a recovery plan for the

eastern timber wolf, and identified potential habitat forwolves in northern New York, Vermont, New Hampshire,and Maine. Subsequent research by Harrison and Chapin(1998) estimated that 48,787 km2 of potential wolf habitatin Maine and New Hampshire could support 488-1,951wolves. Although Harrison and Chapin also identified14,618 km2 of potential wolf habitat, capable of supportinga minimum of 146 wolves, in the Adirondack Park regionof New York, the Adirondacks may be too fragmented andtoo isolated from source populations to sustain a wolfpopulation over the long term (Paquet et al. 1999).

In July 2000, the U.S. Fish and Wildlife Serviceproposed to reclassify the gray wolf across the U.S. underthe ESA. Proposed changes included creating a distinctpopulation segment (DPS) for wolves in the Northeast anddownlisting wolves in the region from Endangered toThreatened. The U.S. Fish and Wildlife Service haspledged to give further consideration to wolf restoration inthe Northeast, and asserts that threatened status, althoughbiologically unjustified, will allow more flexibility inmanaging wolves. At this time, the U.S. Fish and WildlifeService is reviewing public comments on the proposed rule.

Although wolves are physically capable of dispersingthe distance between source populations in southeasternCanada and suitable habitat in the northeastern UnitedStates, several studies indicate that natural recolonization isunlikely (Harrison and Chapin 1998; Wydeven et al. 1998;Paquet et al. 1999). The shortest distance from the closestoccupied wolf habitat in Quebec to potential core habitatin Maine is roughly 70 km; from that in Southern Ontarioto New York, roughly 230 km (Harrison and Chapin1998). But southeastern Canada’s wolves are heavilyexploited outside of protected areas and may haveintrinsically low dispersal rates (Wydeven et al. 1998).Further, wolves dispersing from southeastern Canada face awide range of human and natural barriers. For example, amaintained shipping channel and unconsolidated ice in theSaint Lawrence River, dense human development, andmulti-lane highways parallel to the river are potentialbarriers to dispersal from Quebec to northern Maine andNew Hampshire (Wydeven et al. 1998). Negative attitudestoward wolves—and the consequential killing ofwolves—are other potential barriers (Harrison and Chapin1998). Thus, reintroduction appears to be the mostpromising strategy for reestablishing a viable wolfpopulation in the northeastern U.S.

Ecology and Habitat

Wolves are wide-ranging habitat generalists that havebeen known to disperse hundreds of kilometers. Fritts


(1983) documented dispersals as far as 886 km inMinnesota. Packs inhabit an area of 51 to 555 km2 or moreand are territorial (U.S. Fish and Wildlife Service 1992). Inthe Glacier National Park area, territory size averagesaround 780 km2 (Bangs and Fritts 1993).

Wolf pups typically are born in an undergroundburrow that has been abandoned by another mammal ordug by a wolf. Denning wolves prefer deep soils withadequate drainage and in close proximity to water, and theavailability of beaver (Castor canadensis) may influence densite selection (Paquet et al. 1999).

Eastern timber wolves prey mainly on white-taileddeer (Odocoilius virginianus), moose (Alces alces), and beaver(U.S. Fish and Wildlife Service 1992). At a minimum,wolves require a prey biomass of about 100-kg prey/km2

(Paquet et al. 1999). Research has shown that old, sick,weak, and disabled ungulates are most vulnerable topredation by wolves (U.S. Fish and Wildlife Service 1992).

Beaver are a particularly important alternate preyspecies. In Algonquin Provincial Park, for example, wolveseat moose, deer, and beaver. On the east side of the Park,where most of the wolf studies have been conducted, thediet is a ratio of about 33% moose (some of which isscavenged), 33% deer, and 33% beaver. Wolves will alsoeat snowshoe hare, particularly on the west side of the parkwhere they are more common, and in some instances, bear(CPAWS 2002).

Beaver are also considered a keystone species (Naimanet al. 1986). Major changes occur when a stream section isconverted to a beaver pond, including storage ofprecipitation, an increase in the amount of open canopy inforested areas, and creation of conditions favorable for otherkinds of wildlife dependent on ponds, pond edges, deadtrees and other habitats not present or in limited supply instream systems not modified by beavers (Naiman et al.1988). Beaver are common in North America, withexpanding populations in North America as a whole. Thespecies has an S5 heritage status rank throughout thenortheastern US and southeastern Canada.12

Sensitivities and Threats

Note: Except where otherwise cited, the following section isbased largely on a review by Paquet et al. 1999, pages 30-34.

Human activities significantly influence thedistribution and survival of wolves, with human population

12 NatureServe, The Nature Conservancy and the Natural HeritageNetwork have developed a conservation status ranking system todescribe the relative imperilment, or conservation status, of plants,animals, and ecological communities (elements) on a global (G), national(N), and subnational (state/provincial - S) level. Rankings are assigned,reviewed, and revised according to standard criteria, on a scale from 1 to5, with 1 being “critically imperiled” and 5 being “demonstrablywidespread, abundant, and secure” (NatureServe Explorer 2002).

density and wolves being negatively correlated. Generally,where wolves and people co-exist, people are thepredominant cause of wolf mortality. Even when wolves arelegally protected, the vast majority of wolf deaths arehuman-induced. The absence of wolves in humandominated landscapes most likely reflects high levels ofhuman caused mortality, displacement due to avoidance, orboth. It is widely recognized that human attitudes are acritical factor affecting the ability of wolves to survive in agiven area.

Numerous studies have shown a strong relationshipbetween road density and the absence of wolves. Roadsnegatively affect wolves at local, landscape, and regionalscales by increasing human access (Fuller 1989; Thurber etal. 1994; Paquet et al. 1997; Mladenoff et al. 1995). Roaddensity is considered a key index of human activity and ofthe potential for humans to kill wolves (Fuller et al. 1992).In general, wolves are not found where road density exceeds0.58 km/km2, although Mech et al. (1988, as cited inHarrision and Chapin 1998) found that wolves can survivein areas with road densities as high as 0.73 km/ km2 if theseareas abut large areas with less human access.

In Wisconsin, road density within pack territoriesaveraged only 0.23 km/km2, and few areas exceeded a roaddensity of 0.45 km/km2. The effects of roads play out inmyriad ways. For example, roads may act as mortality sinksvia vehicular collisions or by providing access for huntersand poachers. Roads destroy and alter habitat, and mayfacilitate disturbance. Roads may also be impediments towolf movement.

In addition to roads, other effects of humandevelopment may directly or indirectly threaten wolfpopulations. For example, activities that compact or clearsnow may alter winter movements of wolves by providingtravel routes into areas otherwise inaccessible. Such travelroutes may distort winter home ranges, affect predator/preydynamics, and attract wolves to agricultural and urbanareas where human/wolf conflicts are more likely.Snowmobile trails, like roads, may lead to harassment orkilling of wolves. Railroads are another source of potentialmortality. Lastly, diseases introduced by domestic animalsare a potential threat to wolf populations.

Justification and Focal Value

Keystone: Wolves are top predators that have beenrecognized as important top-down regulators of ecosystems(Terborgh et al. 1999). Recent research from YellowstoneNational Park suggests that the presence of wolves in alandscape may not only affect prey numbers, but also thebehavior of prey species such as elk (Cervus elaphus), whichin turn can affect the recruitment of certain tree species andcause shifts in vegetation communities (Ripple and Larsen2000). Wolves can play a particularly important regulatoryrole in the Adirondacks and Northern Appalachians, where


moose are recovering strongly. There is no natural predatorin Maine today capable of keeping moose populations incheck. Full recovery of Maine’s natural processes, speciesdiversity, and ecosystem integrity will require thefunctional presence of the wolf.

Umbrella: The eastern timber wolf is a wide-ranging,generalist predator that maintains very large territories anddisperses over hundreds of kilometers. Wolf recovery in theNortheast will require the re-establishment of functionalhabitat connectivity to allow for wolf movement andgenetic exchange. Large core areas and linkages designed toprotect wolves should protect habitat for myriad otherspecies as well. Further, providing secure habitat for wolveswill benefit other large carnivores sensitive to humandisturbance.

Recommendations

Note: See Paquet et al. [1999] for an in-depth discussion ofrecommendations and considerations in developing a conservationstrategy for wolves in the Northeast.

Although the eastern timber wolf may not requirewilderness to meet its biological needs, habitatsecurity—or more precisely, security from humanconflict—is widely recognized as an essential element to itslong-term viability. In Algonquin Provincial Park,Ontario, for example, wolves are heavily persecuted whenthey leave the boundaries of the park to follow deer intowinter deer yards (Théberge et al. 1996), although thisshould change with new regulations adopted by theOntario government in late 2001 (CPAWS 2002).Evidence from many regions indicates that wilderness areasand other protected cores provide the highest security forwolves (e.g., Paquet et al. 1999). Although ample suitablehabitat for wolves has been identified in the Northeast,corporate timber interests own much of this habitat. Morepublic wilderness, and private areas in which security fromhuman conflict can be assured, will be necessary toguarantee secure habitat for wolves and other species in thefuture.

The US Fish and Wildlife Service should dedicatesignificant resources to creating and implementingecologically viable recovery plans for wolves in theNortheast. Because wolves require very large areas of securehabitat, unroaded or minimally roaded core protected areaswill be essential to their viability. Further, landscapeconnectivity must be restored and maintained to allow forwolf movement and dispersal. Meanwhile, the easterntimber wolf should continue to be awarded full Endangeredspecies protection status under the Endangered Species Act(ESA) until long-term recovery is successful.

Wydeven et al. (1998) recommend a number of actionsto increase the potential for wolf populations from Quebec

to recolonize the northeastern U.S., and to strengthen anyeventual wolf reintroduction program. These actionsinclude:

• Protection of biotic linkages identified by Harrisonand Chapin (1997).

• Enhanced cooperation on wildlife conservation issuesbetween states in the northeastern U.S. andsoutheastern Canadian provinces.

• Additional research to understand and help improvehuman attitudes toward wolves and predators.

• Increased monitoring of wolves along the southernlimit of their range in Quebec.

• Increased track surveys for wolves and other carnivoresin Maine, northern Vermont and New Hampshire.

• Additional genetic investigations of wolf-coyoterelationships in Quebec and the northeastern U.S.

Paquet et al. (1999) make numerous conservationrecommendations regarding the establishment of wolves inthe Adirondack Park that can be applied to the Northeastin general. Among these recommendations are:

• Focus conservation efforts on ensuring the viability ofthe entire terrestrial carnivore guild.

• Protect habitat quality in areas where there is potentialfor wolf populations to exist.

• Preserve linkages among potential subpopulations ofwolves by protecting probable movement and dispersallinkages.

• Provide for secure latitudinal and elevationalmovements in response to seasonal and long-termclimate change.

• Restore impaired areas that wolves could use, but thatare marginally suitable because of disturbance.

• Take specific management actions where necessary tosustain wolves, including minimizing fragmentation ofwinter ungulate range, controlling potential sources ofdirect mortality on wolves, and providing additionalunroaded or minimally roaded refugia for wolves.

• Assess the effects of human activities that modifynatural snow cover—non-essential roads and trailsshould not be maintained during periods of snowcover.

• Study the effects of habitat manipulation to mimicnatural processes, such as fire. Fire may be critical forsustaining a variety of successional habitats thatsupport ungulates preyed upon by wolves.

• Assess the effects of industrial forestry on wolves andtheir prey. Examples from boreal regions of Canada,where there are relatively low human populations,significant logging operations, sizeable populations ofmoose and other prey, and moderate populations ofwolves, may provide good study areas.


• Consider elevating or burying extensive sections ofhighways that pass through critical wolf habitat andtravel corridors.

Canada Lynx (Lynx canadensis)

Status and History

The lynx originally was distributed from the borealforests of Alaska southeast to Nova Scotia, south tosouthern New England, the northern Midwest, into theRocky Mountains as far south as Colorado and Utah, andwest to the Pacific Ocean (Godin 1983). Lynx populationshave declined considerably over the last 100 years due totrapping and habitat loss and alteration, with declines mostserious in the U.S. portion of its range (U.S. Fish andWildlife Service 1998).

Maine, Montana, and Washington have the largestlynx populations in the lower 48 states, but the species israre even in these states (U.S. Fish and Wildlife Service1998). Estimates of the continental U.S. population are aslow as 700 individuals (Turbak 1998). The U.S. Fish andWildlife Service listed the lynx as Threatened under theESA in March 2000, although the listing notes that lynxremaining in the northeastern U.S., southern Rockies(Colorado and Wyoming) and Great Lakes regions do not“contribute substantially to the persistence of thecontiguous U.S. lynx population” (U.S. Fish and WildlifeService 2000).

Today, the lynx is a rare mammal in Maine, foundalmost exclusively in the boreal forests of the northwesternpart of the state (Krohn et al. 1998). The lynx is a speciesof special concern in Maine due to its low numbers and alack of reliable information on its population size andtrend. There are thought to be less than 200 lynx in Maine,but no reliable population estimates currently exist. Thispopulation, despite its small size, is of national significancebecause it is by far the largest lynx population east of theRocky Mountains. Today, Maine probably supports theonly breeding population in the eastern U.S. (Whitakerand Hamilton 1998; U.S. Fish and Wildlife Service 1999,cited in Ray 2000).

Ecology and Habitat

The lynx is one of several top predators in northernforests, requiring large areas of mature forest that containconifer and/or mixed stands, and small patches ofregenerating forest—typically around 20 years old(Banfield 1974; Parker et al. 1983; Koehler 1990). This catpreys almost exclusively on the snowshoe hare (Lepusamericanus) (Brand and Keith 1979; Parker et al. 1983;Ward and Krebs 1985; Koehler 1990; Poole et al. 1996;O’Donoghue et al. 1998), which thrives in successionalforest patches where there is an abundance of dense coverand woody stems on which to forage.

In periods of hare scarcity (approximately every 8–11years), the lynx switches its diet to squirrels, grouse, andother small mammals and birds (O’Donoghue et al. 1998).In these cyclic population lows, hare numbers can drop to1/300th of their peak and lynx populations can fall by 90%(Poole 1994). The response of lynx populations to haredeclines is typically delayed by a year, with reproduction,and especially recruitment, coming to a virtual halt. In leantimes, lynx leave established territories and seek substituteprey, making them particularly susceptible to trapping(Koehler 1990).

Snowshoe hare populations in the southern portion oftheir range (e.g., Maine) tend to fluctuate less dramaticallythan they do farther north, but also tend to be muchsmaller (Koehler 1990). This is thought to be because thehabitat is marginal near the boundary of the hare’s range.Southern lynx populations are correspondingly low, insome cases 10 times lower than in the north (Koehler1990). Occurring at such low densities, lynx are especiallyvulnerable to trapping mortality, just as they are duringcyclic population lows farther north (Koehler 1990).

Lynx tend to den in mature or old-growth forests withan abundance of coarse woody debris (Koehler 1990;Turbak 1998), and of either hardwood, coniferous or mixedcomposition. Lynx generally favor closed forest canopies tomove through, though they will cross open fields up to 100meters wide (Koehler 1990).

Home range size of lynx can vary considerably fromlocation to location, year-to-year, season to season, andbetween sexes. Home range sizes for individual lynx varyfrom as little as 13 km2 to as much as 177 km2, thoughthey are typically between 40-100 km2 (Koehler 1990).Male home ranges can be almost twice the size of femalehome ranges (Poole 1995).


Lynx are threatened by human activities that fragmentforests, such as large-scale logging, extensive road building,agriculture, and other development. Intact forests areessential to lynx; large clearcuts and other large openingsgenerally are avoided (Turbak 1998). It follows that even-aged forest management practices fail to provide the smallsuccessional patches lynx require. In places wheredisturbance dynamics are mainly driven by humans,however, such as in the sub-boreal southern periphery oftheir range, persistence of lynx populations will depend onprudent land use management practices to provide the“mosaic of closely juxtapositioned successional habitatsrequired by lynx” (Ray 2000). Roads are also problematicbecause they allow easy access to trappers and can result inhigh rates of roadkill. The reintroduction of lynx into theAdirondacks in the early 1990s failed largely because ofroadkill (Turbak 1998).


Lynx populations probably are limited in the southernportion of their range by decreasing snow depths. Only indeep snow do the very large feet of the lynx give it acompetitive advantage over the more aggressive bobcat(Lynx rufus) (Parker et al. 1983; Krohn and Boone 1999).Habitat modifications, such as agricultural clearing andconstruction of major roads, improve human access andmay favor bobcat, coyote and other generalist predators,leading in turn to competition for prey (Ray 2000).

Periodic fires historically have helped to createsuccessional patches in boreal forests, although they are nota primary source of disturbance in boreal ecosystems(Lorimer 1977). Fire suppression, therefore, is detrimentalto lynx populations (Poole et al. 1996). Last, lynxpopulations in Maine are probably particularly vulnerableto (incidental or illegal) trapping mortality because of theirlow densities. Lynx populations in some parts of thecountry have still not recovered from overtrapping thatoccurred 30 years ago (Koehler 1990; Mowat et al. 1996).


Umbrella - The lynx has value as an umbrella species asit is wide-ranging and requires large tracts of mature forest.Protecting the lynx would help protect a suite of speciesthat depend on boreal forest, from snowshoe hares to three-toed woodpeckers (Picoides tridactylus).

Recommendations

The following management recommendations shouldhelp to improve the survival of lynx populations in Maine:• Establish core protected areas in Maine that are tailored

to meet the needs of the lynx including large,unfragmented tracts of mature forest that arefunctionally connected to other such tracts.

• Allow natural disturbance regimes to proceed in orderto create small patches of successional forest. Habitatmodifications that favor competing species such asbobcat and coyotes should be avoided in areasinhabited by lynx.

• Minimize or avoid logging in lynx habitat to limitfragmentation and disturbance.

• Keep roads in lynx habitat to a minimum and closethem wherever possible.

Ultimately, large protected areas will best serve theneeds of the lynx, and large public cores should beestablished as opportunities arise.

The U.S. Fish and Wildlife Service should take furtheraction to protect the lynx in Maine and elsewhere in thecontinental U.S. Distinct Population Segments should beestablished for each discrete lynx region, including theNortheast. All threats to the species in this region shouldbe considered, and critical habitat designated asappropriate. The lynx’s protection status should be

increased to Endangered. Trapping of Maine’s lynxobviously is not appropriate given their rarity, but trappingof other species (e.g., coyotes) also should be carefullyregulated—if not banned—in lynx habitat to prevent theincidental take of lynx. Finally, an accurate populationestimate for lynx in Maine is necessary, as well as long-termmonitoring to assess population trends and to ensurerecovery.

Eastern Cougar (Puma concolor couguar)

Status and History

The eastern cougar—also commonly known in NewEngland as the catamount, panther, painter, mountain lion,and puma—once ranged from New Brunswick south to theCarolinas and west to Illinois (Young and Goldman 1946).Historically reported throughout New England (Bennett1996), cougars were heavily persecuted by early Europeansettlers and were the target of widespread bountiesthroughout the region. Bounties were established inConnecticut as early as 1694, in Massachusetts in 1742(Bennett 1996), and in Vermont in 1779 (Altherr 1994).These bounties, together with the dramatic loss of forestedhabitat and severe decline in deer populations, led to thevirtual extirpation of the eastern cougar by the late 1800s(see Taverna et al. 1999); the cougar was declared extinct inMaine in 1899 (Bennett 1996). The eastern cougar waslisted as Endangered under the ESA in 1991.

The current status of the eastern cougar is an issue ofgrowing interest and speculation. By the 1950s, sightingswere becoming increasingly frequent (Taverna et al. 1999).As a result of escalating sighting reports, the U.S. ForestService and the U.S. Fish and Wildlife Service co-funded astudy in 1978 to examine the possible survival of theeastern cougar in the southern Appalachians (Downing1996a). Although this study was unable to confirm theexistence of self-sustaining cougar populations in theeastern U.S. north of Florida, physical field evidence ofcougars is beginning to accumulate and the U.S. Fish andWildlife Service now acknowledges that there are at least afew wild individuals living in the East (Bolgiano et al. inpress). The U.S. Fish and Wildlife Service published anEastern Cougar Recovery Plan in 1982, but has notapproved its implementation (Taverna et al. 1999).

Reliable sightings suggest that there could be a smallnumber of cougars of unknown origin living in NewBrunswick and Nova Scotia (Stocek 1995). Cougarsightings in New England also continue to this day(Bennett 1996). The Eastern Puma Research Network hascompiled the following sightings in New England between1983 and 1993: 69 in Maine, 31 in New Hampshire, 29 inVermont, and 23 in Massachusetts (Bennett 1996).According to retired U.S. Fish and Wildlife Servicebiologist Robert L. Downing: “Though Maine reports the


cougar as nonexistent, I have seen a huge scat and asampling of the state’s sighting reports and know howmuch of it remains roadless (no paved, public roads atleast), so I must place Maine in the optimistic category…”(Downing 1996b). Maine officials have publicly announcedthat cougars probably exist in that state (Downing 1996a).But the origin of such cougars remains a mystery—there isconsiderable speculation they are escaped or releasedpets—and the presence of a viable breeding population ofeastern cougars is questioned (Bennett 1996).

Ecology and Habitat

Although studies of cougars in the western U.S.associate cougar habitat with steep, rugged terrain, thehabitat of the Florida panther—the only documented Pumaconcolor population in the East, is predominantly within lowelevation mixed swamp forest (Maehr 1990, in Taverna etal. 1999). Other than in southern Florida, there is no gooddata on the habitat requirements of cougars in the easternU.S.; and southern Florida is unlike any habitat whereadequate information on habitat relationships exist (J. Ray2002). In general, cougars may occupy a wide variety ofhabitats, but are now typically associated withmountainous or remote undisturbed areas. It has beendocumented that physically rugged terrain providesstalking cover and den sites, thus is a commonly mentionedattribute of good mountain lion habitat (Young 1946;Seidensticker et al. 1973; Koehler and Hornocker 1991;Fitzgerald et al. 1994; Beier 1996). Deer serve as thecougar’s primary food in many regions. Cougars are highlyopportunistic, however, and also prey on various large andsmall mammals, insects, and reptiles (Hall 1981). Insouthern California, Beier et al. (1995) found that, onaverage, an adult cougar killed 48 large and 58 smallmammals per year, and fed for an average of 2.9 days on asingle large mammal.

Cougar dens, selected for inaccessibility to otherpredators, are located in secluded areas among rocks or inextremely dense vegetation (Beier et al. 1995). Cougardensity is usually not greater than 3-4 adults/100 sq km2

(Kitchener 1991). Adult males typically are solitary, withhome ranges varying between 40-325 km2, depending onprey density, location of other cougars, and landscape type(Taverna et al. 1999). Home ranges of males are twice aslarge as those of females (Anderson et al. 1992, cited inBeier et al. 1995). In studying migratory patterns ofcougars in southern California, Pierce et al. (1999) foundthat some animals migrate seasonally while others do not,depending upon the movement patterns of the primaryprey species.

Beier (1993 and 1996) estimates that, in the absence ofimmigration in the western U.S., patches of suitablehabitat in the range of 1000-2200 km2 is necessary toensure the persistence of a population of mountain lions for

at least 100 years, and that these minimum areas wouldsupport 15-20 adult cougars. He stresses that these shouldbe considered “minimum areas”, and that much larger areasor periodic immigration or translocation of lions would berequired for longer-term persistence of the population.Emphasizing the importance of immigration to smallpopulations, Beier (1993) predicts that a mountain lionpopulation could persist in an area as small as 600-1600km2 provided up to 3 males and 1 female immigrated tothe area per decade. Beier (1993) cautions that his modelresults apply to mountain lion densities observed in theSanta Ana Mountains of California (1 adult per 100 km2),and should be adjusted for local adult lion densities andother parameters elsewhere.


Cougars are shy, solitary, and wide-ranging felids thatrequire remote habitat undisturbed by humans. In theeastern U.S., the cougar has been severely threatened, if notextirpated by habitat alteration and loss. Taverna et al.(1999) point out that cougars prefer forested habitat,despite the fact that deer densities are higher in “edge”habitats. Cougars also have been shown to select areas withno recent timber sales and minimal human residences (VanDyke 1986a).

Van Dyke et al. (1986b) found that cougars crossedsmall dirt roads more frequently than improved dirt andhard-surfaced roads, and suggested that cougars avoid areaswith an abundance of improved roads. Roads are a well-publicized threat to Florida panthers, which also arerapidly losing habitat to agriculture and development(Downing 1996b). In modeling minimum habitat areasand corridors for cougars, Beier (1993) concludes thatsurvival rates for a population inhabiting a large area willprobably decrease due to increased highway mortality anddecreased dispersal success. Numerous studies havesuggested that the protection and maintenance ofmigratory corridors may be integral to cougar conservation(e.g., Beier 1993; Pierce et al. 1999).


Umbrella: Extrapolating from Beier (1993, 1996), theEastern cougar probably has the largest area requirementsof all species profiled for the Maine Wildlands NetworkDesign. This quality makes the cougar an ideal umbrellafor the conservation of other native species. Beier (1993,1996) cautions that his model results apply to mountainlion densities observed in the Santa Ana Mountains ofCalifornia (1 adult per 100 km2 [39 mi2]), and should beadjusted for local adult lion densities and other parameterselsewhere. Nonetheless, they serve as a guide for cougarconservation planning. The Maine Wildlands NetworkDesign identifies conservation areas that are sufficiently


large to accommodate the minimum habitat requirementsof the mountain lion.

Keystone: As a major predator of ungulates, easterncougars likely exerted a top-down influence on white-taileddeer, and potentially on caribou. By preying on ungulatesand killing mesopredators like bobcats, mountain lions, aspart of a “suite” of top predators, may contribute to thetop-down regulation of ecosystems. Mountain lionpredation may 1) dampen oscillations of prey populations,producing a more even distribution of prey across thelandscape, which would in turn reduce potential habitatdamage from over-browsing or over-grazing; 2) enhancebiological diversity by contributing to the suppression ofmesopredator populations; and 3) promote ecosystemstability. Empirical evidence that mountain lions play akeystone role in ecosystems in the Northeast is largelylacking, however—due to their long absence orscarcity—so the hypothesis regarding their keystone role isbased largely on ecological theory. Nevertheless, recentstudies on forested islands in Venezuela suggest thatextirpation of another large felid, the jaguar (Panthera onca),may be resulting in a hyperabundance of seed predators andgeneralist herbivores, leading to an impoverishment ofnative tree species (Terborgh et al. 1999, 2001). Recoveryof the eastern cougar would almost certainly restore animportant functional influence to Maine’s ecosystems.

Recommendations

According to information compiled by Bolgiano et al.(in press), it is possible that cougars may already bebreeding in the Mid-Atlantic and Northeast regions of theU.S. Given the recovery of the Northeastern forests andample prey populations in this region, the U.S. Fish andWildlife Service should develop a revised Eastern CougarRecovery Plan for the entire eastern U.S. Concurrently,conservation groups should initiate public and hunter-education programs, and begin to study current publicattitudes toward this species.

Core areas should be designed and implemented suchthat they provide ample wild habitat for secure breedingand refugia for source populations. Linkages should bedesigned with sufficient width to be used by dispersingcougars. Further, wildlife overpasses or underpasses shouldbe integrated into any new highway or major road projects.In this vein, areas of suitable habitat that cross major roadsshould be identified, and overpasses installed to facilitatethe natural recovery of cougar populations to this region.

American Marten (Martes americana)

Status and History

The American marten currently is distributedthroughout the boreal and taiga zones of Canada, Alaska,and the continental United States. Marten occur from coastto coast, with a southern boundary extending along theRocky Mountains as far south as New Mexico, throughmidwestern states such as Minnesota and Michigan, to theNorthern Forest of New England (Clark et al. 1987). Thismember of the weasel family historically was found as farsouth as West Virginia in the eastern U.S. (Buskirk andRuggiero 1994). In Maine, marten are currently found inthe northern two-thirds of the state (Krohn et al. 1998).

Marten populations are demonstrably secure in boththe U.S. and Canada, with Heritage Status rankings of 5 inboth countries (Krohn et al. 1998). Marten appear to bethriving in northern Maine, the Adirondacks, andCanadian provinces to the north; the species has a Heritageranking of 5 in Maine and Quebec, and 4 in NewBrunswick. Marten populations are faring much worse tothe south, however: they are imperiled in New Hampshire,imperiled or extirpated in the Canadian MaritimeProvinces, and extirpated in Vermont, Massachusetts, NewJersey, Pennsylvania, West Virginia, Ohio, Indiana, andIllinois. Marten populations in Maine and the Adirondacksare critical because they are the southernmost viablepopulations in eastern North America.

Prior to European settlement, the marten was one ofthe most abundant carnivores in the eastern U.S.(Thompson 1991). Yet by the early 1900s the logging offorests and their conversion to agricultural land had causedthe disappearance of the marten from much of the region(Buskirk and Ruggiero 1994). Overtrapping of marten alsocontributed to their decline. Marten populations have nowrebounded in the Adirondacks and Northern Appalachiansdue to forest re-growth and successful reintroductionefforts. Maine’s marten are once again trapped in largenumbers, with an annual kill of 3,000-5,000 animals overthe last two decades (Krohn et al. 1998).

Ecology and Habitat

The marten, a medium-sized weasel, preys primarilyon small mammals hunted on the ground or in trees.Marten most commonly prey on microtine rodents, such assouthern red-backed (Clethrionomys gapperi) and meadow(Microtus pennsylvanicus) voles, but will also consumesnowshoe hare, deer mice (Peromyscus leucopus), red squirrels,and seasonal fruits (Bissonette et al. 1997).

Although marten generally prefer mature coniferousforests, they also use mature deciduous and mixed forests.Marten are tied to mature forests primarily because ofrelated structural features such as large trees, snags, and


logs, which are used for denning, and piles of coarse woodydebris (CWD), which are used as winter resting sites(Buskirk et al. 1989; Payer and Harrison 1999, 2000). Inaddition, coarse woody debris probably allow marten toaccess hunting areas beneath snow (Corn and Raphael1992). Habitat features such as snags, logs, or coarse woodydebris are much more frequent in mature forests than inyounger forests. Studies in Maine have shown, however,that marten can be found in young forests or forestswithout closed canopies (Chapin et al. 1997). This is likelyto occur only in stands where structural diversity is high,such as those where spruce budworm outbreaks havesignificantly reduced canopy cover (Chapin et al. 1997). Instands with a history of budworm-caused mortality, thereare often many more defoliated tree boles, and more coarsewoody debris generally, than in clearcuts of a similar age(Payer and Harrison 2000)


Marten do not fare well in heavily logged areas, aslogging reduces available habitat, and they rarely forage inclearcuts (Snyder and Bissonette 1987). In addition,scientists have found that marten population densities aremuch lower in forests cut 40 years ago than in uncut forests(Thompson 1994). Marten also forage much moresuccessfully in mature forests than in young ones(Thompson and Colgan 1994). These mustelids avoid smallforest patches, or patches isolated from other forests (Snyderand Bissonette 1987; Chapin et al. 1998). In a studyconducted in Maine and the Rocky Mountains, martenwere completely absent from landscapes that were less than70% forested (Chapin et al. 1998) even when forest patchesare well connected (Hargis et al. 1999). When averagenearest neighbor distances between non-forested patchesare less than 100 meters, forest interior conditions thatcontain habitat attributes required by martens tend to belost, even though prey densities in regenerating clearcutsabound (Hargis et al. 1999, cited in Ray 2000).

Marten are easily trapped, and their reproductive ratesare low and the age at sexual maturation high bymammalian standards; so they are slow to recover frompopulation level impacts (Buskirik and Ruggiero 1994,cited in Ray 2000). Marten were nearly eliminated inMaine in the 1930s because of overtrapping, and therecontinues to be concern in the state about trapping pressure(Hodgman et al. 1994, cited in Ray 2000).


Umbrella - The marten is a carnivore that requires largetracts of forest with complex vertical and horizontalstructure, although not all that forest must be mature(Payer and Harrison 1999). As Payer and Harrison (2000)note “marten have been declared an ecological indicatorspecies for forest ecosystem integrity in New Brunswick,

Ontario, and several national forests in the western U.S.This suggests that identification of specific featuresaffecting habitat suitability for marten might also benefitother forest-dependent wildlife.” Protection of martenwould thus help protect a suite of species that depend onintact forests, such as snowshoe hares, northern goshawks(Accipiter gentilis), spruce grouse (Dendragapus canadensis),black-backed (Picoides arcticus) and three-toed woodpeckers,and red (Loxia curvirostra) and white-winged (Loxialeucoptera) crossbills.

Habitat Quality Indicator—Some of the same factorsthat make marten an umbrella species also make it ahabitat quality indicator. As discussed above, marten donot fare well in heavily logged areas, as logging reducesavailable habitat, and they rarely forage in clearcuts (Snyderand Bissonette 1987). As also noted above, martenpopulation densities are much lower in forests cut 40 yearsago than in uncut forests (Thompson 1994). Marten alsoforage much more successfully in mature forests than inyoung ones (Thompson and Colgan 1994). These mustelidsavoid small forest patches, or patches isolated from otherforests (Snyder and Bissonette 1987; Chapin et al. 1998).

Recommendations

Protected areas tailored to meet the needs of martenshould include many large, unfragmented forest tracts thatcontain significant areas of mature, coniferous forest. Inharvested forests, logging should be limited to no morethan 25% aerial cover and clearcuts should be eliminatedor at least grouped to avoid habitat fragmentation.Structural features such as snags, coarse woody debris, andlive mature trees should be maintained. Payer and Harrison(1999) recommend that:

Harvested stands be managed to maintainoverstory features above the thresholds required bymarten. We suggest that basal areas of live-treesand snags (>13 cm dbh [> 5 inches dbh]) bemaintained above 18 square meters per hectare (80square feet per acre), and that at least 25% of thestanding basal area be composed of snags.Furthermore, our data indicate that marten selectstands where mean tree height is > 9 meters (> 30feet), thus we recommend that the mean height ofresidual trees and snags be maintained at aminimum of 9 meters (30 feet).

Even in areas with carefully planned logging, martentrapping may need to be eliminated or at least limited toprevent a synergistic threat to the population, as the“effects of trapping appear confounded with forestharvesting because trappers use logging roads to accessresidual marten habitat (Ray 2000).”


Northern River Otter (Lutra canadensis)

Status and History

The northern river otter ranges throughout NorthAmerica north of Mexico, excluding the extremesouthwestern U.S. The population trend of otters probablyis stable in much of its range, and reintroduction andmanagement efforts are boosting populations in underpopulated areas. River otters are considered commonthroughout the state of Maine and are managed as a gamespecies.

Along with beaver and gray wolf, the river otter onceoccupied one of the largest geographical areas of any NorthAmerican mammal. In the late 1800s and early 1900s,however, wetland destruction, pollution andoverexploitation for furs were devastating to NorthAmerican river otter populations (Ray 2000). Populationswere most drastically reduced where human populationswere dense, agricultural or industrial practices wereparticularly intense, wetlands were naturally sparse, oroligotrophic waters could not support a healthy prey base(Polechla 1990).

Ecology and Habitat

Northern river otters are habitat generalists, using avariety of aquatic habitats, including estuaries, lakes,ponds, rivers, streams, marshes, and swamps. Otters areparticularly abundant in low gradient areas with largerivers, wetland complexes, and beaver ponds. Such areastypically produce an abundance of prey and provideisolation and den sites (Maine Cooperative Fish andWildlife Research Unit 1998). Urban areas are typicallyavoided, although on Mt. Desert Island there is evidencethat river otters cross residential areas to move betweenponds and the ocean (Maine Cooperative Fish and WildlifeResearch Unit 1998).

Primarily foragers in shallow waters, river ottersopportunistically feed on crustaceans, shellfish,amphibians, and other non-aquatic foods, but fish comprisethe bulk of their diet (Larivière and Walton 1998;Whitaker and Hamilton 1998 cited in Ray 2000). Somestudies in Maine show that otters associate strongly withbeaver ponds, largely because this habitat provides den andresting sites, stable water levels, and abundant vegetativecover (Dubuc et al. 1990).


Otters are relatively resistant to human impacts, butvulnerable to overtrapping and to degradation of aquatichabitats (Toweill and Tabor 1982). They are scarce inheavily settled areas, particularly if waters are polluted(Toweill and Tabor 1982). Otters may in fact be able to

tolerate a large variety of threats as long as food availabilityremains high (Ray 2000).

Ray (2000) cautions however that althoughmanagement practices aimed at increasing beaver habitathave also improved otter habitat in the northeastern U.S.,habitat change continues to be the major cause of concernfor this species. Key vulnerability factors include (Kruuk1995):

• Linear home ranges requiring large areas;• Reliance on water, also badly needed by humans;• Importance of bank vegetation, also sought after

by humans.

Ray (2000) notes further that:

By virtue of their aquatic existence and position atthe top of the foodchain, otters are particularlyvulnerable to pollution in aquatic ecosystems(Melquist and Dronkert 1987; Foley et al. 1988).They readily accumulate high levels of mercury,organochloride compounds, and other chemicals tolevels significantly above environmentalconcentrations (see references in Larivière andWalton 1998). The decline of otter in Europe hasbeen attributed in large part to the introduction oforganochlorine insecticides for use in sheep dipand as seed dressings (Dunstone 1993). In onestudy of sympatric otters and mink, ottercontained greater PCB concentrations than didmink, suggesting that the exposure to thiscompound was greater for otters than mink, thatotters were more tolerant of PCBs than mink, orthat otters were less efficient than mink ineliminating PCBs (Foley et al. 1988). Recent workin Nova Scotia showed much higher mercuryconcentrations in otters captured in inland vs.coastal waters, and scientists are investigatingwhether declines reported by trappers working ininland areas are related to this condition.


Umbrella - River otters serve as an umbrella species fora broad variety of aquatic habitats because of their need forlarge linear home ranges, clean standing or running water,and at least a moderate degree of streambank vegetation.The umbrella value of the otter is enhanced by itssensitivity to aquatic and airborne pollutants.

Habitat Quality Indicator—As Ray (2000) observes,above, by virtue of their aquatic existence and position atthe top of the foodchain, otters are particularly vulnerableto pollution in aquatic ecosystems. They may thereforeserve as broader indicators of habitat and water quality in


riparian areas. Also, because of their sensitivity to airbornepollutants, they may also serve as indicators of air quality.All of these qualities are particularly important in thenortheastern US, especially in areas where mercuryconcentrations are highest. “If otter, which are moreresilient than loon (by virtue of the fact that otter exhibitmore mobility between lake and stream systems—loons aremore faithful to a particular lake), begin to show effects [ofmercury pollution], as they do in Nova Scotia, then we arereally in trouble (Ray 2002).”

Recommendations

Wetland complexes with beaver ponds situated alonglarge rivers are areas of high otter productivity and shouldbe targeted for protection. Forestry and agriculture in andaround wetlands and riparian areas should be modified totake into account the vulnerability factors identified above.Best management practices include wide buffer stripsaround riparian areas and fencing livestock out of streamzones. Because of their sensitivity to water and airbornepollutants, a systematic otter monitoring program shouldbe instituted throughout the Maine Wildlands Network.Such a program would sample not only population trendsbut also the levels of mercury and organochlorinecompounds, among other chemicals, in otter populations.Finally, because recent history suggests that this species isvulnerable to overexploitation, trapping patterns andeffort—especially by otter “specialist trappers”—should beclosely monitored (Ray 2000).

Northern Goshawk (Accipiter gentilis)

Status and History

The northern goshawk is a large accipiter that nests inmature forests throughout the Northern Hemisphere, fromthe U.S. to the British Isles, to the Himalayas. In NorthAmerica, the goshawk ranges from Alaska east to theCanadian Maritime Provinces, and south to California,central Mexico, the Rocky Mountains, the Great Lakesregion, West Virginia, and New England. Goshawks aregenerally year-round residents, although they occasionallymigrate out of northern parts of their range. Globally,northern goshawks are considered stable, with a NaturalHeritage ranking of 5.

Goshawk populations undoubtedly declined in colonialtimes as forests in the Northeast were cleared (Kennedy1997). With the restoration and maturation of forests inthe Northeast as whole, the nesting range of the goshawk isexpanding (Squires and Reynolds 1997). The goshawkpopulation in Maine may be declining, however (MaineCooperative Fish and Wildlife Research Unit 1998).

Ecology and Habitat

Goshawks are birds of prey that generally hunt in theinteriors of forests, preying primarily on birds and smallmammals (Grzybowski and Eaton 1976), andsupplementing their diet with reptiles, amphibians andoccasional invertebrates. They will also hunt along forestedges and small openings. The goshawk is capable ofcatching birds in the air, but also hunts them on theground or in vegetation.

Goshawks generally use large tracts of mature or old-growth forest (Bosakowski and Speiser 1994; Woodbridgeand Detrich 1994) with little human disturbance(Hayward and Escano 1989; Squires and Ruggiero 1996;Squires and Reynolds 1997). In Maine, goshawks prefer tonest in the interior of extensive, remote coniferous or mixedforests (Maine Cooperative Fish and Wildlife ResearchUnit 1998). They usually nest in the largest treespossible—often in deciduous species such as beech, birch,or poplar. Nest sites are typically in a stand with highcanopy closure and sparse groundcover (Squires andRuggiero 1996), and near water. Nesting densities are lowfor this raptor, 4 - 11 pair/100 km2 in the western U.S. andas low as 1 pair/100 km2 in the Northeast (Squires andReynolds 1997). Nesting home ranges of goshawks in theWest average 2,400 ha. In a study in Arizona, goshawknests were, on average, three kilometers apart (Reynolds etal. 1994). There is some evidence in Europe that goshawksare becoming more acclimated to agriculturalenvironments, nesting near farms and towns. The sametrend may be developing in the U.S., with reports ofnesting goshawks in the suburbs of Boston (Palmer 1988a).


Forest management practices are the primary threat topopulations of nesting goshawks, despite their apparentabilities to adapt to agricultural and suburbanenvironments (Squires and Reynolds 1997). Even selectivelogging can harm goshawk populations. Logging activitiestoo close to goshawk-nesting sites can cause parents toabandon their nests (Boal and Mannan 1994; Squires andReynolds 1997). Logging that targets large trees or reducescanopy cover significantly degrades the nesting habitat ofthis raptor. In addition, the opening up of the canopy oftenresults in the replacement of goshawks by red-tailed hawksor great-horned owls (Erdman et al. 1998). Great-hornedowls are a problem because they prey on both young andadult goshawks (Boal and Mannan 1994; Erdman et al.1998; Rohner and Doyle 1992).


Umbrella - The goshawk is an excellent umbrella formature forests because of its large area requirements. Thesingle nesting density estimate for the eastern U.S. is 1.17


pairs per 100 km2 (Kimmel and Yahner 1994, cited inSquires and Reynolds 1997). Protecting the goshawkwould protect large expanses of the Northern Forest andthe myriad species that rely on its integrity.

Habitat Quality Indicator - The goshawk is an indicatorof forests that are free of extensive logging and in whichecological processes such as fire are relatively intact.

Recommendations

The primary action required to conserve goshawks is tophase out or regulate forest management in current andpotential goshawk habitat in Maine. Large clearcuts shouldeither be eliminated or severely limited in such habitat.Clearcuts that do occur should be clustered to minimizehabitat fragmentation. Any logging should be carefullyplanned to provide goshawks with many seral stages offorest to maximize habitat diversity and prey density. Largetrees and old growth stands should remain untouched. TheU.S. Forest Service recommends that managementactivities should be excluded from a 170-ha area around agoshawk nest during the breeding season (Reynolds et al.1992; Graham et al. 1994). Naturally occurring firesshould be allowed as well, as they foster habitat diversity.In some areas prescribed burns may be necessary to achievethis objective.

Protected areas tailored to conserve goshawks shouldinclude large stands of mature forest with significant standsof conifer or mixed forest and should be connected by suchhabitat to other cores. In addition, cores should includemicrohabitat features such as large trees, small openings,woody debris and snags. Finally, it is critical to begin long-term monitoring of goshawk populations because no trenddata are currently available for this species.

Red-shouldered Hawk (Buteo lineatus)

Status and History

The red-shouldered hawk is a medium-sized buteo thatfrequents woodlands in southern Canada and the U.S. Thisraptor breeds in two distinct areas, with the first stretchingfrom central California south to Baja California; the second,from Nebraska to the Great Lakes states to southernQuebec and New Brunswick, and south to Florida, Texasand eastern Mexico. Red-shouldered hawks in northernareas migrate to spend the winter south of a figurative linerunning from Kansas to Pennsylvania to southern NewEngland.

Although red-shouldered hawks appear to be relativelysecure globally, they are much less numerous than theywere 100 years ago in most of their range. In fact, globalnumbers are likely at their historic lows, with an estimated3,000-10,000 individuals representing the entire species.Once the most common woodland hawk east of the RockyMountains, the red-shouldered hawk has declined

significantly in the northern part of its range. This declineoccurred as forests were cleared for agriculture anddevelopment in the 19th and early 20th centuries, and hasonly begun to reverse in some areas since the 1970s(NatureServe Explorer 2002). Hawks in midwestern statessuch as Illinois, Indiana, Wisconsin, and Ohio declined byas much as 65 to 95 percent in Christmas Bird Counts from1950 to 1969 (Brown 1971).

As regenerating forests have begun to provide moresuitable habitat for the red-shouldered hawk, populationsmay have stabilized in some areas of the Northeast inrecent years (Titus and Fuller 1990). The species is rankedas uncommon in Maine, however—far less common thanwould be suggested by GAP-predicted habitat—and isspeculated to be undergoing a moderate population declinein this state. Currently, red-shouldered hawks breed insouth and central Maine (Maine Cooperative Fish andWildlife Research Unit 1998).

Ecology and Habitat

The red-shouldered hawk frequents mature forests,foraging for a variety of prey, including small mammals,reptiles, amphibians, the occasional bird, and someinvertebrates (Palmer 1988b). The red-shouldered hawkgenerally requires large areas of mature hardwood or mixedforest (Bednarz and Dinsmore 1982; Evers 1992) thatcontain wetlands and large trees for nesting. Nesting areasare usually in dense, mature stands that have a highcanopy, are in close proximity to water (Maine CooperativeFish and Wildlife Research Unit 1998), and are far fromhuman use areas (Hands et al. 1989). In eastern NorthAmerica, nests generally are far from forest edges, and atleast 43 mostly deciduous tree species may be used,suggesting that the size and shape are more important thanthe actual species (Bednarz 1979; Apfelbaum and Seelbach1983; Titus and Mosher 1987; Palmer 1988b; Ebbers1989).

Parents will often use the same nest year after year, andsuccessive pairs of birds may occupy a red-shouldered hawkterritory for decades. Scientists estimate that a pair of red-shouldered hawks typically needs 100-250 ha of breedingterritory (Evers 1992).


Forestry practices are the main threat to the red-shouldered hawk. Clearcuts are detrimental because theyreduce the area of preferred habitat; too much selectivecutting may result in replacement of this species by thered-tailed hawk (Buteo jamaicensis). Indeed, the red-shouldered has been replaced by the red-tailed hawk as themost common hawk in much of its range, primarilybecause of logging practices that favor the red-tailed hawk(NatureServe Explorer 2002). Red-tailed hawks prefer, oraccept, forest patches that are smaller, closer to roads, and


have less canopy cover (Bednarz and Dinsmore 1982), andsmaller crown diameter than those preferred by the red-shouldered hawk. Logging that fragments forests, removesmature trees, thins dense stands, and has short rotationperiods puts the red-shouldered hawks at a competitivedisadvantage against the red-tailed hawks. In addition, red-shouldered hawks are sensitive to disturbance caused bylogging near their nests (Bryant 1986).

Other threats to the red-shouldered hawk includerecreation and the loss of wetlands. This raptor is asecretive nester, and as development increasinglyencroaches upon forested areas, disturbance from humanactivities has pushed the hawks into remotewoodlands—thereby diminishing their habitat.Predictably, red-shouldered hawks generally avoid nestingnear roads (Bednarz and Dinsmore 1982).


Umbrella - The red-shouldered hawk is an effectiveumbrella species for mature hardwood and mixed forestsbecause of its large area requirement. Protecting this raptorwould protect large tracts of northern hardwoods and theother species that rely on this habitat.

Habitat Quality Indicator -The red-shouldered hawk isan indicator of unfragmented forests subject to no orminimal logging or other forms of human disturbance.

Recommendations

Restoration and recovery planning for red-shoulderedhawks should receive higher priority in the Northeast. Theconservation of red-shouldered hawks requires carefulregulation of forest management. Large clearcuts should beavoided in prime habitat, and clearcuts—if allowed atall—should be clustered to minimize fragmentation.Selective cutting should be planned to avoid the removal ofmature hardwood trees and the thinning of dense forestpatches, and to minimize road building. Maintaining acrown closure of greater than 70 percent may prevent red-tailed hawks from displacing red-shouldered hawks.Forested and open wetlands should be left intact.

Active red-shouldered hawk nests should be bufferedagainst any human disturbance, although it is unclear howlarge the restricted zone should be. Evers (1992) hasrecommended that a distance of at least 90 m from the nestsite be kept free from human disturbance. Existing roadsshould be closed to public use during the breeding season.

Ecological cores designed to accommodate red-shouldered hawks should encompass large stands of maturehardwood or mixed forest and should be connected by suchhabitat to other cores. As mentioned above, at least 100 to250 hectares of mature forest probably are required for eachnesting pair, although hawks will use smaller patches ifthey are near large tracts.

Existing red-shouldered hawk populations should beclosely monitored, and research conducted on minimumarea requirements for a breeding pair, competitiondynamics between red-shouldered and red-tailed hawks,and the optimum size of restricted areas around nests.

Common Loon (Gavia immer)

Status and History

The common loon is one of five loon species that breedin North America, but is the only one breeding in thenortheastern U.S. The species breeds from Iceland,Greenland, Canada and Alaska, south to California, NorthDakota, the Great Lakes area, southern New England, andNova Scotia. Common loons winter along the Pacific coastfrom the Aleutians to Baja California, and along theAtlantic from Newfoundland to Florida to Texas.

Common loons have declined throughout much of theU.S. over the last 150 years, with their range graduallycontracting northward (McIntyre 1988a). Overall,populations of this loon species are probably stable, with aNatural Heritage rarity ranking of 5 for the globalpopulation, a similar ranking for breeding populations inCanada, and a rarity ranking of 4 for breeding populationsin the U.S. (NatureServe Explorer 2002). Loon numbersmay be increasing in North America, with Breeding BirdSurvey data showing a statistically significant annualincrease of 2.2% from 1966 to 1989 (Droege and Sauer1990). Estimates place the continental loon population at500,000 to 600,000 adults, with most of these loons inCanada and Alaska and only an estimated 18,000 to thesouth (McIntyre 1988a). In the U.S., loons breed in Alaskaand 13 other northern states (Ehrlich et al. 1992).

Loons are uncommon in the Northeast, with Mainebeing the only state with a Heritage ranking higher than 3.The population of breeding loons is probably secure inMaine, with a population trend that appears to be stable(Maine Cooperative Fish and Wildlife Research Unit1998). The loon is listed as endangered in Vermont,threatened in New Hampshire, and as a species of specialconcern in Massachusetts and New York. Loon populationsare generally stable or increasing in these states due toconservation efforts that are reducing human impacts(Rimmer 1992). The potential for loon recovery in vacantportions of their range is considered high (Rimmer 1992).There is some dispute regarding these positive trends.Some populations in eastern Canada have shown verylow—or no—reproductive success in recent years. Thisappears to stem from a loon’s allegiance to a particularlake—a loon will remain on a lake even if the lake isdead—which in turn makes them particularly susceptibleto contaminants such as mercury, and to the effects of acidrain on a lake. Thus, air pollution is probably a


significantly greater threat to loons than local habitat ornest loss (Ray 2002).

Ecology and Habitat

Common loons are largely piscivorous, hunting for fishin the clear waters of large lakes, rivers, and marshes. Thesebirds are well-adapted to chasing fish underwater, havingsolid bones and a weight of up to 6.3 kg, and rear-set feetto aid swimming (McIntyre 1988a). Loons hunt in shallowwater, typically at a depth of 5 m or greater. Loons eatsmall to medium-sized fish of many species, preying mostfrequently on yellow perch, but are opportunistic and willeat any suitable prey they can readily see and capture(McIntyre 1988a). Water visibility of at least 3-4 m isnecessary in order for loons to see their prey.

Loons need large stretches of open water partiallybecause they are so heavy; they need to flap and run 200-400 meters before becoming airborne. They also require asignificant area for foraging, with territories ranging in sizefrom 40-500 hectares. Lakes smaller than 80 hectares areonly able to support one mated pair (Johnsgard 1987).

Loons also require suitable nesting areas. Loons nest onsmall islands, in marshy coves, in wetlands at inlets oroutlets, in deadwaters of streams up to 800 m from lakes,and on mainland shorelines (Maine Cooperative Fish andWildlife Research Unit 1998). Loons probably preferislands because of the dearth thereon of nest predators.They place their nests close to water's edge because theirlegs are set so far back on their body that they can barelywalk. Females lay two eggs, incubate them for a month,and then raise and feed their young for 8-12 weeks(McIntyre 1988a). They typically use shallow, secludednursery coves in which to raise their young (McIntyre1983). In the fall, Maine loons migrate to the Maine coastand south, where they hunt for fish in shallow ocean waters(McIntyre 1988a). Juvenile loons probably inhabit coastalareas until they mature at 3-4 years old (McIntyre 1988b).


There are many threats to common loons, includinghabitat loss and degradation, disturbance from recreation,predation, and fluctuating water levels, and air and waterpollution. Increased human activity associated withshoreline development can depress nesting success. Further,water levels in reservoirs that fluctuate as little as 20 cmcan flood nests (McIntyre 1988b). In the best case, thisscenario causes loons to renest and prolong the strenuousparenting period; at worst, it results in complete nestfailure. Wakes from motorboats can have the same effect(Vermeer 1973).

Nest losses are more harmful later in the breedingseason, since loons will not renest after a certain point(McIntyre 1988a). Recreation near loon nests can causeparents to temporarily flee the area, exposing eggs to

predators such as raccoons (Procyon lotor), blue jays(Cyanocitta cristata), gulls (Larus spp.) and American crows(Corvus brachyrhynchos) (McIntyre 1988a). This problem isexacerbated by the fact that such predators have increasedsignificantly near many lakes due to development. There issome evidence that loons can habituate to recreationalactivity, with loons on developed lakes abandoning theirnest in response to disturbance much less readily than loonson undeveloped lakes (Titus and VanDruff 1981; McIntyre1988a).

Abandoned fishing gear can also harm loons, which canbecome entangled in monofilament fishing line and ingestlead sinkers—the latter are usually fatal (Okoniewski andStone 1987). Loons can also accumulate high levels ofmercury in their tissues. Most mercury either precipitatesout of the atmosphere or is released from chlor-alkali andwood pulp processing plants (McIntyre 1988a). Scientistsbelieve that high mercury levels may have caused a die-offof up to 7,500 loons in coastal Florida in 1983 (Alexander1985). Acid rain that lowers the pH of lakes also lowerslake productivity, thereby reducing the loon's prey base(Rimmer 1992). Finally, oil spills in coastal areas canjeopardize wintering and juvenile loons


Habitat Quality Indicator - Loons thrive in lakes that arelargely free of the turbidity caused by erosion andeutrophication, free of the effects of acid rain, are relativelyundeveloped, and have relatively natural fluctuations inwater levels. The species can apparently tolerate somehuman disturbance, but fares better in areas where it isminimized.

Recommendations

Potential loon nesting habitats, including smallislands, marshy coves, and mainland shorelines, should beprotected from development. Protecting existing nest sitesis important because loons will often reuse them for manyyears (Strong et al. 1987). In addition, nursery areas need tobe buffered from development. Recreation should beprevented near loon nests, and the speed of motorboats andthe size of their engines should be limited to prevent largewakes from flooding nests.

In reservoirs hosting nesting loons, water levels shouldbe carefully regulated during the summer to avoid floodingor stranding nests. Artificial nesting platforms that rise andfall with water levels may improve nesting success on lakesthat lack natural islands and have poor shoreline nestinghabitat, fluctuating water levels, or a history of lowproductivity (Wood 1979). Island nesting sites andartificial nest platforms are particularly important on moredeveloped lakes because they offer protection from theartificially high populations of mammalian mesopredatorsassociated with development. Ecological reserves designed


to restore and protect loons should encompass all nearbylarge lakes—especially those that are undeveloped. Furtherresearch should be undertaken on the ecology of winteringand subadult loons in the coastal areas of Maine.

Finally, because of the sensitivity of the species to airand water pollution and of concerns about populationdeclines in eastern Canada, the reproductive success ofloons in Maine should be closely tracked, and these datashould be shared with other states and provinces in theregion.

Atlantic Salmon (Salmo salar)

Status and History

The Atlantic salmon spans a broad range in theNorthern Hemisphere, historically breeding from Spainnorthward and eastward along the coast of Europe toScandinavia and western Russia, including populations inthe Baltic Sea. This historic range extends westward fromScandinavia and the Norwegian Sea to the British Isles,Iceland, Greenland, eastern Canada, and south along thecoastal regions of Canada and New England (Parrish et al.1998).

Atlantic salmon in the U.S. were historicallyconcentrated in Maine; they occupied almost every coastalriver and the larger, inland rivers of the state. Salmon werealso found in at least 11 rivers in southern New England(U.S. Fish and Wildlife Service 1999a). The HousatonicRiver in western Connecticut was the southern limit ofbreeding Atlantic salmon in North America. Salmon arebelieved to have occurred in up to 34 rivers in Maine (U.S.Fish and Wildlife Service 1999a), with the larger riverssuch as the Androscoggin, Kennebec, and Penobscotproducing the majority of fish. The Connecticut andMerrimack Rivers were highly productive salmon rivers insouthern New England (U.S. Fish and Wildlife Service1999a).

The Atlantic salmon’s range has been graduallyshrinking. Of the 53 major watersheds worldwide thatwere once inhabited by salmon, populations are nowextirpated in 8, extirpated but being reintroduced in 15,declining in 14, and stable in only 16 (Mather et al. 1998).The Atlantic salmon was extirpated in southern NewEngland by 1865, and Maine’s population was reduced to2,000 returning adults by the 1940s (U.S. Fish andWildlife Service 1999a). These declines occurred in the faceof conservation and reintroduction efforts that began asearly as the mid-19th century. The U.S. Fish and WildlifeService (1999a) cites dams, overfishing, and waterpollution as the primary causes of salmon decline.

The U.S. Fish and Wildlife Service listed the Gulf ofMaine population segment of Atlantic salmon asEndangered under the Endangered Species Act inNovember 2000 (U.S. Fish and Wildlife Service 2000). In

early 2002, the National Academy of Sciences confirmedthat Maine’s wild Atlantic salmon form a distinctpopulation segment, one genetically distinct fromCanadian salmon (Young 2002). Fewer than 300 adultsnow return annually to Maine’s rivers (Stevens 1999)—afraction of the estimated 500,000 adults that returned eachyear to rivers in the United States in pre-colonial times(Malakoff 1998).

Ecology and Habitat

The Atlantic salmon’s use of habitat is complex due tothe many requirements at its different life stages.Essentially, these anadromous fish require habitat largelyfree of human influence in all reaches of a riversystem—from headwaters to tributaries to the main stemsof rivers, and to the watersheds that influence ripariancharacteristics. Eggs are laid in gravel beds (redds) in deepriffles that must be free of sedimentation (Poff and Huryn1998). Fry need pools and an abundance of large woodydebris (LWD) in headwaters (Poff and Huryn 1998). Parrrequire fast riffles for foraging and deep pools foroverwintering (Poff and Huryn 1998). Smolts, thedevelopmental stage of salmon most vulnerable topollution, use the salmon habitat that is most exposed tohuman impacts—the mainstem of rivers (McCormick et al.1998). Finally, returning adult salmon need estuaries inwhich to rest, and rivers and tributaries free of dams orother obstructions (Parrish et al. 1998).


Scientists have identified many factors responsible forsalmon decline, although the relative importance of eachthreat is still unclear. Dams, for example, are a significantcause of salmon decline in Maine, excluding salmon fromabout half of available habitat (Beland 1984). The negativeeffects of impoundments may be lessened by fish ladders(NatureServe Explorer 2002).

Land uses such as logging, agriculture, and urbandevelopment all negatively affect salmon by increasingsediment load in rivers and by increasing the extremes offlow intensity, thereby resulting in stronger floods andlower water levels in winter (Poff and Huryn 1998). Thelogging of riparian areas also increases stream temperaturesand reduces LWD (Poff and Huryn 1998). Finally,pollution and pesticides associated with forestry,agriculture and urban development cause chemicalpollution of salmon habitat (National Marine FisheriesService 1999).


The negative effects of salmon farming on wildAtlantic salmon have been extensive (U.S. Fish andWildlife Service 1999a). Researchers estimate that 25-40%of the salmon in the North Atlantic are now of farm origin(U.S. Fish and Wildlife Service 1999a). Escaped fish fromsalmon farms can disrupt redds, compete with wild salmonfor food and habitat, breed with wild salmon (causing adeterioration of genetic diversity; Nielson 1998), and serveas a vector of disease and parasites (McGinn 1999). Thethreat of salmon farming to wild populations was broughtto the fore by a recent ecological disaster in Maine: inDecember 2000, more than 100,000 farm-raised salmonwere accidentally released into Machias Bay after a stormdestroyed a cage containing over 170,000 one-year-oldsalmon. This escape was widely publicized as the largestever documented in the U.S. or the Canadian MaritimeProvinces.

Other anthropogenic threats to wild Atlantic salmonpopulations include acid rain (Parrish et al. 1998) andglobal warming (Hansen and Quinn 1998). Overfishing, anhistoric problem, has been largely eliminated as a threat toNorth American populations of this fish.


Umbrella: The Atlantic salmon is an umbrella speciesfor aquatic organisms that require clean water andrelatively undisturbed river habitat.

Habitat Quality Indicator: Atlantic salmon are excellentindicators of intact riverine ecosystems, in that they aresensitive to changes throughout the river systems theyinhabit, from headwaters to estuaries. If salmon arethriving in a river, it is more likely that the river isecologically healthy and its watershed minimally affectedby human disturbance. This species may also indicatebroader ecological health, as scientists believe the warming

of the North Atlantic due to global climate change may bereducing salmon populations.

Foundation: A foundation species is one that, likekeystone species, enriches ecosystem function in a uniqueand significant manner. Unlike keystone species,foundation species occur at much higher populationdensities. Although current numbers are low, thehistorically high numbers of Atlantic salmon in Mainequalify it as a foundation species, the only one used as afocal species in the Maine Wildlands Network.

Recommendations

The Atlantic salmon is in dire straits due to thecombined effects of many human activities. As a result,salmon recovery is complex and will require long-termrestoration and protection efforts. The U.S. Fish andWildlife Service is currently developing a recovery plan foreight rivers in Maine, most of which are in the eastern partof the state: the Dennys, Machias, East Machias,Narraguagus, Sheepscot, Pleasant, and Ducktrap rivers, andCove Brook. Because the umbrella value of this species is sohigh, however, salmon recovery should be pursued wellbeyond the areas proposed by U.S. Fish and WildlifeService, targeting as much available habitat as possible. Atminimum, conservation actions will need to:• Prevent dams from obstructing the upstream

migration of potential spawning adults and killingsmolts and adults in their downstream migration (thismay require dam removal).

• Restrict urban, agricultural, and silvicultural activitiesthat cause sedimentation and chemical and thermalcontamination of salmon rivers.

• Place a moratorium on salmon aquaculture or requiresalmon farms to a) prevent escapees of domestic salmoninto the wild and b) use salmon as genetically close tonative stocks as possible.

maine wildlands network visionforeman, john gallo, steve gatewood, harvey locke, chris mcgrory...

Documents