:02g(SStUt;09
fCm00:S:
Applications, 12(2), 2002, pp. 390-397
Ecological
the Ecological Society of America
:
Z
0
? i)tt
t2002by
LANDSCAPEECOLOGYAND FORESTMANAGEMENT:
DEVELOPINGAN EFFECTIVEPARTNERSHIP
STANBOUTINI3 AND DARYLLHEBERT2
Sciences, Universityof Alberta,
AlbertaPacific Forest IndustriesLtd., and Departmentof Biological
2E9
T6G
Canada
Edmonton,Alberta,
South Creston,British Columbia,Canada VOBIG2
2EncompassStrategicResourcesInc., RR#2599 Highway21,
their research applicable to
Abstract. Landscape ecologists have been eager to make
contributed to shaping the
has
ecology
landscape
how
examine
forest management. We
research in forested
ecology
Landscape
way forest management is currently practiced.
issues,
fragmentation
of
study
the
(1)
areas:
general
two
into
ecosystems can be divided
the
(2)
and
conservation;
species
on
which focuses on the effects of forest fragmentation
dynamics and the
patch
on
focuses
which
models,
projection
development of landscape
processes. Fragmentation issues
effects of spatial arrangement of patches on ecosystem
but research to date has overmanagers,
forest
of
minds
the
have become priorities in
We suggest that the
conservation.
emphasized the effects of landscape structure on species
change on
landscape
of
effects
threshold
of
study
the
research focus should move toward
in
conservation
species
on
the relative influence of habitat loss and habitat configuration
important
becoming
rapidly
are
models
projection
forest-dominated landscapes. Landscape
great promise as a means to bring
tools in forest management planning, and they hold
ability to produce future landscapes
The
together.
landscape ecologists and forest managers
these to landscapes produced by
compare
to
and
scenarios
management
under different
and scientists on understanding
managers
both
focus
to
natural disturbance regimes will help
and ecological processes.
features,
landscape
activities,
human
among
the key interactions
management;forest projection
Key words: conservationbiology; disturbancedynamics;forest and managers.
scientists
of
models;fragmentation;landscapeecology; partnerships
INTRODUCTION
The disciplinesof naturalresourcemanagementsuch
as fisheries, wildlife, and forestry have been eager to
adoptthe science of landscapeecology. And why not?
Landscapeecology appearsto bringa freshperspective
to age-old problemsby encouragingmanagersto expandthe scale at which solutionsare sought.Although
managershave been painfullyawarethatthe traditional
focus on local populationsstudiedat small spatialand
temporalscales is problematic,the tools and science
needed for a broaderperspective have been slow to
develop.This appearsto be changingas remotesensing
and GeographicalInformationSystems (GIS) technology allow us to obtain and analyze larger and larger
amountsof spatialdata.At the same time, spatialecology has begun to take shape throughconcepts related
to metapopulations,edge effects, patchdynamics,and
percolationtheory (FormanandGodron1986, Gardner
and O'Neill 1991, Gilpin and Hanski 1991).
Landscapeecologists have also been eager to seek
widerapplicationsfor theirworkby addressingapplied
Manuscript received 25 September 2000; revised 27 April
2001.
2001; accepted 9 May 2001; final version received 4 June
For reprintsof this Invited Feature, see footnote 1, p. 319.
:3E-mail: stan.boutin @ualberta.ca
problems.Muchof this workhas focusedon landscapes
thathave been highly alteredby humans,and it is no
coincidencethat issues of species conservationfigure
prominently.The workof landscapeecologists has become increasinglyinfluentialin conservationbiology
and,in fact, one could argue that the marriageof the
two disciplines is largely complete. Can the same be
said for landscapeecology and forest management?It
certainlyseems like a possibility, given that forestry
has immense potential to alter landscapesand biodiversityconservationwithinforestedlandscapeshas become a priority.In addition,forestry has a history of
"spatialconsciousness" broughtaboutby the need to
plan road development,cut sequences, and long-term
wood supply (Mladenoffand Baker 1999).
In this essay, we discuss how landscapeecology has
contributedto shapingcurrentpracticesin forest management.We thinkthatlandscapeecology has muchto
offer forest management,but this potentialhas yet to
be realized because both sides have not formed a real
partnershipto solve problems. The intention of this
paperis to providethe practitionerwith a synopsis of
some relevantkey researchthrustsin landscapeecology. In addition,we providethe researcherwith a practitioner's perspective on how landscape ecology research may be made more relevant to forest manage-
390
000SIt;Xi:tA
ment. We need to move from a situationin which the
scientist casts stones at the fortress of conventional
practice to one of true partnershipwhereby both the
scientist and practitionerengage in solving problems.
The paperis intendedfor landscapeecologists and forest managers who are serious about changing forest
practicesthroughthe applicationof new science. Landscape ecology's impact on forest managementwill be
assessed by the influencethatit has on changingactual
forest managementpractices(Hobbs 1997).
WHATIs LANDSCAPE
ECOLOGY?
Although the term "landscapeecology" is broadly
familiarto most forest managers,thereis considerable
confusion as to what it encompasses.It might be best
to start simply by defining a landscapeas a spatially
heterogeneous area (Turnerand Gardner1991). The
importantpoint here is the spatial natureof the heterogeneity:we view landscapeecologists as being primarily interestedin how spatial heterogeneityaffects
ecological processes. Landscapeshave emergentmeasurementsthat tend to be associatedwith the size, distribution, configuration,and connectivity of patches
(Weins et al. 1993), whereas Lidicker (1995) listed
emergentpropertiesof landscapessuch as edge effects,
interpatchfluxes of energy, nutrients,and organisms,
and stability of patch configuration.Confusion arises
when landscapeis used to designate a general spatial
scale or level of ecological organization.King (1997)
provides a particularlyclear discussion of why it is
importantto be cautious when using the term "landscape" in these contexts. Spatial scales (Bissonette
1997) and hierarchytheory (King 1997) are strongly
intermeshedwith landscapeecology, but for the purposes of this paper we will focus on how spatial heterogeneity affects ecological processes; the spatial
scale will be primarilyone of multiple forest stands
(patches).
CURRENT
FOREST
MANAGEMENT
AND
LANDSCAPE
ECOLOGY
Forest managementhas undergonea majorconceptual shift over the past 10 yearsthatcan be summarized
as a transitionfrom a focus on high-yield production
of fiber and selected wildlife species to supplying a
wide array of values, including maintenanceof biodiversity (Kohm and Franklin1997). Ecosystem management has emerged as the broad approachused to
achieve this objective, and a fundamentaltenet is that
success dependson managingat large spatial,andlong
temporal,scales (FEMAT 1993). These spatial scales
areoften equatedto landscapescales andthe immediate
assumptionis that landscapeecology is a fundamental
part of ecosystem management.However,working at
a particularspatial scale, by itself, is not enough to be
doing landsc-apeecology. Instead,the focus should be
t$tSBS:S
:wE;
on how spatial heterogeneity affects ecological processes ratherthan on scale per se.
Despite managing for a wider arrayof values, the
actual operationallevers available to forest managers
remain the same, namely, harvest rate, cutblock size
and shape, cut sequence, and silvicultural practices
(cutting and regenerationmethods). As stated by one
reviewer,managershave long been able to design future forests (landscapes)to meet the requirementsof a
groupof mills. In a similarfashion,they also have been
capable of simultaneouslyincorporatingthe needs of
a handful of wildlife species if the species' habitat
requirementsare known. If forest managershave been
practicing landscape planning, have they been using
landscapeecology to do so? In otherwords, how often
does our understandingof how spatial heterogeneity
affects ecological processes play into forest management decisions? We discuss two general lines of research in the landscape ecology literaturethat have
been influentialin forest management.We call these
the "forest fragmentation"and "patchdynamics"approaches. The former is focused on how forest fragmentation affects biodiversity conservation, whereas
the latteris focusedon patchdynamicsandspatialmodeling of habitat succession following disturbancein
forested landscapes.
FOREST
FRAGMENTATION
Habitatfragmentationoccurs when a specifichabitat
is successively divided into pieces to form a mosaic of
patches that vary in size7 shape, and connectedness.
Fragmentationis a common outcome of human resource development,particularlyin regions converted
to agriculture.A primaryfocus of landscapeecology
has been the studyof the persistenceof species in specific fragments (patches) and in the landscape as a
whole. Forest cutting, as it is currently practiced
throughoutthe world, tends to fragmentforest habitat
because complete standsare not harvestedin their entirety.Instead,cutblockstend to be of uniformsize and
shape, and relatively small relative to existing forest
patches. The result is a patchworkof cut-and-leave
forest familiarto anyone who has flown over actively
managed forest regions. Landscape ecologists have
been quick to warn forest managersthat such an approachmighthave negativeconsequencesfor the maintenance of forest species. Old-growthforest and its
associatedspecies have receivedthe most attentionbecause it is the older age classes that are most likely to
be truncatedby short rotation practices designed to
maximizetimberyields. The question,then, is whether
or not fragmentationcreated by forest harvesting is
significantenough to warranta change in forest practices andif so, whatshouldthe new practiceslook like?
The fragmentationconcept used in landscapeecology has two components.These are overallhabitatloss
(the total amountof suitablehabitatremovedfrom the
00!(;Aff0!ffErSSlU0i0!
fsSft ffff<Rfff0i;
X0X0!:
W00
20;
)
0ESSV 0
:C0Nft
f
00
0tAXt
!
W
f
landscape) and habitatconfiguration(patch size, isolation) (Haila 1986). Both habitatloss and changes in
configurationcould affect species presence (Andren
1994, Fahrig 1997), and it is importantto separatethe
two because forest managersmay focus on very different practices,dependingon the relative importance
of each component.There is little controversyin the
statementthat habitat loss means a reduction in the
average abundanceand overall distributionof species
using that habitat,and that once the amountof habitat
dropsbelow a critical threshold,the likelihood of species persistencebecomes zero (Lande 1987). If habitat
loss is the principledriverin species loss, forest managers could predicthow well theirpracticeswill maintain various species simply by predictingthe amount
of varioushabitattypes in futureforests. The question
of "How much is enough?" would still be important,
but we would not need landscapeecologists to provide
the answer.
Landscapeecologists however,have arguedthathabitat amountalone is not adequateto answerthis question; instead,we also need to considerpatchsize, configuration,andthe natureof the interveningmatrix(for
a review, see Fahrig2002). This added dimensionincreasesthe complexityof forestplanningconsiderably,
and before this change is warranted,landscapeecologists mustprovidestrongevidence thatmanaginglandscapc configurationmakes a significantdifference to
biodiversitymaintenanceover and above managingfor
habitatloss alone.
Forestfragmentationstudies have focused primarily
on the effect of patchsize andisolationon the presence
of selected species. Fewerstudieshave triedto separate
effects of habitat loss from spatial configuration.A
numberof approacheshave been taken,the most direct
being the experimentalfragmentationof a landscape,
whereby fragmentsof variable sizes and degrees of
isolation are created. These experiments have been
summarizedby Debinski and Holt (2000), who concluded that there was a surprisinglack of supportfor
the predictionthat smaller fragmentswould maintain
fewer species, have higher turnoverrates, and experience more severe edge effects.
On first impressions, it would seem that fragmentationexperimentscould provideforest managerswith
direct guidelines as to the appropriatepatch size for
remnantold-forestpatches.Althoughsome mightargue
that these experiments simulate conditions that will
arise as forest harvesting develops (small patches of
old forest in a matrixof unsuitablehabitat),this is only
partially true for the following reasons. The experiments follow a traditionalANOVA design; given the
effortrequired,treatmentrangeandreplicationarelimited (Debinski and Holt 2000). In most cases, selected
experimentalpatch sizes are much smaller than the
patches that would realisticallybe left by forest companies. Small patcheshold fewer species. This is well
documented, but the important issue is to determine if
thresholds in patch size effects exist, particularly within the range of patch sizes that forest companies are
capable of creating. Unfortunately, current fragmentation experiment designs have limited resolution at
these relevant scales.
In the case of fragmentation experiments in forested
regions, there is an additional complication. The experimentally created fragments are usually placed within a landscape that has considerable original habitat
remaining. This would not necessarily be the case as
the harvest rotation proceeds (Schmiegelow et al.
1997). Most fragmentation experiments have been followed for a short time relative to forest succession
scales. Immediately following the creation of the treatment, the matrix becomes clear-cut forest, possibly the
most inhospitable matrix for many old-forest dwellers.
However, unlike forest remnants within an agricultural
matrix, clearcuts regrow to forest, making the difference between the patch and matrix less obvious. Both
the nature of the matrix and the proportion of habitat
in the landscape are modifiers of patch size and isolation effects (Andren 1994). Consequently, current results of fragmentation experiments should not be extrapolated to what may happen in future forests.
All in all, fragmentation experiments create the best
conditions to test for the effects of patch size on the
ability of that patch to contain species of interest, but
the generality of the results will be limited. Apart from
the general rule that bigger is better, there are no other
prescriptions that managers could follow. Our statements are not meant as criticisms of the scientific merit
of the experiments. We are simply saying that their
design is not likely to give a forest manager direction
on how much old forest should be maintained and what
its configuration should be.
There are a growing number of observational fragmentation studies that take advantage of fragments created by human activities. Although these studies lack
the controlled design of fragmentation experiments
(McGarigal and Cushman 2002), they allow a much
broader range of patch sizes, matrix types, and configurations to be explored. Much of the work involves
birds as study organisms, and the results have been
thoroughly reviewed a number of times (Andren 1994,
Freemark et al. 1995). Although the details vary somewhat, the general conclusion is that smaller or more
isolated fragments hold fewer species. Most observational studies have been conducted in landscapes where
natural forests exist as small patches surrounded by
agriculture or tended conifer forests. McIntyre and
Hobbs (1999) have operationally defined fragmented
landscapes as those where natural habitat has dropped
below 60% of the landscape; and relict landscapes are
those with < 10Woof original habitat. Much of the work
on fragmentation has been conducted in relict landscapes where edge effects, patch size, and patch con-
i f;;w
CwSr;
:00gA#Bt0000SAbM0
figurationhave strongeffects on species persistencein
the remaininghabitat(Saunderset al. 1991). McIntyre
and Hobbs (1999) suggest that forestry operationsin
nativeforests createexamplesof variegatedlandscapes
wherenaturalhabitatstill represents>60% of the landscape. As researchersbegin to expand the range of
landscapesstudied to include the extensive forests of
western and northernNorth America, patch size and
isolation effects and edge effects are less pronounced
or nonexistent (Andren 1994, Freemarket al. 1995,
McGarigaland McComb 1995, Drapeauet al. 2000,
Schmiegelow and Monkkonen 2002). These results
raise the very real possibility that the fragmentation
issues that have preoccupied landscape ecologists
working in relict landscapesmay not be a priorityin
the variegated landscapes created by forest planning
(McIntyreand Hobbs 1999).
Fahrig (1997) has also suggested that the emphasis
on habitatconfigurationis "misplaced"and that conservationefforts shouldfocus on reducinghabitatloss.
She used a spatiallyexplicit populationmodel to show
that total habitatamounthad a far greaterinfluenceon
species persistence in landscapesthan did configuration. Configurationhad little effect as long as suitable
habitatmade up >20% of the landscape.Some observational studies have tried to de-couple the effects of
configurationfromthose of habitatloss. Andren(1994)
reviewed studies of birds and mammalsand concluded
that the total area of suitable habitat was of greater
importancethan spatial configuration,particularlyin
landscapes with > 30% of suitable habitat left. The
greaterimportanceof habitatamountrelative to configurationseems to be a consistentpattern,at least for
forest birds (McGarigaland McComb 1995, Trzcinski
et al. 1999, Drapeauet al. 2000, Flatheret al. cited in
Fahrig2002). However,Villardet al. (1999) foundthat
fragmentationand habitat amount had roughly equal
influence in eastern deciduous forests within an agriculturalmatrix.
Therehas been one otherapproachto studyingfragmentationeffects. This involves the use of experimental model systems (EMS), whereby landscapesare artificially created at scales that researcherscan effectively replicate (Ims et al. 1993, Wolff et al. 1997).
Unlike large-scale manipulativeexperimentsor comparativemensurativeexperiments,EMS actuallystudy
how spatial heterogeneityaffects ecological processes
as opposed to inferring process from patterns (see
McGarigal and Cushman 2002). These studies have
revealed some interestingeffects of spatialconfiguration on population processes, but their relevance to
forest managementremainsto be determined.We are
skeptical of their utility for two reasons. Given that
spatial scale appearsto be so importantto landscape
ecology and forest management,one cannot assume
thatit will be straightforward
to "scale up" fromEMS
to forest landscapes. Secondly, given the importance
:0W0An
itE0000thSSlEg;00000
of the relationshipbetweenthe patchmatrixandanimal
movement,it is not clearhow resultsfroman artificially
createdmatrixcan be appliedto the dynamicmatrices
found in forested systems.
FRAGMENTATION
STUDIES
ANDFOREST
MANAGEMENT
PRACTICES
Based on results of the fragmentationresearchjust
outlined, we suggest the following operationalguide
for forest managersworking in regions where forests
of differentage will remainas the predominantcover
type in the region. Forest planning to conserve biodiversity should focus on maintaininghabitatamount;
there is little need to take configuration(patch size,
corridors)into accountunless habitatsof interestdrop
below 20-30% of the landscape.Monkkonenand Reunanen (1999) advise against managingaccordingto a
thresholdrule because critical thresholdsare species
specific and the 20-30% thresholdmay be an underestimatefor many species. We are not suggesting that
this thresholdshould serve as a guide for the amount
of habitatrequiredto maintaintarget species. Rather,
we suggest thatcurrentinformationsupportsthe working hypothesis that forest managersneed not worry
aboutpatchconfiguration
untilhabitatloss reaches7080%. In other words, the amountof habitatshould be
the primarydriver in forest planning, and it is only
when projectedloss of habitatis substantialthat configurationshould also be considered.
Given these recommendations,we provide a cautionarynote. Most forest landscapesare more complicated than the dichotomoushabitatand matrixdesign
of landscapemodels andexperiments.In particular,forests subject to large-scale naturaldisturbanceevents
are naturallyfragmented,and old forest may naturally
comprise <20% of a landscape(Bergeronand Harvey
1997). Does this mean that landscapeconfigurationeffects are always present, or does the amount of old
foresthave to dropto <30% of "natural"levels before
configurationbecomes important?Currentlandscape
models are of little help in this case because the predicted effects depend on how individual species respond to the landscapematrix(Fahrig2002). The key
parameterssimply have not been measuredin variegated forest systems and this continues to representa
major challenge to landscape ecologists. It is clear,
however,that any practicethat makes the patchmatrix
more hospitable will greatly reduce the potential for
fragmentationeffects associatedwith individualmovementsbetweenpatches(Fahrig2002). So-called "New
Forestry" approachesthat leave some forest structure
on newly cut areas may hold promise in this regard
(FEMAT1993, Franklinet al. 1997).
To summarize,thereis little evidence to suggest that
forest managersshouldplace a priorityon habitatconfigurationwhen planning for conservation of biodiversity in landscapeswhere forests will comprise the
fffeN:e Sef!tfffft00Ul00
0tS
fff00 S00ff!ffF0 7: 000SU
jO00!;j
j0 jUSSu
t tSiV0
ly! f
majorityof the landscapein the future.Instead,managinghabitatloss alone is probablythe most reasonable
guide. However,modelsandempiricaldatasuggestthat
there is a thresholdrelationshipbetween habitatloss
and effects of configuration.It remainsunclear,however,whatthe exact thresholdlevel is, but it is probably
below 50%. We suggest that researchersand forest
managers identify these thresholds under conditions
that are likely to exist in futuremanagedforests.
LANDSCAPE
EFFECTS
ON SPECIES
HABITAT
USE
If forest managersare to conserve selected species
by maintainingadequateamountsof habitat,speciesspecific habitat requirementsneed to be identified.
Landscapeecology has addedanotherdimensionto this
process by raising the possibility that spatial configurationmightaffect habitatuse. Some species havebeen
identifiedas "interior"specialists, or as having minimum patch size requirements(Whitcombet al. 1981,
Freemarkand Merriam1986, Hansen et al. 1993; see
also Villard 1998). This has two implicationsfor forest
management.First, projectionsof habitatsupply for a
particularspecies would have to be readjustedto exclude patchesbelow a minimumsize. Second, cutting
plans would need to be designed to create patches of
adequatesize and shape to meet the requirementsof
"landscape-sensitive"species. These adjustmentsare
not substantive,given that patch size is commonlybeing trackedin most GIS forest inventories.
Along with patch size and configuration,another
landscape variable that can affect habitat use is the
juxtapositionof two habitatpatches. Juxtapositionof
habitats has received relatively little attention from
landscapeecologists, but it is interestingto note that
much of the traditionaldesign of forest cutblock size
and shape was actually driven by the perceived need
to provide the appropriatejuxtapo-sitionof forage and
cover for ungulates (Rempel et al. 1997). It is likely
that more research will reveal that some species are
associatedwith patch types arrangedin a certainfashion. The questionwill then be whetherthis addedhabitat requirementwouldbecome an additionalconstraint
on cutblock layout.
Accommodatingthe local and landscapehabitatrequirementsof a limited numberof species while maintaining wood supply is certainly possible, as long as
the numberof species considereddoes not become too
large. However, Monkkonen and Reunanen (1999)
point out thatit is impracticalto generalizeaboutlandscape effects on species becauseof differencesin scale,
life history characteristics,and responsesto landscape
matrix. They recommend "applying case-by-case information"as a result.Althoughthis approachmay be
possible for conservationof selected species, we think
that it is unrealisticfor the broaderobjective of maintenanceof biodiversity.This "fine-filter"approachinevitably leads to makingprescriptivedecisions thatfa-
iSSSig04tSStU!0000uRVE0S
S;X005
0' VSfE 00 0;
t:
vor one species over another.As an example, Bender
et al. (1998).found that the effect of patch size on
populationsize was negative for interior,but positive
for edge, bird species. The effects differed between
migratoryand residentspecies and betweencarnivores
andherbivores.It wouldbe impossibleto managethese
species on a case-by-case basis. Landscapeecologists
mustseek generalguidesfor planningforestlandscapes
if they wish to change forest practices.
PATCH
DYNAMICS
ANDFOREST
LANDSCAPE
PROJECTION
MODELS
The study of fragmentationeffects in managedforests has tended to take a somewhatstatic view of the
spatial nature of forest patches. For example, landscapes are modeled as patches within a matrix, with
the spatial location, patches, and matrices remaining
constant.This seems perfectlyreasonablein relictlandscapes, but managedforest landscapesare much more
dynamic,and it is the dynamicsof patch mosaics per
se that may hold the key to maintenanceof species
diversity (Pickett and Rogers 1997). Landscapeecology has played a majorrole in our ability to describe
the spatial arrangementof importantelements at the
regional scales necessary for forest management(Perera and Euler 2000). It also has the potential to help
us understandthe reciprocaleffects of spatial pattern
on ecological processes (Pickettand Cadenasso1995).
Forestershave always had some appreciationfor the
large spatial, and long temporal,natureof their business. Forest inventories are essential for calculating
availablewood supply,and the spatial location of that
wood supply is crucial for determiningroad construction andcut sequence.However,untilrecently,the spatial map of forest inventorywas largely a static snapshot thatwas updatedat regularintervalsas forestharvesting and planting proceeded. There was really no
way to projectcurrentpracticesinto the futureto catch
a glimpse of what the forest would look like some 30100 yr into the future, nor were there tools to make
rapidcomparisonsof landscapemetricsbetweenforest
landscapessubjectedto differentpractices.This is now
feasible and landscapeecology has had a majorrole to
play in this development.
Spatially explicit landscape projection models are
rapidlybecoming partof every forest manager'stoolkit. There are now many versions of these types of
models, and Mladenoff and Baker (1999) provide a
good summaryof the evolution of their development.
The models are intended to take a spatially explicit
currentvegetationinventoryand projectit into the future. The rules for doing this are drawnfrom an understandingof vegetationsuccession,forestharvesting
and silvicultureplans, andnaturaldisturbancessuch as
fire and insect outbreaks.There are majorchallenges
to producingrealisticlandscapes.These relateto "scaling up" fromthe individualtree or standlevel to broad
w0
::0+bi
landscapes,
and to capturing the
spatial nature of many
ecological processes.
Many forest processes
to involve a spatial
are likely
are beginning to component, and a number of studies
explore this aspect
me 1994, Roland
(Turner and Romand Taylor 1997,
Greene and Johnson
1999, Li 2000). The
when spatial pattern challenge will be in determining
truly
because its inclusion in matters (Turner et al. 1995),
increasescomputing timelandscape projection models
Landscape projection immensely.
models are strategic in
and,as such, their
nature
utility is not in tracking
ina landscape.
exact changes
Rather, they provide a
howforest
general guide to
landscapes might look, "on
average," under
different
management practices. As
such, they provide
avitaltool for
planning at large scales
and
management
scenarios to be projected allow various
into the future.
These
hypothetical landscapes can
then
be compared
using
a wide range of
descriptive metrics
and
Marks 1995). Doing so
(McGarigal
quickly reveals how
cutting
forestpractices act to change
landscape elements such
as
patchsize, amount of
edge,
and degree of
edness
connect(Franklin and Forman
1987, Spies et al.
Wallin
et al. 1994). In
1994,
addition, this
toformulation
of hypotheses as to approach also leads
how these practices
actually
affect ecological
processes.
be
However, it should
stressedthat many of the
proposed
spatial effects on
ecological
processes are still at the
hypothesis stage. The
challenge
to landscape
ecologists is to find
to
test
creative ways
these hypotheses at
the large spatial
which
scales upon
they are proposed to
operate.
DIFFERENT
FORESTMANAGEMENT
SCENARIOS
PRODUCE
DIFFERENT
LANDSCAPE
PATTERNS:
WHATDOES IT
MEAN?
Landscape
projection models
produce future landscapes
that have clearly
different landscape
pending
metrics deon the forest
management scenario
The
question
employed.
becomes "What do
forest managers do
with
thisinformation?"
In any type of
resource management,
the key to success
is to know the
relationship
functional
between the values being
the
managed for and
conditions
that managers
actually manipulate.
lowing
on this, landscape
Folecologists should work
forest
with
managers
to determine how
spatial heterogeneity
affects
ecological processes, and
how forest management
might affect these
functional
Armed
withthis
relationships.
information, it should then
for
be
one
to functionally
link the values for possible
manage
which we
andthe
management practices
able.
actually availUnfortunately,
current research in
ogy
is at the
landscape ecolnot
yet
stage that allows us
to do this.
Fragmentation
and patch
dynamics studies remain
largely
at
the pattern
description stage; the
tional
crucial funcrelationships
between spatial pattern
logical
and ecoprocesses
and the effects of
forest practices on
these
processes,
remain unknown.
How
shouldforest
management proceed?
Swanson
::!W00
et al. (1993)
proposedan approachthat
a solution. It is
might provide
based on designing
forest management
approachesto maintainthe
rangeof natural
inhabitattypes that
variability
are created by the
physicalfactors and
interactionof
disturbanceregimes. Natural
turbances
dissuch as fire have a
characteristicfrequency,
size,and severity
within a given region,
turn,
and this, in
producesthe forest
vegetation and age distributions
observedon a landscape.
The natural
regime
also plays a major
disturbance
role
in creatingthe
pattern
of patches, edges,
and connectivity natural
alandscape.
Managinglandscapeeffects presenton
underthis approach
becomesa matterof trying
to patternlandscapes
with
human activities after
those created by
disturbances
natural
.
Tryingto maintainthe
rangeof
landscapes
is a "coarse-filter" naturalvariabilityin
approachto management
basedon a key
ed
tothe landscapesassumptionthatspecies areadaptcreatedby natural
gimes.
disturbancereThe emphasisis on
large-scalegeneralpatterns
ofhabitat
mosaics ratherthanon
meetingthe
needs
of individual
species (Hunter 1993).fine-filter
proach
The apremainsto be tested, but
it has gained
erable
considsupport,particularlyin
forests where large disturbances
such as fire are
common features (Hunter
1993,
Bergeron and Harvey
Perera
and Baldwin 2000). 1997, Angelstam 1998,
It is attractive
does
notrequirea detailed
because it
understandingof how landscape
features affect processes.
Instead,
isguided
by the comparison
of managed management
those
landscapesto
created by historical
disturbancepatterns.This
approach
allows both the manager
and the researcher
toidentify
areas that require
immediate
the
case
attention.In
of the manager,this
might
new
cut
patternsto more closely mean designing
match disturbance
patterns;
in the case of the
researcher,it helps to direct
research
prioritiesto key landscape
parametersandprocesses.
This raises an
importantpoint. Forest
ment
managepractices
are continually
undergoingchange and
landscape
ecologists mustbe in tune
with those changes.
Otherwise,
the experimentsthat
makeperfectsense
now
will
seem trivial under
future
To
summarize,
we see a growing regimes.
forest
managers
and landscape interactionbetween
ecologists in the area
of
understanding
how ecological
processesareaffected
by
spatial
heterogeneity.There is a great
to
foster
opportunity
this
interactionby
ment
of landscape focusing on the developrealistic
projectiontools. At present,
these
tools
still lack key
functional
tween
relationshipsbespatial
heterogeneityand ecological
Asthese
processes.
relationshipsare developed,
an interim approach
may
be to use the range
of natural
in
patch created
variability
dynamics
by disturbanceand
sion
as
a
succesmanagement
a common guide. Doing so would provide
frameworkthat managers
and landscape
ecologists
can use to determine
priorities.
:n
GE
t:S:S?
t00
CONCLUSIONS
of "fraghave been critical of the contribution
We
is due
This
studies to forest management.
mentation"
this,
By
landscape.
real
to a failure to see the
partially
ffor
management
forest
of
that the implications
mean
we
current
must be considered within realistic
biodiversity
of Scandinavia
forest landscapes. The forests
future
and
the future of
into
give us a clear glimpse
Australia
and
northwestern
and
Canada
of
the extensive forests
what
studies
States could look like, and fragmentation under
United
happen
to
a clear indication of what is likely
give
opportunity
conditions. However, there is immense
these
future
different
very
create
practices now to
change
to
work
must
ecologists
conditions. Landscape
forest
that
practices
with forest managers to develop
closely
is
fragmentation an
notcreate future forests where
do
ecologists
landscape
to
We think that the challenge
issue.
is to deterforests
managed
in
fragmentation
studying
change where
possible thresholds of landscape
mine
compoimportant
configuration becomes an
landscape
If such
habitat.
of
loss
over and above the absolute
nent
begin
then
can
managers
can be identified,
thresholds
accordingly.
plantheir activities
to
contribution of landWehave been less critical of the
of forest landscape
ecology to the development
scape
that such models
say
to
not
models. This is
projection
forest landplanning
to
solutions
provided the
have
fealandscape
The important aspects of how
scapes.
scibe
to
yet
have
affect ecological processes
tures
projection
landscape
entificallyvalidated. However,
managers and
provides a framework-to focus
modeling
spatial interactions at the
scientistson the important
scales. We believe that
relevantspatial and temporal
managers can benforest
landscape ecologists and
both
disturbance
natural
historical
from considering how
efit
act to shape
features
physical
regimesand underlying
time. The patterns crelandscapepatterns in space and
for designing
template
atedshould provide the basic
understandfor
and
scales,
forestpractices at landscape
to the interplay
adapted
be
may
inghow organisms
and landscape features.
betweenecological processes
ACKNOWLEDGMENTS
Villard for organizing this
We are grateful to Marc-Andre
helpful comments on the
special feature and for providing Gordon Baskerville for
thank
manuscript. We also wish to
into forest management issues.
sharing some of his insights
CITED
LITERATURE
fragmentation on birds
habitat
of
Effects
Andren, H. 1994.
proportions of
different
with
and mammals in landscapes
71:355-364.
suitable habitat: a review. Oikos and restoring biodiverAngelstam, P. K. 1998. Maintaining developing natural disby
sity in European boreal forests
of Vegetation Science 9:593turbance regimes. Journal
602.
and L. Fahrig. 1998. Habitat
Bender, D. J., T. A. Contreras, meta-analysis of the patch
a
decline:
population
loss and
.
size effect. Ecology 79:5 17-533
1997. Basing silviculture
Bergeron, Y., and B. D. Harvey.
an approach applied to the
naturalecosystem dynamics:
on
Quebec. Forest Ecolboreal mixed-wood forest of
southern
92:235-242.
Management
and
ogy
ecological properties:
J. A. 1997. Scale-sensitive
Bissonette,
Pages 3-31 in J. A.
meaning.
context, current
historical
ecology: effects
editor. Wildlife and landscape
Bissonette,
New York, New
Springer-Verlag,
scale.
ofpatternand
USA.
York,
A survey and overD. M., and R. D. Holt. 2000.
Debinski,
Conservation
of habitat fragmentation experiments.
view
14:342-355.
Biology
Savard, Y. Bergeron,
P., A. Leduc, J.-F. Giroux, J.-P.
Drapeau,
disturbances
Landscape-scale
W. L. Vickery. 2000.
and
of boreal mixed-wood
communities
bird
in
changes
and
Ecological Monographs 70:423-444. loss and fragforests.
habitat
L. 1997. Relative effects of
Fahrig,
of Wildlife
on population extinction. Journal
mentation
61:603-610.
Management
on the exL. 2002. Effect of habitat fragmentation
Fahrig,
12:
Applications
Ecological
threshold: a synthesis.
tinction
346-353.
Assessment Team).
(Forest Ecosystem Management
FEMAT
an ecological, ecomanagement:
1993.Forest ecosystem
of the Forest EcoReport
nomic,and social assessment.
U.S. Government
Team.
systemManagement Assessment USA.
D.C.,
Washington,
PrintingOffice,
Landscape ecology.
R. T. T., and M. Godron. 1986.
Forman,
USA.
York,
JohnWiley, New York, New
Thornburgh, and J. C.
J. F., D. R. Berg, D. A.
Franklin,
approaches to
silvicultural
Alternative
Tappeiner.1997.
systems. Pagharvest
retention
timberharvesting: variable
Creating
editors.
Franklin,
F.
J.
es111-139 in K. Kohm and
Island Press, Washington,
century.
21st
the
for
aforestry
D.C., USA.
Creating landscape
J. F., and R. T. T. Forman. 1987.
Franklin,
consequences and
ecological
patterns by forest cutting:
1: 19-28.
Ecology
Landscape
.
principles
and J. R. Probst.
K. E., J. B. Dunning, S. J. Hejl,
Freemark,
research, confor
perspective
1995. A landscape ecology
T. E. Martin
in
381421
Pages
servation, and management.
management of neoand
Ecology
editors.
Finch,
and D. M.
synthesis and review of critical
tropical migratory birds, a
Press, New York, New York,
issues. Oxford University
USA.
of
Merriam. 1986. Importance
Freemark,K. E., and H. G.
in temassemblages
bird
to
area and habitat heterogeneity
Conservation 36: 115perate forest fragments. Biological
141.
O'Neill. 1991. Pattern, process,
Gardner,R. H., and R. V.
for
of neutral landscape models
use
the
and predictability:
and R.
Turner
G.
M.
in
289-307
landscape analysis. Pages
methods in landscape
H. Gardner, editors. Quantitative
New York, USA.
York,
New
ecology. Springer-Verlag,
Metapopulation
1991.
editors.
Hanski,
Gilpin, M. E., and I.
and emperical investigations.
dynamics: theoretical models
Academic Press, London, UK.
1999. Modelling recruitGreene, D. F., and E. A. Johnson.
banksiana, and Picea
Pinus
tremuloides,
ment of Populus
boreal forest.
mixedwood
the
in
mariana following fire
29:462-473.
Research
Canadian Journal of Forest
land birds in forest fragments:
Haila, J. 1986. North European
315-319 in J. Verner, M.
Pages
effects.
evidence for area
Wildllfe 2000: modeditors.
Ralph,
L. Morrison, and C. J.
vertebrates. Uniterrestrial
of
eling habitat relationships
Madison, Wisconsin, USA.
Press,
Wisconsin
of
versity
B. Marks, and D. L. Urban.
Hansen, A. J., S. L. Garman,
vertebrate diversity
managing
for
1993. An approach
Applications 3:
Ecological
across multiple-use landscapes.
481-496.
and the future of landHobbs, R. J. 1997. Future landscapes
W
::!S:2
1995. Landscape
S::g0t:
Cadenasso.
T. A., and M. L.
S.
systems. SciPickett,
ecological
in
37:
Planning
and Urban
spatial heterogeneity
ecology:
Landscape Ecology
ecology.
scape
269:331-334.
Patch dynamics:
ence
H. Rogers. 1997.
as spatial
1-9.
T. A., and
S. K.
function. Pagdisturbance regimes Conserand
Pickett,
Natural
structure
L., Jr.1993.
M.
of landscape
Biological
Hunter,
and landforests.
transformation
Wildlife
the
boreal
editor.
for managing
models
in J. A. Bissonette,
Springer-Ver101-127
scale.
es
pattern and
65:115-120.
Predicting space
vation
ecology:effects of USA.
scape
and P. Wegge. 1993. voles (Microtus
York,
J. Rolstad,
New
R. A.,
can
York,
Ims,
M. J. Gluck.
lag, New
R. Rodgers, and
to habitat fragmentation:model system (EMS)
responses
effects
use
S., P. C. Elke, A.
R.
Rempel,
experimental
natural-disturbance
and
serve as an
in boreal forest.
WiIdlife
oeconomus)
of
Timber-management
urogallus)
Journal
1997.
evaluation.
grouse (Tetrao
cappercaillie
habitat: landscape
for
moose
on
Conservation 63:261-268.a guide to system struc61:517-524.
species
Biological
Management
theory:
1997. Insect parasitoid
1997.
W.
A. Hierarchy Pages 185-212 in J. A. BisKing,
J.,and P. D. Taylor. different spatial scales. Nature
Roland,
at
wildlife
for biologists. landscape ecology: effects of
ture
to forest structure
respond
editor. Wildlife and
1991.
sonette,
New York, New York,
386:710-713.
and C. R. Margules. a reand scale. Springer-Verlag,
D. A., R. J. Hobbs,
pattern
fragmentation:
Saunders,
ecosystem
a
of
consequences
USA.
editors. 1997. Creating
Biological
Biology 5:18-32.
K.A., and J. F. Franklin, Island Press, Washington,
Conservation
Kohm,
S. J. Hannon.
view.
century.
21st
C. S. Machtans, and
the
A.,
K.
for
F.
forestry
fragmentation?
Schmiegelow,
to forest
USA.
responses.
Are boreal birds resilient
D.C.,
in demographic mod1997.
of short-term community
Extinction thresholds Naturalist 130:624study
1987.
R.
Lande,
experimental
An
American
populations.
Habitat
territorial
78: 1914- 1932.
of
els
Ecology
Monkkonen. 2002.
perF. K. A., and M.
with reference
avian
635.
simulation
Schmiegelow,
landscapes:
and their
in dynamic
L. Euler,
D.
fragmentation
2000.
Applications
C.Fire regimes
and
Perera,
Li,
H.
loss
A.
Ecological
Pages 115-140 in
from the boreal forest.
Ontario.
of a managed landto
spectives
editors. Ecology
Onin
Thompson,
D.
1994. DyI.
and
12:375-389.
of forest landscapes
and G. A. Bradshaw.
patterns and processes Columbia Press, Vancouver,
T. A., W. J. Ripple, managed coniferous forest landscape:
Spies,
University of British
and pattern of a
4:555-568.
tario.
namics
Ecological Applications and J. H. Cissel.
something
Columbia, USA.
in
British
concept:
scapeOregan.
Wallin,
landscape
O.
The
D.
W. Z., Jr. 1995.
F. J., F. A. Jones,
J. Lidicker, Jr.,
for ecosystem
Lidicker,
Swanson,
Pages 3-19 in W.
Univariability implications and P. S. BourNatural
ecology.
andsomething new.
old
1993.
Jensen
E.
mammalian
M.
in
Pages 89-103 in
Landscape approaches
Minnesota, USA.
health assessment.
management.
editor.
forest ecosystem
Press, Minneapolis,
Eastside
evaluMinnesota
and appliof
editors.
Comparative
versity
geron,
principles
2002.
management:
K., and S. Cushman.
II. Ecosystem
Oregon, USA.
study of habitat
Volume
McGarigal,
the
to
Portland,
USDA Forest Service, G. Merriam. 1999. Indeof experimental approaches Applications 12:335cations.
ation
and
the
effects. Ecological
M. K., L. Fahrig,
fragmentation
Trzcinski,
fragmentation on
forest cover and
of
spatial
Applicaeffects
pendent
345.
birds. Ecological
1995. FRAGSTATS:
strucof forest breeding
K., and B. J. Marks.
distribution
McGarigal,
for quantifying landscape Report
program
analysis
9:586-593.
Technical
tions
pattern
editors. 1991. Quantitative
Service General
M. G., and R. H. Gardner,Springer-Verlag, New York,
Turner,
ture.USDA Forest
USA.
Portland, Oregon,
in landscape ecology.
Relationships bemethods
PNW-GTR-351.
W. C. McComb. 1995.
USA.
and
Oregon
1995. EcoK.,
York,
the
in
New
McGarigal,
and R. V. O'Neill.
and breeding birds
M. G., R. H. Gardner, scales: ecosystems and landlandscape structure
Turner,
65:235-260.
tween
Monographs
at broad
conlogical dynamics
CoastRange. Ecological
A framework for
(Supplement):S29-S34.
R. Hobbs. 1999.
and
BioScience
relevance
S.,
its
Landscape dynamics
scapes.
and
McIntyre,
W. H. Romme.1994.
effects on landscapes
and
G.,
human
Biology
M.
Ecology 9:59-77.
Turner,
ceptualizing
Conservation
Landscape
models.
research
incrown fire ecosystems.
species, edge-avoidtomanagement and
Spatial
M.-A. 1998. On forest-interiorin avian conservation.
1999.
Villard,
13:1282-1292.
editors.
Baker,
and dogmas
D. J., and W. L.
ance, area sensitivity,
approaches and apMladenoff,
Fraglandscape change:
New
forest
115:801-805.
of
York,
Auk
New
modeling
and G. Merriam. l999.
University Press,
M.-A., M. K. Trzcinski, birds: relative influence of
Villard,
plications. Cambridge
on forest
mentation effects
on landscape occupancy.
York,USA.
Development of forest
W. L. Baker. 1999.
J.
cover and configuration
D.
and
in
D.,
woodland
1-13
Mladenoff,
13:774-783.
approaches. Pages modeling of
Conservation Biology
Marks.1994. Landscape
and landscape modeling
Spatial
Swanson, and B.
J.
F.
L. Baker, editors.
O.,
W.
D.
generation rules:
and
Wallin,
applications.
pattern
and
Mladenoff
in
change: approaches New York, USA.
response to changes Ecological Applications 4:
pattern
forest landscape
York,
in forestry.
Press, New
land-use legacies
Cambridge University Reunanen. 1999. On critical threshIms.
P.
and
M.,
569-580.
perspective.
Horne, and R. A.
Monkkonen,
a management
OiC. Stenseth, B. Van
N.
A.,
connectivity:
ecology.
J.
Weins,
and landscape
olds in landscape
mechanisms
Ecological
1993.
Oikos 84:302-305.
2000. Spatial patterns
C. Robbins,
D. J. B. Baldwin.
74-99
kos 66: 369-380.
M. K. Klimkiewicz,
Perera,A. H., and
of Ontario. Pages
of forest
landscape
R. F., J. F. Lynch,
Effects
forest
editors.
Whitcomb,
1981.
managed
in the
and D. Bystrak.
and I. D. Thompson,
forest.
Euler,
L.
deciduous
D.
L..Whitcomb,
B.
processes
eastern
Perera,
in A. H.
patterns and
avifauna of the
on
editors,
landscape:
Sharpe,
M.
Comanaged
fragmentation
D.
of British
Ecology of a
R. L. Burgess and
landscapes.
in Ontario. University Canada.
Pages 125-205 in
in man-dominated
of forest landscapes
Columbia,
dynamics
British
island
USA.
Forest
in
Vancouver,
York,
ecology
New
Press,
lumbia
Landscape
New York,
1997. Effects
Springer-Verlag,
D. L. Euler. 2000.
H.
and W. D. Edge.
Pages 3-11 in A.
Perera, A. H., and
and deO., E. M. Schauber,
J.
behavior
introduction.
an
the
Wolff,
on
Ecology
11:
forest management: and I. D. Thompson, editors.
and fragmentation
loss
Biology
habitat
of
Conservation
forest
Perera, D. L. Euler,
and processes of
of gray-tailed voles.
patterns
mography
of a managed landscape: University of British Columbia
945-965.
landscapes in Ontario. Columbia, Canada.
British
Press, Vancouver,
d+w0!9