Natural disturbance emulation in boreal forest ecosystem management — theories, strategies, and a comparison with conventional even-aged management 1 1 This article is one of a selection of papers from the 7th International Conference on Disturbance Dynamics in Boreal Forests.

Abstract

Natural disturbance emulation (NDE) has been proposed as a general approach to ecologically sustainable forest management. We reviewed the concepts, theories, and strategies related to NDE in boreal forest management. We also reviewed publications that discussed NDE in the boreal forest in general and those that specifically compared NDE-based management with conventional even-aged management. The papers generally focused on northern North America and landscape-scale wildfire as the main disturbance factor, whereas information from Eurasia was exclusively theoretical. Within this limited scope, NDE was generally found to have a positive effect on biodiversity in terms of forest structure and species diversity when compared with conventional even-aged management. Studies on timber supply and social implications of NDE were so few that they preclude generalizations. We conclude that the ecological and economic performance of NDE as a management approach still remains poorly examined. To advance the development of NDE, particular attention should be given to (1) augmenting the knowledge base on natural range of variability of unmanaged forest ecosystems and evaluating the validity of this information in a changing climate, (2) fostering multidisciplinary research with better integration of ecological theory to both integrative and analytical research on NDE, and (3) better integration of socioeconomic concerns, adaptive management schemes, and international collaboration into NDE initiatives.

Full text

REVIEW / SYNTHÈSE
Natural disturbance emulation in boreal forest
ecosystem management —theories, strategies,
and a comparison with conventional even-aged
management1
Timo Kuuluvainen and Russell Grenfell
Abstract: Natural disturbance emulation (NDE) has been proposed as a general approach to ecologically sustainable forest
management. We reviewed the concepts, theories, and strategies related to NDE in boreal forest management. We also re-
viewed publications that discussed NDE in the boreal forest in general and those that specifically compared NDE-based
management with conventional even-aged management. The papers generally focused on northern North America and land-
scape-scale wildfire as the main disturbance factor, whereas information from Eurasia was exclusively theoretical. Within
this limited scope, NDE was generally found to have a positive effect on biodiversity in terms of forest structure and species
diversity when compared with conventional even-aged management. Studies on timber supply and social implications of
NDE were so few that they preclude generalizations. We conclude that the ecological and economic performance of NDE as
a management approach still remains poorly examined. To advance the development of NDE, particular attention should be
given to (1) augmenting the knowledge base on natural range of variability of unmanaged forest ecosystems and evaluating
the validity of this information in a changing climate, (2) fostering multidisciplinary research with better integration of eco-
logical theory to both integrative and analytical research on NDE, and (3) better integration of socioeconomic concerns,
adaptive management schemes, and international collaboration into NDE initiatives.
Résumé : L’imitation des perturbations naturelles (IPN) a été proposée comme approche générale pour un aménagement fo-
restier écologiquement durable. Nous avons examiné les concepts, théories et stratégies reliés à l’IPN pour l’aménagement
des forêts boréales. Nous avons aussi examiné les publications qui ont traité de façon générale de l’IPN dans la forêt boréale
et celles qui ont comparé spécifiquement l’aménagement basé sur l’IPN à l’aménagement équienne conventionnel. Les arti-
cles ont généralement mis l’accent sur le nord de l’Amérique du Nord et les feux à l’échelle du paysage comme principal
facteur de perturbation alors que l’information provenant de l’Eurasie était exclusivement théorique. Dans ce champ d’appli-
cation limité, l’IPN a généralement eu un effet positif sur la biodiversité en termes de structure forestière et de diversité des
espèces comparativement à l’aménagement équienne conventionnel. Les études sur l’approvisionnement en bois et les impli-
cations sociales liées à l’IPN sont si peu nombreuses qu’elles excluent toute généralisation. Nous concluons que les résultats
d’ordre écologique et économique que produit l’IPN comme approche d’aménagement demeurent encore faiblement étudiés.
Pour faire progresser le développement de l’IPN, une attention particulière devrait être accordée à (1) l’augmentation des
connaissances de base sur l’amplitude naturelle de la variabilité des écosystèmes forestiers non aménagés et l’évaluation de
la validité de cette information dans un contexte de changements climatiques, (2) la promotion de la recherche multidiscipli-
naire avec une meilleure intégration de la théorie écologique dans la recherche intégrative et analytique sur l’IPN et (3) une
meilleure intégration des préoccupations socioéconomiques, des schémas d’aménagement adaptatif et de la collaboration in-
ternationale dans le cadre des initiatives qui portent sur l’IPN.
[Traduit par la Rédaction]
Introduction
The idea of using natural forest structures and develop-
mental patterns as a basis for sustainable forest management
was first developed in Europe during the 19th century (Jo-
hann 2006; Jonsson et al. 2011). Although the definition and
understanding of what is “natural”have changed over time,
the concept of following nature’s ways has persisted (e.g.,
Received 11 November 2011. Accepted 10 April 2012. Published at www.nrcresearchpress.com/cjfr on 3 July 2012.
T. Kuuluvainen and R. Grenfell. Department of Forest Sciences, P.O. Box 27, 00014 University of Helsinki, Helsinki, Finland.
Corresponding author: Timo Kuuluvainen (e-mail: Timo.Kuuluvainen@helsinki.fi).
1This article is one of a selection of papers from the 7th International Conference on Disturbance Dynamics in Boreal Forests.
1185
Can. J. For. Res. 42: 1185–1203 (2012) doi:10.1139/X2012-064 Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

Leibundgut 1978, 1981). Nevertheless, more narrow utilitar-
ian philosophies have undoubtedly shaped mainstream silvi-
culture in the 20th century.
Although many pioneering early 20th century studies on
European boreal forests were carried out in near-pristine for-
ests, these studies were mostly motivated by the desire to re-
veal patterns of forest development that could be translated
into intensive management practises (e.g., Sirén 1955; Jons-
son et al. 2011). In the aftermath of World War II, silvicul-
ture became increasingly utilitarian, a development hastened
by the growing demand for timber for the booming forest in-
dustry coupled with the combination of improved machinery
and availability of cheap fossil fuels. Rapid industrialization
and inexpensive energy created the illusion that forests could
be indefinitely modified to boost timber production. Conse-
quently, silviculture adopted methods more akin to agricul-
tural crop production, such as preference for monocultures,
genetic breeding, intensive crop manipulation, and the use of
fertilizers and pesticides.
Since the 1990s, increasing ecological knowledge, societal
awareness, and sensitivity to the adverse ecological impacts
of intensified forestry has aroused fresh interest in more eco-
logically sensitive, nature-based forest management ideas
(Spence 2001). During the last 10–15 years, this core princi-
ple has guided the development of concepts such as “nature-
based forestry”and “continuous-cover forestry”in Europe
(Pommerening and Murphy 2004; Diaci 2006), “new for-
estry”in North America and elsewhere (Kohm and Franklin
1997) as well as the proposal of “managing forests as com-
plex adaptive systems”(Puettmann et al. 2009). These man-
agement ideas have been sparked by, and benefitted from, an
increasingly robust knowledge base of forests as ecological
systems (e.g., Esseen et al. 1997; Kuuluvainen 2002a,
2002b; Lindenmayer and Franklin 2002; Seymour et al.
2002; Gauthier et al. 2009). Natural disturbance emulation is
one such management idea.
Natural disturbance emulation (NDE) aims to implement
such forest management practices that would reproduce and
maintain the essential forest structural features that are
present in the forest under a natural disturbance regime. In
essence, the goal is to design harvesting patterns to emulate
the main ecological outcomes of natural disturbances (Gauth-
ier et al. 2009). It is essential to consider the defining charac-
teristics of the disturbance regime, such as the variability of
disturbance size, severity, and quality as well as temporal re-
peatability at the landscape scale (Perera et al. 2004). The
underlying general hypothesis is that by emulating the forest
structural dynamics created by natural disturbances, the ma-
jority of the process and species diversity of the ecosystem
will be better safeguarded than under conventional even-aged
management. This of course depends on how well, and to
what extent, the ecological functions of natural disturbances
can be emulated with forest management.
The NDE approach to forest management can also be put
under the heading of “biomimicry”through which natural
ecosystem function is used to inspire the design of sustain-
able management systems. This is not to be confused with
the moral stance of biocentricism (“nature knows best”).
Rather, the aim is to learn from natural systems and apply
this knowledge to attain specific goals such as long-term sus-
tainability. After all, natural forests are prime examples of re-
silient systems, which have been able to persist and adapt
through millennia to evolve an amazing degree of diversity
and complexity, despite catastrophic environmental changes.
In this paper, we first review ecological theories related to
NDE as a forest management approach. Secondly, we discuss
the main strategic models that seek to use the natural boreal
forest, and particularly its disturbance regimes, as a guide to
ecologically sustainable forest management. Thirdly, we pro-
vide an overview on papers that generally discussed NDE
without comparison with even-aged management. Fourth, we
offer a systematic review of studies that compared the NDE
approach and conventional even-aged clearcutting based
management in the boreal forest. This review was not in-
tended to be exhaustive; instead, our aim was to explore the
most relevant body of research. Finally, we discuss the future
challenges of managing boreal forest ecosystems based on
the NDE approach and examine the prerequisites for an inter-
disciplinary research agenda.
Theoretical concepts related to natural
disturbance emulation
The need for a new approach to forest management has
been driven by recent advances in ecological theory, which
have, in turn, led to better understanding of the dynamics of
unmanaged forest ecosystems and increased awareness of the
significant differences between natural and managed forests
(Sousa 1984; Bergeron et al. 2002; Kuuluvainen 2002b,
2009; Cyr et al. 2009; Puettmann et al. 2009; Kneeshaw et
al. 2011). In particular, since the 1970s, there has been a
growing recognition of the importance and ubiquity of natu-
ral disturbance in forest ecosystems. This recognition oc-
curred first in relation to tropical and temperate forests and
later for boreal forests (Pickett and White 1986; Attiwill
1994; Kuuluvainen 1994; Kneeshaw et al. 2011). Shortly
thereafter, the concept of using NDE as a management ap-
proach started to appear rather rapidly in the research litera-
ture (Attiwill 1994; Angelstam 1998; Bergeron et al. 2002;
Seymour et al. 2002; Perera et al. 2004; North and Keeton
2008). This proliferation of material on NDE was precipitated
by a growing realization that natural disturbance is integral to
the proper functioning of ecological communities and not
merely an external, antagonistic influence (Pickett and White
1986; Clark 1989).
Contemporary novel approaches to forest management uni-
versally recognize the fact that the maintenance of ecosystem
integrity, resilience, and biodiversity must be underpinned
and guided by unambiguous ecological principles (Gunder-
son and Holling 2002; Drever et al. 2006; Gauthier et al.
2009). What NDE suggests is that disturbance ecology is the
key to understanding, re-creating, and perpetuating these im-
portant ecosystem properties (Fig. 1) (Attiwill 1994; Pickett
and White 1986; Gauthier et al. 2009). It is therefore vital to
understand the mechanism whereby disturbance facilitates the
existence of diverse species assemblages.
Perhaps the most general systems framework for forest dis-
turbance ecology and NDE is provided by the “Panarchy”
concept, which is a conceptual framework explaining how
complex systems maintain resilience through change and
adaptation over time (Gunderson and Holling 2002; Drever
et al. 2006). Other concepts closely related to the NDE in-
1186 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

clude Hunter’s et al. (1988) metaphor of “coarse filter”and
“fine filter”and the “natural range of variability”(NRV) con-
cept (Fig. 1) (Landres et al. 1999). The “coarse filter man-
agement”strategy is based on the hypothesis that by
maintaining natural forest structures and dynamics, viz. habi-
tats at multiple scales, the corresponding diversity of species
and variability in processes (e.g., hydrological, nutrient, and
carbon cycles) of the ecosystem can also be safeguarded
(Hunter et al. 1988). Based on this idea, Hunter (1993) pre-
sented a theoretical discourse on how the size and distribu-
tion of natural stand-replacing wildfire disturbances
(Heinselman 1973; Cogbill 1985) could be used to devise a
spatial harvesting model in Quebec and Labrador, Canada.
Modifying timber harvests to mimic the whole range of
likely fire sizes was concluded to be integral to success
(Hunter 1993).
The NRV (sometimes “historical range of variability”)isa
concept that refers to the spatiotemporal oscillations of eco-
system structures and processes that would be expected to
occur in the absence of human impact (Landres et al. 1999;
Cyr et al. 2009; Keane et al. 2009). The NRV-based manage-
ment can be regarded as a dynamic realization of the coarse
filter principle. Nevertheless, the concept is somewhat ambig-
uous, as it is strongly dependent on the spatial and temporal
scales of observation. Furthermore, many disparate defini-
tions and methods are used to define the NRV. Moreover,
the question has been asked whether knowledge of past eco-
systems will still be applicable in the future with current rap-
idly changing environmental conditions (Keane et al. 2009).
Despite these shortcomings, the concept of NRV can be re-
garded as fundamentally important to NDE and is likely to
retain that status in the future (Keane et al. 2009).
Both the coarse filter and NRV, and NDE-based manage-
ment as their applied derivative, acknowledge the fact that
for the time being, we lack crucial knowledge of functioning
of forest ecosystems; the responses of forest ecosystems may
therefore be unpredictable. Hence, NDE can be regarded to
be congruent with the “precautionary principle”where risks
of adverse effects are avoided by refraining from modifying
the ecosystem structures beyond their natural limits of varia-
bility (Gauthier et al. 2009).
The “intermediate disturbance hypothesis”(IDH) (Connell
1978) goes a step further by proposing a mechanism for the
coexistence of species in disturbed environments. As such, it
also represents one of the most promising theoretical frame-
works for NDE. This hypothesis states that species richness
is maximal under intermediate levels of disturbance. Here,
“intermediacy”usually refers to frequency, but it can also re-
fer to intensity, extent, or duration of disturbance (Shea et al.
2004). IDH is based on the assumption that under intermedi-
ate disturbances, both early- and late-successional species can
survive, which will result in high diversity. The IDH is based
on two fundamental principles: (1) if disturbances are rare
events, then competitive exclusion results in dominance of
late-successional species and low species diversity and (2) if
disturbances are very frequent, then most late-successional
species are eliminated, again resulting in low diversity. It
therefore logically follows that a disturbance frequency that
lies between these two extremes would allow the continued
existence of patches that host late-successional species as
well as the creation of the disturbed patches needed by
early-successional species. The hypothesis intrinsically as-
sumes that there is a trade-off between the colonizing ability
of species and their competitive ability. This means that there
are distinct early-successional pioneer species and late-suc-
cessional species (Wilson 1994). It is essential to notice that
here, IDH is interpreted as a between-patch spatial metapopu-
lation process. Thus, what IDH essentially hypothesizes is
that when NDE succeeds to maintain the patch dynamics
within its NRV, it will also safeguard native species popula-
tions and their dynamics.
Strategic approaches and proposed models
The theoretical concepts and models discussed above do
not offer a means for direct application of NDE within forest
management practice. Instead, strategic models and principles
that translate abstract theories and hypotheses into systems
that can be applied in real-world scenarios are required. Re-
cent books by Burton et al. (2003), Perera et al. (2004), and
Gauthier et al. (2009) provide some step by step guidelines
of how to put NDE into action. However, so far, only a few
specific strategic models have been proposed that detail how
to actually transform contemporary models of natural forest
disturbance dynamics into practical forest management solu-
tions. These models are strategic in the sense that they do
not merely reiterate the desire to emulate natural forest dy-
namics but instead go one step further by making concrete,
practical proposals. Examples of such strategic models are
provided by (1) the “patch-corridor-matrix”model, (2) the
“ASIO model”, and (3) the “multicohort model”.
(1) The “patch-corridor-matrix model”is based on the
patch-mosaic paradigm of landscape ecology, which states
Fig. 1. Illustration of the natural disturbance emulation (NDE)
model and related ecological concepts in forest management. The
core of the approach is stand- and landscape-level emulation of nat-
ural disturbance –succession cycles in forest management, which
can comprise both silvicultural and more restorative treatments.
Knowledge of the natural range of variability (NRV) at multiple
scales is fundamental to NDE because it is used to ensure that the
essential spatiotemporal habitat variability at the landscape level (the
coarse filter) is maintained within the bounds of the NRV. Combin-
ing NDE-based management with specific habitat protection/restora-
tion (the fine filter) measures aims to ensure the maintenance of
native diversity of ecosystem types and populations of species.
Kuuluvainen and Grenfell 1187
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

that landscapes are composed of mosaics of distinct patches
(Forman 1995; Wiens 1997). The paradigm conceptualizes
the forest area as being a mosaic composed of three ele-
ments: firstly, patches, which are somewhat homogenous,
nonlinear areas that differ significantly from their surround-
ings (in forest ecosystems, these are usually valuable habitats
such as key biotopes or protected areas), secondly, corridors
or stepping-stones, which are of the same type of habitat as
patches, and, lastly, the surrounding matrix, which is often
assumed to be unsuitable or low-quality habitat (Forman
1995). The background of this management model lies in the
theory of island biogeography (MacArthur and Wilson 1967)
and in the more recently developed metapopulation theory
(Hanski 1999). Another somewhat complimentary strategic
model is the “landscape continuum model”(McIntyre and
Hobbs 1999). Whereas the patch-corridor-matrix model is
generally concerned with the spatial arrangement of these el-
ements, the landscape continuum model (McIntyre and
Hobbs 1999) recognizes that distinct patches may not always
be easy to determine. The landscape continuum model sug-
gests that landscape-scale variability in vegetation cover that
results in functional heterogeneity for a particular species or
assemblage is more significant than the spatial configuration
per se. In a practical context, both the spatial configuration
and the functional variability of a given landscape are signifi-
cant. The key principles of both models have been applied
with some success in the boreal forest where the protected
core habitats comprise moist fertile sites with low disturbance
frequency and high species richness (Class A in the ASIO-
model; see below and Fig. 2), while the managed sites have
a higher disturbance frequency and low species richness, do-
minated by species better adapted to disturbance. It can be
debated whether the patch-corridor-matrix model and contin-
uum model actually represent true disturbance emulation be-
cause of the extremely simplified representation of the
landscape elements and spatially limited view. The sole focus
of the basic model on the spatial arrangement of landscape
elements poorly conforms to contemporary understanding of
the multiscaled dynamic structural complexity of forest eco-
systems.
(2) The “ASIO model”was developed in Sweden in the
1990s as a strategic conceptual model for emulating natural
disturbance in forest management (Fig. 2) (Rülcker et al.
1994; Angelstam 1998). The principles are informed by re-
search on the role and impact of natural disturbances, espe-
cially fire, in the northern boreal forest (Zackrisson 1977).
This research suggested that in Fennoscandia, “natural”fire
frequency was closely connected to a gradient of site fertility
and soil moisture, ranging from low fire frequency in moist
herb-rich forests to high fire frequency in dry sandy heaths
dominated by Scots pine (Pinus sylvestris L.) (Fig. 2). The
landscape can thus be divided into four categories along the
site moisture and fertility gradient where fire is correspond-
ingly absent or occurs “seldom”,“infrequently”,or“often”.
In each of these categories, different silvicultural and harvest-
ing techniques are used to imitate the impact of fire and other
disturbances on forest structure and dynamics. A benefit of
the ASIO model is that it maps directly onto a site classifica-
tion system. The ASIO model can be linked with various
management and cutting regimes for each category that are
congruous with the assumed frequency and effect of fire (or
absence of fire) on stand structure and succession. It has been
suggested that in practice, herb-rich sites with lowest fire fre-
quency, which have high biodiversity and conservation value,
should be set aside as reserves (Fig. 2) (Angelstam 1998).
The ASIO model is similar to both the patch-corridor-ma-
trix and landscape continuum models in that it leaves the
core areas outside management and recognizes the signifi-
cance of both form and function in the disturbance-driven
landscape. A potentially diverse set of silvicultural tools can
be used to emulate the fire-impacted structures that are typi-
cal of natural forests (Fig. 2). The assumption that fire fre-
quency is dictated solely by local site properties is a
somewhat controversial and contested aspect of the ASIO
model because it has been demonstrated that in reality, land-
scape structure also plays a role (Hellberg et al. 2004). In
practice, the ASIO model has therefore generally been used
more as a theoretical reference where conventional clearcut-
ting is simply used across most of the forest landscape. How-
ever, in the light of our knowledge of the diverse impact fire
on forest structure in Fennoscandian conditions, this simplifi-
cation is clearly not consistent with the goal of “emulating
fire”(Kuuluvainen 2009). How well the ASIO model emu-
lates natural forest thus depends on both the level of knowl-
edge of natural forest dynamics and the degree of willingness
to fully implement the model (Kuuluvainen 2009).
(3) The “multicohort model”was presented by Bergeron et
al. (1999) and Bergeron et al. (2002) based on research on
forest fire and successional dynamics carried out in Quebec,
eastern Canada (Fig. 3). It is a strategic model that explicitly
considers the effects of NDE at multiple spatial scales from
the stand to the landscape level. The dominant disturbance
agent is stand-replacing fire with variable periodicity (Ber-
geron and Harvey 1997). The fundamental concept is that
the successional development of a post-fire forest moves
through distinct phases, which can be termed “structural co-
horts”, under the potential influence of further disturbance
events (Fig. 3). The natural proportions, or range of variabil-
ity, of the different “structural cohorts”representing the vari-
ous post-fire successional stages are derived from a
spatiotemporal analysis of the historical landscape disturb-
ance regime. Where stand-replacing disturbances are infre-
quent, the landscape is dominated by old forest, whereas
young forest dominates when severe disturbances are fre-
quent. An important distinction from the ASIO model is that
fires are assumed to occur randomly across the landscape in
the original formulation of the multicohort model (Bergeron
et al. 2002).
Based on the analysis of natural landscape disturbance dy-
namics, different silvicultural methods are applied at the
stand scale to maintain the natural proportions and spatial
pattern of “structural cohorts”at the landscape level (Fig. 3)
(Bergeron et al. 2002). Stands are subjected to different treat-
ments depending on which “structural cohort”and processes
should be emulated. Younger stands can be treated with thin-
ning and partial cutting. Older stands, representing late-suc-
cessional stages, could be subjected to partial or gap cutting.
Clearfelling with appropriate retention maintains a suitable
proportion of open habitat and young forest age classes in
the landscape.
It was suggested that this approach could maintain the for-
est landscape within the natural range of structural and com-
1188 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

positional variability (NRV). As with the ASIO model, the
multicohort model also highlights the importance of areas
set aside solely for conservation surrounded by a buffer zone
with high tree retention. These principles were evaluated as
providing a robust and efficacious basis for biodiversity con-
servation and have since provided a working framework for
further studies (Bergeron et al. 1999, 2004, 2007; Harvey et
al. 2002).
Review of studies on NDE: a systematic
approach
We aimed to review scientific publications that either gen-
erally discussed and evaluated the NDE approach or specifi-
cally compared it with conventional clearcutting management
in boreal forestry. In the latter case, we specifically asked
whether NDE was shown to be more effective than tradi-
tional clearcut forestry with respect to ecological and socioe-
conomic sustainability.
The principles of systematic review were applied to the
collation of relevant studies (e.g., see Pullin and Stewart
2006; Pullin et al. 2009). We gathered the relevant publica-
tions by searching the BIOSIS, CAB, CSA, and ISI databases
using the following search string: (“coarse filter”or (“natural
disturbance”and (emulat* or imitat* or model))) and (forest)
and (boreal). The studies were published prior to May 2011.
The database search retrieved 166 publications excluding
double hits. Some additional studies were found from the
bibliographies of retrieved studies and by consulting fellow
researchers. The retrieved studies were then assessed for rele-
vance to the study. In addition, we only accepted those stud-
ies that had been peer reviewed and published in a scientific
journal or book. On the basis of these criteria, a total of 74
studies were identified.
Fig. 2. Illustration of the principle of the ASIO model (Angelstam 1998). The landscape is divided into four categories along a site moisture
and fertility gradient, with assumedly different natural fire frequency intervals, i.e., where fire is absent or occurs “seldom”,“infrequently”,or
“often”. The ASIO model may can be implemented with various management and cutting regimes for each category that are congruous with
the assumed frequency and effect of fire (or absence of fire) on stand structure and succession. (Drawing: J. Karsisto.)
Fig. 3. Illustration of the principle of the multicohort model (Bergeron et al. 2002). The model aims to emulate natural disturbances and
succession to maintain the structural features typical of the natural forest. In this approach, the forest area is divided into “structural cohorts”
representing different post-fire stand successional stages. Different cutting methods and intensities with variable final harvest rotations are
applied to maintain landscape-level forest structures and age distributions similar to those that would exist under a natural disturbance regime.
(Drawing: J. Karsisto.)
Kuuluvainen and Grenfell 1189
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

A total of 29 out of the 74 studies either generally dis-
cussed and evaluated the NDE approach or specifically com-
pared it with conventional clearcutting management in boreal
forestry. Of those, 16 were review or discussion studies that
provide general evaluations of NDE (“soft”results, Table 1).
The remaining 13 studies used empirical methods or simula-
tions to compare NDE with conventional even-aged forestry
(“hard”results, Table 2). The studies were further sorted and
tabulated by study type (empirical, simulation, theoretical,
strategic, or review), geographic area, dominant disturbance
agent, focal subject (timber supply, ecology, or social issues),
focal species, temporal and spatial scale, and dominant tree
species (Tables 1 and 2).
All but one of the reviewed studies were from North
America, and most considered the impact of large-scale dis-
turbances, mainly stand-replacing fires (Tables 1 and 2). Out
of the 29 studies, 26 primarily focused on wildfire. The land-
scape scale was considered by 16 studies, the stand scale by
20, and the regional scale by 3, with several of these studies
addressing more than one scale. Some mention was made of
gaps, but no study focused primarily on this scale. The tem-
poral scales of the simulations ranged from 20 to 1000 years.
Review and discussion papers on NDE
In the following section, we give a brief overview of the
review and discussion papers that evaluated NDE as a man-
agement strategy without producing tangible results or for-
mulating specific strategic management models (Table 1).
The papers were grouped into three categories: (1) general
evaluations, (2) papers that focus on ecological outcomes,
and (3) papers that refer to social aspects of NDE. There
have been several papers of this type and they can be valua-
ble for formulating and discussing the NDE approach in a
broader multidisciplinary research and societal application
context.
(1) General evaluations of NDE were provided by Euler et
al. (2004) and Burton et al. (2006). Euler et al. (2004) exam-
ined NDE in Canada and explained that while limited evi-
dence exists to show that NDE is beneficial for biodiversity,
there is evidence that traditional even-aged management has
a negative impact on biodiversity. The natural disturbance
based management principles of Attiwill (1994) were ex-
plained to provide a justification for NDE, and the spatially
explicit disturbance index developed by Seymour et al.
(2002) was advanced as a suitable quantitative tool for the
approach. Aside from this “precautionary approach”, another
factor that was suggested to be intrinsic to NDE was the
“conservation ethic”perspective (sensu Leopold 1949; also
see Klenk et al. 2009). This concept extends the remit of
management to include ecosystem values and services that
cannot be evaluated in purely scientific or economic terms
(Euler et al. 2004).
This concept was also discussed by Burton et al. (2006) in
their general examination of a decade of progressive manage-
ment efforts across Canada. The authors detailed the potential
range of alternative deliverable ecosystem services, which in-
cludes biodiversity conservation, building community ca-
pacity, hunting, water supply, late-successional habitat,
carbon sequestration, and recreation. Again invoking Attiwill
(1994), NDE was suggested to be the logical benchmark for
sustainable management. However, it was recognized that
there is a lack of clear management guidance, possibly due
to NDE being regarded as an untested hypothesis coupled
with an incomplete understanding of natural disturbance
ecology. Moreover, the difficulty of prescribing market val-
ues for nontimber services was recognized as a further prob-
lem, resulting in management plans incorporating nontimber
values merely as constraints rather than products per se (cf.
Armstrong et al. 1999, 2003; McCarney et al. 2008). As a
consequence, the most promising solutions are those that pro-
vide win-win solutions with immediately demonstrable eco-
nomic and environmental benefits. Burton et al. (2006)
concluded that while NDE may maintain a wider range of
ecosystem services, more evidence is needed to build indica-
tors. Furthermore, all services will ultimately be subject to
economic, cultural, and institutional restrictions (Burton et
al. 2006).
(2) Ecological aspects of NDE were reviewed or discussed
by Harvey et al. (2002), Drever et al. (2006), Bergeron et al.
(2007), Thorpe and Thomas (2007), Samways (2007), Bou-
chard et al. (2008), Long (2009), and Zwolak (2009).
Harvey et al. (2002) presented a case study from a multi-
cohort NDE trial in Lake Duparquet Forest, southern Que-
bec. The native disturbance regime was taken to be
dominated by a 140-year fire cycle of variable severity (Ber-
geron 1991; Dansereau and Bergeron 1993; Bergeron et al.
1999), with some secondary influence from spruce budworm
(Morin et al. 1993; Bergeron and Leduc 1998). It was con-
cluded that the approach may offer a number of advantages
over conventional management through promoting nontimber
values such as habitat creation and recreation. Furthermore,
the possibility that partial cuts may alter stand compositional
trajectories in favour of mixed stand types was noted as hav-
ing the potential to benefit seed bank renewal, natural regen-
eration establishment, site productivity, and soil nutrient
levels. However, it was noted that due to high natural varia-
bility of stand structural and compositional dynamics, the
creation of typical stand age structures does not guarantee
that the NRV of stand characteristics will be present. Further-
more, as the approach was designed for a region naturally do-
minated by large catastrophic fires, it is unclear how it would
perform in regions with contrasting disturbance regimes
(Harvey et al. 2002).
Drever et al. (2006) ask whether NDE-based forest man-
agement is effective for maintaining ecological resilience.
The Panarchy concept (Holling 2001; Gunderson and Holling
2002) was employed as a theoretical basis, explaining both
the concept of ecological resilience and how the variability
resulting from natural disturbance maintains resilience. Resil-
ience was described as being an important tool for overcom-
ing both theoretical and practical barriers to NDE-based
management by providing a conceptual argument for the sus-
tainability of NDE and through prompting ecological process
inspired management over many temporal and spatial scales.
The authors concluded that “resilience thinking”(Walker and
Salt 2006) could be a very useful approach for developing
and promoting NDE (Drever et al. 2006).
Bergeron et al. (2007) examined how NDE could be used
to replicate the near-natural wildfire-driven forest landscapes
of Quebec’s Clay Belt across managed landscapes. The dom-
inant disturbance regime was taken to be fire of variable se-
1190 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

Table 1. The examined review and discussion papers dealing with natural disturbance emulation that did not deliver original results.
Study Study type Temporal scaleaSpatial scale Location Focal subject Disturbance agent Tree speciesb
Long 2009 Review Stand, landscape Canada Ecology Wildfire, wind, insects
Burton et al. 2006 Review Canada Ecology Wildfire
Thorpe and Thomas 2007 Review Stand Canada Ecology Wildfire
Hebert et al. 2003 Review Intermediate Stand, landscape Canada Sociology Wildfire
Bergeron et al. 2002 Strategic Stand, landscape Quebec, Canada Ecology Wildfire Pop tre, Bet pap, Pin ban,
Abi bal, Pic gla
Harvey et al. 2002 Strategic Stand, landscape Quebec, Canada Ecology Wildfire Abi bal, Pic gla, Bet pap,
Pin ban, Pic mar
Klenk et al. 2009 Theoretical Canada Sociology
Zwolak 2009 Theoretical Intermediate,long Stand USA, Canada Ecology Wildfire
Bouchard et al. 2008 Theoretical Long Quebec, Canada Ecology Wildfire, insects Pop gla, Bet pap, Pin ban,
Abi bal, Pic gla
Belleau et al. 2007 Theoretical Long Stand, landscape Canada Ecology Wildfire
Drever et al. 2006 Theoretical Stand, landscape,
ecosystem
Various Ecology
Samways 2007 Theoretical Ecology
Euler et al. 2004 Theoretical Stand, landscape Canada Ecology Wildfire Pin ban, Pic mar
Note: The papers are classified as review, strategic, or theoretical papers based on their structure and approach. The papers are characterized in terms of temporal and spatial scales involved, geographic
location, focal subject, main disturbance agent, and main tree species.
aShort temporal scale 0–5 years, intermediate temporal scale 6–50 years, and long temporal scale >50 years.
bPic gla, Picea glauca; Pic mar, Picea mariana; Pin ban, Pinus banksiana; Pop tre, Populus tremuloides; Bet pap, Betula papyrifera; Pic abi, Picea abies; Pin syl, Pinus sylvestris; Abi bal, Abies balsami-
fera.
Kuuluvainen and Grenfell 1191
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

Table 2. The reviewed studies that made a direct comparison between natural disturbance emulation based management and conventional management using clearcutting.
Study Study type
Temporal
scaleaSpatial scale Location
Focal
subject
Disturbance
agent Focal species Tree speciesb
Van Wilgenburg and
Hobson 2008
Empiricial Short Landscape Alberta, Canada Ecology Wildfire Birds Pop tre, Pin ban, Abi
bal, Pic gla
Hannam et al. 2006 Empiricial Short Stand Alberta, Canada Ecology Wildfire Microbes Pic gla, Pop tre, Pic
mar, Pin con
Stuart-Smith et al. 2006 Empiricial Short Stand British Columbia,
Canada
Ecology Wildfire Birds Psu men, Pic gla, Pin
con, Pop tre
Harrison et al. 2005 Empiricial Short Stand Alberta, Canada Ecology Wildfire Birds Pic gla, Pin con, Pic
mar, Pop tre
Hogberg et al. 2002 Empiricial Short Stand Alberta, Canada Ecology Wildfire Bats Pic gla, Pic mar, Pop
tre
Shinneman et al. 2010 Simulation Long Landscape Minnesota, USA,
Ontario, Canada
Ecology Wildfire Pin ban, Pic gla, Pop
tre, Bet pap
McCarney et al. 2008 Simulation Long Stand, region Alberta, Canada Timber
supply
Wildfire Pic gla, Pop tre, Pic
mar, Pin con
Nielsen et al. 2008 Simulation Long Landscape Alberta, Canada Ecology Wildfire Bears Pic gla, Pin con, Pop
tre
Bergeron et al. 2007 Simulation Stand, landscape Quebec, Canada Ecology Wildfire Pop tre, Bet pap, Abi
bal, Picea gla
Carlson and Kurz 2007 Simulation Long Landscape Alberta, Canada Ecology Wildfire Pop tre, Pic gla, Pic
ari, Pin ban
Rempel et al. 2007 Simulation Long Stand, landscape Ontario, Canada Ecology Wildfire Birds
Wintle et al. 2005 Simulation Long Landscape Ontario, Canada Ecology Wildfire Birds Pin ban, Pic spp., Abi
bal, Bet pap
Bergeron et al. 2004 Simulation Intermediate Stand, landscape Quebec, Canada Ecology Wildfire
Armstrong et al. 2003 Simulation Long Stand, region Alberta, Canada Timber
supply
Wildfire Pic gla, Pop tre, Pic
mar, Pin con
Armstrong et al. 1999 Simulation Long Stand, region Alberta, Canada Timber
supply
Wildfire Pic gla
Delong and Tanner 1996 Simulation Long Stand, landscape British Columbia,
Canada
Ecology Wildfire Pic gla, Pin con
Note: The studies are grouped by study approach (empirical or simulation) and characterized in terms of the temporal and spatial scales involved, geographic location, focal subject, main disturbance agent,
focal species group studied, and main tree species.
aShort time scale 0–5 years, intermediate time scale 6–50 years, and long time scale >50 years.
bPic gla, Picea glauca; Pin con, Pinus contorta; Pic mar, Picea mariana; Pin ban, Pinus banksiana; Abi bal, Abies balsamifera; Psu men, Psuedotsuga menziesii; Pop tre, Populus tremuloides; Bet pap,
Betula papyrifera.
1192 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

verity (Johnson and Van Wagner 1985), which results in a
considerable percentage of the forest landscape being older
than the native fire cycle. It was proposed that the multico-
hort management model (Bergeron et al. 1999, 2002) could
provide a means to bring managed forest landscapes closer
to naturally disturbed landscapes. Comprehensive benefits
were predicted for wood production by reducing paludifica-
tion and enhancing biodiversity through maintaining more
heterogeneous forest landscapes (Bergeron et al. 2007).
Thorpe and Thomas (2007) reviewed the impact of partial
harvesting on residual tree growth and mortality from trials
in various areas in Canada. They explained that without in-
formation on future yields, it is difficult to tell not only the
economic feasibility of partial cutting but also the biodiver-
sity impact. They found that an increased post-harvest mor-
tality risk, particularly with sparse retention, translates into
economic loss and that the common positive growth response
of residuals may not offset this loss. As commonly used sim-
ulation models make assumptions that are not consistent with
these findings, the authors therefore concluded that models
must be improved to allow the accurate prediction of dynam-
ics needed to assess the economic and biological impact of
partial harvesting.
Samways (2007) discussed the general principles of NDE
for insect conservation. While not exclusively considering
forest ecosystems, the study generally considered insect con-
servation theory and strategy, concluding that NDE-based
management was a fitting practical approach. With an ex-
plicit foundation in the coarse–fine filter approach and meta-
population theory (Hanski 1999), it was concluded that NDE
has a role to play in insect conservation through maintaining
high levels of habitat heterogeneity, not only in reserve areas
but also the in the surrounding matrix, thus reducing the con-
trast between the two. Indeed, the metapopulation trio of
maintaining suitable patch size, quality, and connectivity,
which allow sustainable metapopulation dynamics to occur,
was seen to be compatible with an NDE-based management
strategy. One discussed problem with the approach was how
to define the temporal envelope of emulation. Further prob-
lems were anticipated in areas with high fragmentation where
emulation of large-scale disturbances is not possible due to
the small size of remnant habitat fragments. In such instan-
ces, it was concluded, emulating natural conditions may still
serve a useful purpose in meeting specific local conservation
goals. Overall, NDE was suggested to have many potential
benefits for insects through reducing fragmentation and creat-
ing suitable habitat (Samways 2007).
Bouchard et al. (2008) used a 150-year record of regional
data to assess the theoretical impacts of emulating large-scale
disturbance pulses due to wildfire (Girardin et al. 2004; Gre-
nier et al. 2005), spruce budworm (Bouchard et al. 2005,
2006), and birch mortality. It was found that in the hemibor-
eal Témiscamingue region of western Quebec, pulses of
large-scale disturbance lasting from one to a few decades
dominate the disturbance regime. The proposed alternative
management scenario was to mimic the spatiotemporal char-
acteristics of this regime with temporal clustering of intensive
harvest operations over 10–20 years, with 50- to 100-year
respite periods. The rationale offered is that large-scale dis-
turbances are an important element of ecosystem function
but have been poorly investigated and rarely considered for
emulation. One critical shortcoming was that in instances
where it is difficult to predict future natural disturbance re-
gimes, such management risks moving the ecosystem further
outside the NRV compared with traditional management. It
was concluded that emulation of pulsed severe disturbance
should therefore be restricted to areas where natural disturb-
ance has been well below the upper boundary of NRV. An-
other potential problem was that pulsed disturbance will lead
to an inconsistent timber supply, although it was noted that
most timber supply is, in fact, rather inconsistent and fluctu-
ating timber supply could be more profitable, as it would al-
low supply to follow market variation. It was also noted that
heavily degraded areas could be used as traditionally man-
aged areas to maintain timber supplies between harvest
pulses. Despite the theoretical positive impacts on biodiver-
sity, the approach was not fully endorsed by the authors be-
cause of the uncertainties in determining the NRV and
impacts on timber supply. It was concluded that more study
on mimicking large-scale disturbance, and effect on timber
supply, is urgently required (Bouchard et al. 2008).
Zwolak (2009) used meta-analysis to compare the effects
of stand-replacing fire and several modes of forest harvest
on deer mice (Peromyscus maniculatus (Wagner, 1845)) and
red-backed voles (Myodes gapperi (Vigors, 1830)) in western
and northern boreal forests and eastern temperate forests in
North America. Theoretically, deer mice numbers should in-
crease after stand-replacing fire and red-backed voles should
decline. Results suggested that while the expected changes
occurred with stand replacement, either by harvesting or
wildfire, the effects were less pronounced with clearcutting.
This suggests that while the structural changes resulting
from the two disturbance agents may be comparable, incon-
sistencies exist that may be significant for particular species.
One suggested difference was that severe fires also consume
the ground vegetation; however, burning after clearcutting
failed to elicit a significant difference in the response of ei-
ther species. Due to this disparity, facilitating the occurrence
of natural disturbance was seen as a critical component of
NDE initiatives. It was therefore concluded that emulating
wildfire with clearcutting alone may not benefit all target
species (Zwolak 2009).
Finally, Long (2009) reviewed the principles of NDE from
a general North American perspective, specifically addressing
the “why, what, and how”of NDE. The justification for the
approach was that it builds resilience and biological diversity.
The author emphasized the importance of building resilience
at a time when there is a growing scientific consensus that
atypical disturbance events are becoming more common.
The study draws attention to the fact that NDE often revolves
round a single main disturbance agent —predominantly
stand-replacing fire —and that other agents and interactions
of multiple agents are rarely simulated. Evidence of direct
benefit from NDE was given by the examples of the substan-
tial increase in numbers of Kirtland’s Warblers (Dedroica kir-
tlandii (S.F. Baird, 1852)) (Kashian et al. 2003) and
landscape-scale reduction in crown fire occurrence (Rocca-
forte et al. 2008). Indirect economic benefit, for example
from crown fire suppression, was thought to be an underval-
ued factor. Climate change, habitat fragmentation, and degra-
dation were thought to be obstacles to NDE implementation;
Kuuluvainen and Grenfell 1193
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

however, it was concluded that socioeconomic issues may
represent the greatest challenge (Long 2009).
(3) Social aspects were specifically treated in only two pa-
pers. Hebert et al. (2003) reviewed a series of case studies
from across Canada with the main aim of illustrating the fea-
sibility of integrating social dimensions into sustainable for-
est management initiatives. Past management options were
claimed to have suffered from being too prescriptive, ecolog-
ically unambitious, and often disproportionate to the desired
outcome. Alternative adaptive management options were pre-
sented and illustrated through real case studies. These novel
management options included tenant farming schemes and
community-based comanagement boards. Conclusions were
reached about the participatory process, emphasizing the im-
portance of ensuring that involvement represents the true di-
versity of the stakeholder community and making provision
in policy and law for any agreements to be binding. It was
also concluded that the tangible benefits must be delivered
to the local community and also that understanding of the lo-
cal land and people must be integrated into the management
initiatives. The need for education of associated groups —in-
cluding local stakeholders, industry representatives, private
forest managers, and governments —was highlighted and
“model forests”were promoted as an effective education tool
to meet these ends. The importance of institutional support
from provincial forest agencies, certification schemes, and
national and international standards bodies was also empha-
sized. Predictably, the main limitations to the success of
adaptive management were perceived as being time and
money. It was thus concluded that committed investment
from industry and government is pivotal to alternative man-
agement schemes such as NDE. In summary, it was shown
that effective integration of social benefits and ecological
benefits is only possible in a conducive administrative and fi-
nancial environment (Hebert et al. 2003).
In a generally critical evaluation, Klenk et al. (2009) sug-
gested that NDE, as with many scientific disciplines, is influ-
enced by moral and ethical values that are obscured by the
strictly scientific characterizations presented in the literature.
It was further argued that the generally accepted definitions
of desirable “natural variability”are arbitrary historical refer-
ence points and, furthermore, may not be appropriate over
the whole landscape, as they ignore the influence of climate
change. It was also thought that NDE could lead to a lack of
stakeholder participation, as culture is seen as unnatural. The
main conclusion was therefore that forest management deci-
sion making should take a more democratic and participatory
approach, one that is more adaptive and not dominated by
the desirability of scientific benchmarks of naturalness
(Klenk et al. 2009).
In conclusion, the preceding review and discussion ar-
ticles, which generally focused on North America, universally
acknowledged the potential benefits of NDE for biodiversity.
Despite the perceived ecological benefits, there was no con-
sistent attempt to make a link to existing ecological theories
or to form a theoretical basis for NDE. Indeed, the NRV was
the only consistent theoretical concept, although its validity
under current climate change was also questioned. Support-
ing theory was confined to general concepts such as Atti-
will’s (1994) disturbance-based management principles,
coarse–fine filters, and ecosystem resilience. Specific strat-
egies that were referred to were the multicohort model (Ber-
geron et al. 1999, 2002) and the patch-corridor-matrix model.
Although disturbance agents other than fire and agent inter-
action were discussed (e.g., Long 2009), the focus was gener-
ally on stand-replacing wildfire emulation. Some potential
nontimber economic benefits were discussed, for example re-
ducing the impact of atypical disturbance, seed bank renewal,
natural regeneration growth, site productivity enhancement,
shortened cutting cycles, reduced road requirement, and re-
versing paludification. Some social benefits were also antici-
pated, for example recreation services and greater potential
for participation. However, socioeconomic issues were also
consistently identified as a potential problem and barrier to
NDE. Mortality risk and inconsistent or reduced timber sup-
ply were potential economic burdens. In general, social bar-
riers, such as resistance to change, institutional climate, and
lack of participation, were regarded as more significant bar-
riers than theoretical or scientific issues.
Studies comparing NDE and conventional
even-aged management
Timber supply implications
The timber supply implications of NDE in comparison
with conventional management were studied by Armstrong
et al. (1999, 2003) and McCarney et al. (2008). Armstrong
et al. (1999) modelled the possible effects of NDE manage-
ment on timber supply over 100 years in Alberta, Canada
(Table 2). Allowable cuts were based on an annual nonspa-
tially modelled stand-replacing wildfire disturbance rate. Eco-
nomic loss was found to arise from harvesting immature
stands and the erratic timber supply caused by imitating a
fluctuating natural disturbance rate. It was concluded that the
timber supply chain would be severely negatively affected by
NDE (Armstrong et al. 1999).
Armstrong et al. (2003) examined how to reconcile the
needs NDE and timber supply in Alberta. Annual disturbance
rate (Armstrong et al. 1999) was discarded as the sole emula-
tion guide. The allowable harvest was instead determined by
the wildfire-driven NRV of habitat for five key vertebrate
species over 200 years. The habitat needs for the key species
were used as constraints on maximum timber yield. The non-
spatial model demonstrated that many NDE requirements,
such as preserving late-successional stands, may incur unrea-
sonable costs for the timber industry (Armstrong et al. 2003).
McCarney et al. (2008) augmented the model used by
Armstrong et al. (2003) to study the implications of NDE
for timber and carbon management in Alberta over 125 years.
It was shown that carbon sequestration initiatives may simul-
taneously assist in meeting ecological goals due to the preser-
vation of late-successional forest. Carbon price fluctuation
and unpredictability, with the potential for economic and en-
vironmental impact, was the main shortcoming of the pro-
posed scenario. Nevertheless, it was demonstrated that a
high carbon price may alleviate the economic impact of
NDE where dual management for both carbon and timber is
possible (McCarney et al. 2008).
The papers examining the effects of NDE on timber supply
are limited to a series of related studies on wildfire emulation
in Alberta, Canada. Over a time scale of 100–200 years, it
was demonstrated that NDE is likely to negatively impact
1194 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

management for timber, although carbon production may
provide a means to offset losses. Nontimber services were ei-
ther included as limits to timber production or evaluated with
carbon units. Carbon trading therefore represents one way in
which the nontimber values and services associated with
NDE could be evaluated. However, the extremely narrow fo-
cus of the studies prevents any general conclusion being
made as to the impact of NDE on timber supply.
Ecological implications: forest structure and composition
A total of 13 studies compared the ecological implications
of NDE-based management with conventional even-aged
management. Delong and Tanner (1996) compared the land-
scape-scale attributes caused by stand-replacing wildfire and
conventional harvesting over 200 years in lodgepole pine (Pi-
nus contorta Douglas ex Loudon) and white spruce (Picea
glauca (Moench) Voss) dominated forest in British Colum-
bia, Canada. Large stand-replacing wildfire has disappeared
from the landscape and dispersed harvesting produces a very
different landscape mosaic structure. The study investigated
whether large harvest blocks were geometrically consistent
with burnt patches and also compared the amount of undis-
turbed habitat. It was found that the harvested patches had
simpler boundaries and less remaining mature forest com-
pared with burnt patches. Furthermore, wildfire patches in-
creased in complexity with size, whereas harvested patches
greater than 500 ha became less complex. It was therefore
suggested that the aggregation of small harvest patches
(<500 ha), with green retention, may result in a more natural
landscape structure. Economic savings from operational cen-
tralization were a further benefit. NDE was thus found to be
ecologically legitimate and cost-effective, although care has
to be taken to ensure that large wildfires are indeed part of
the local disturbance regime (Delong and Tanner 1996).
Bergeron et al. (2004) modelled past, present, and future
stand-replacing wildfire frequency in Quebec, Canada, using
50 years of data from various literature sources. The natural
occurrence of post-fire stands at the landscape scale was de-
termined, which translates into allowable levels of even-aged
management in the management context. Using simulations,
they suggested that climatic effects and fire suppression have
extended the fire return interval, which has reduced the share
of post-fire stands in the landscape. It was thus concluded
that even-aged management could be used to produce a
more natural landscape structure, although this does not en-
sure a suitable share of late-successional forest (Bergeron et
al. 2004).
Belleau et al. (2007) used a simulation model to examine
the effects of wildfire frequency and severity over 1000 years
on the proportion, size, and spatial distribution of even-aged
stands in various Canadian boreal regions from northern Al-
berta to southeastern Labrador. Such information could be
used to guide even-aged management for NDE. The results
showed that where fire return periods are long, a sufficient
suite of ecological ecosystem services cannot be supported
by even-aged management. The use of green retention areas
in large clearcut areas was encouraged as a potential means
of offsetting the impacts. NDE was concluded to benefit bio-
diversity through keeping disturbance within the NRV and
using retention areas (Belleau et al. 2007).
Carlson and Kurz (2007) produced 150-year landscape-
scale simulations for Alberta, Canada, to assess the likely ef-
fects of increasing aggregation of management areas to emu-
late the spatial signature of wildfire. The currently prevalent
dispersed management paradigm has lead to landscape frag-
mentation. Spatially aggregated harvest was shown to result
in a less fragmented landscape structure, although late-suc-
cessional interior forest fell from 13% to below 1% due to
harvesting targeting mature stands. Protected zones and par-
tial cutting would therefore be needed to maintain a suitable
share of late-successional interior forest. It was also found
that aggregated harvest may yield inconsistent ratios of de-
ciduous to coniferous timber, which may not be viable for
the timber industry. Aggregating harvest blocks in this man-
ner was therefore found to have potential ecological and eco-
nomic risks despite reducing landscape fragmentation
(Carlson and Kurz 2007).
Shinneman et al. (2010) assessed NDE of wildfire using
200-year landscape simulations in Minnesota, USA, and On-
tario, Canada. Structural and compositional NRV benchmarks
were employed to compare the effects of NDE with conven-
tional regional management. Conventional management was
found to be significantly less effective at maintaining struc-
ture and composition than NDE. In addition, it was found
that conventional management was more likely to lead to
landscape-scale discontinuities in structure and composition
between separate forest management areas. NDE was thus
found to offer advantages over present practice at the land-
scape scale (Shinneman et al. 2010).
The preceding papers were in broad agreement that NDE
results in a more natural forest structure compared with
even-aged management. The main risks were loss of late-suc-
cessional habitat, economic impacts, and the difficulty of es-
tablishing the NRV. Many of the potential benefits discussed
in the review section were not considered, such as seed bank
impact and reversing paludification. Operational advantages
from centralized operations were discussed, but problems
with inconsistent timber supply, caused by altering stand
compositional trajectories, were also mentioned. Furthermore,
the landscape scale dominated the studies and little mention
was given to more local effects. As many of the benefits
mentioned in the reviews were at a local scale, for example
natural regeneration growth and site productivity enhance-
ment, more focus on this scale is urgently needed.
Ecological implications: species and communities
Wintle et al. (2005) produced a hybrid metapopulation and
forest dynamics model to compare the effects of current man-
agement with emulation of locally dominant stand-replacing
wildfire (Rowe and Scotter 1973) at the landscape scale in
Ontario, Canada, over 160 years. The Brown Creeper (Cer-
thia americana Bonaparte, 1838) was selected as an indicator
species because, as a cavity nester, it may be sensitive to the
loss of snags and old trees needed for foraging and nesting.
The model showed that NDE consisting of retention and nat-
ural regeneration (in line with local regulatory guidance) was
better for the Brown Creeper than standard management.
However, the validity of the model was questioned due to its
inherent assumptions and the stochastic nature of wildfire.
Nevertheless, the model suggested that NDE may deliver bio-
diversity-related ecosystem services (Wintle et al. 2005).
Harrison et al. (2005) investigated the impacts of partial
Kuuluvainen and Grenfell 1195
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

harvesting on songbird communities at the stand scale from
NDE in Alberta, Canada. The studies comprised part of the
EMEND (Ecosystem Management by Emulating Natural Dis-
turbance) project, which incorporates live tree retention.
Songbird counts were made over 2 years. The results sug-
gested that low retention levels, in the order of 10%–20%,
do not support avian diversity in the short term. Even with
high retention, certain species were still negatively impacted.
Mention was also made of the importance of the spatial con-
figuration of retention; more specifically, clumps of retention
may be better for songbirds than dispersed retention. In gen-
eral, however, retention was found to increase songbird num-
bers (Harrison et al. 2005).
Songbird communities in post-harvest and post-wildfire
stands with varying residual tree density were investigated in
the Rocky Mountains of British Columbia, Canada, by
Stuart-Smith et al. (2006). Stand-replacing wildfire was noted
as the historically dominant disturbance mode (Van Wagner
1995). The authors examined differences in songbird com-
munities in stands of similar time since disturbance and also
assessed if the communities resulting from each disturbance
type became more similar over time. It was found that logged
and burned stands have similar songbird communities and
that the degree of similarity was affected by the density and
type of retention trees. Broadleaved tree species were partic-
ularly important for many members of the focal group. How-
ever, some species were absent from harvested stands. The
importance of providing a range of retention densities and
species compositions was therefore demonstrated. It was con-
cluded that green retention may have a positive effect on
songbirds (Stuart-Smith et al. 2006).
Rempel et al. (2007) used a spatially explicit habitat model
to compare the impact of different forest management options
on songbird populations over three different time scales (100,
150, and 200 years) in Ontario, Canada. It was discussed that
using such long temporal scales introduced a large amount of
uncertainty into the model and that the assumed socioeco-
nomic stasis is unlikely to exist in reality. The habitat needs
of a focal group, consisting of 15 songbird species, marten,
and caribou, defined the boundaries of a “coarse filter”.It
was found that some management options that used nonna-
tural spatial patterns or led to atypical species composition
had a detrimental effect on the focal species. In contrast,
wildfire-inspired patterns incorporating expanses of intact in-
terior forest matrix were found have a positive effect. Emula-
tion of wildfire was therefore found to be more beneficial to
the focal species than other management options (Rempel et
al. 2007).
Van Wilgenburg and Hobson (2008) investigated the im-
pact of emulating wildfire using single-pass harvesting on
forest birds in Alberta, Canada. The study compared the
composition and abundance of avian communities at the
landscape scale in single-pass harvested and multipass har-
vested stands with those of post-fire stands, which had either
been salvage-logged or left untouched. It was found that after
single-pass harvesting, bird abundance and community com-
position were similar to levels and composition found after
wildfire. However, both management schemes led to some
deviation from the NRV. Retention trees were suggested as a
means to offset the impacts from harvesting. Nevertheless,
single-pass harvesting, which imitated wildfire, was found to
be more beneficial than locally used multipass harvesting for
forest birds (Van Wilgenburg and Hobson 2008).
Nielsen et al. (2008) used forest harvest simulations to pre-
dict the variability in suitable habitat for grizzly bears at the
landscape scale over a 100-year period in Alberta, Canada.
Periodic stand-replacing wildfire dominates the natural native
disturbance regime (Andison 1998). It was found that NDE
would increase the share of early-successional forest in the
landscape, which may benefit grizzly bears. However, road
density was found to increase with the size of harvested
areas, and the density after 100 years was over three times
the suggested tolerable maximum for grizzly bears. Success-
ful application of NDE would therefore necessitate limits on
the proliferation of roads (Nielsen et al. 2008).
Hogberg et al. (2002) researched the effects of clearcutting
with retention on bats in Alberta, Canada, as part of the
EMEND project. They surveyed the bats 1.5 years after har-
vest in locations that were either clearcut, on the edge of a
retention area, or on the edge of a clearcut area. Retention
was concluded to provide important commuting habitat for
smaller species, while benefits for larger bats were not evi-
dent in this study (Hogberg et al. 2002).
Hannam et al. (2006) researched the effects of partial har-
vesting for NDE on forest floor microbial community struc-
ture in Alberta, Canada. The study area was part of the
EMEND study area. The three stand treatments were 50% re-
tention, 20% retention, and clearcutting,, and the commun-
ities were sampled 4.5 and 5.5 years after harvesting. Stand
species composition was found to affect microbial commun-
ity composition, whereas the retention level had no effect. It
was therefore concluded that retention levels were not impor-
tant for microbes but retention species composition may be
significant (Hannam et al. 2006).
The preceding studies investigated the effects of wildfire
NDE on different focal species mostly in Canada. Retention
was found to have a generally positive effect. Many of the
studies focused on songbirds and universally demonstrated a
positive influence from NDE. However, the importance of
the spatial pattern of retention was underscored (Harrison et
al. 2005; Stuart-Smith et al. 2006; Rempel et al. 2007). Fur-
thermore, some inconsistencies were found between the avian
communities of emulation stands and natural post-fire stands
(Harrison et al. 2005; Stuart-Smith et al. 2006; Van Wilgen-
burg and Hobson 2008). More effort is therefore needed to
better understand the functional inconsistency between stands
affected by natural wildfire and emulation management. As-
sociated impacts such as road proliferation should also be
considered because these may offset the advantages of altered
management regimes (Nielsen et al. 2008).
Discussion
Natural disturbance emulation as management and
research agenda
NDE has been suggested as a general approach to ecologi-
cally sustainable forest ecosystem management (Attiwill
1994; Bergeron et al. 2002; Korpilahti and Kuuluvainen
2002; Gauthier et al. 2009). The seed for this development
of NDE approach in forest research and management was
sown by ecological research that discovered the pivotal role
of disturbance in structuring the diversity and functioning of
1196 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

ecological communities (Sousa 1984; Pickett and White
1986).
The underlying hypothesis of NDE is that the use of suit-
able management techniques that maintain more natural
structures (the “coarse filter”, Hunter et al. 1988) and associ-
ated dynamics (NRV) better safeguards biodiversity in com-
parison with conventional even-aged management. This
hypothesis is obviously based on knowledge and assumptions
concerning the evolutionary adaptation of species and eco-
systems to past disturbance regimes. This is a reasonable ar-
gument and can serve as a useful ad hoc approach to
sustainable forest management, but the danger may be that it
impedes the acquisition of deeper understanding of the man-
aged ecosystem. This can happen if NDE is based on an am-
biguous “nature knows best”attitude, without an effort to
disentangle the mechanisms of how disturbance relates to im-
portant ecosystem properties such as species diversity and re-
silience.
One advantage of NDE is that it offers both an integrative
(holistic) approach and strategic models for forest ecosystem
management. Established silvicultural tools with some modi-
fications can be readily used within the NDE framework, as
demonstrated by strategic models such as the multicohort and
ASIO models (see Figs. 2 and 3). These models facilitate the
application and development of NDE for practical manage-
ment by applying the principles of adaptive management and
“learning by doing”, with validation monitoring as a means
of evaluating the efficacy of management (Walters and Hol-
ling 1990). However, scientific research is indispensable for
acquiring a deeper understanding of the mechanisms through
which the ecosystem responds to natural disturbances and
NDE. This is also important because future environmental
conditions may not have direct analogues in the past to act
as model precedents (Keane et al. 2009).
Role of concepts, theories, and strategies
NDE research can be positioned in between two comple-
mentary approaches of ecological research, one integrative
(holistic), aiming at understanding the systemic properties
and responses of ecosystems, and the other analytical (reduc-
tionistic), “a science of parts”(Holling 1998). To reconcile
these two approaches in developing NDE, it is necessary to
have suitable conceptual and theoretical tools.
The Panarchy concept provides perhaps the most general
systemic framework for forest disturbance ecology and its de-
rivative NDE (Gunderson and Holling 2002; Drever et al.
2006). Panarchy is a conceptual framework describing the
cross-scale interactions and changes of complex adaptive sys-
tems. However, because of its holistic nature, it cannot be
easily translated into testable hypotheses. Other important
theoretical concepts related to NDE include the “coarse fil-
ter”(Hunter et al. 1988) and the NRV (Landres et al. 1999).
These two concepts are closely related: the NRV essentially
adds a temporal dimension to the coarse filter concept when
applied to NDE. Both concepts are fairly abstract, but they
can be regarded as indispensable for the development of
NDE-based management applications. The philosophy and
application of NRV in conservation and sustainable forest
management have been discussed in detail by Keane et al.
(2009).
The IDH (Connell 1978) represents an analytical approach
in that it proposes a within- and between-patch mechanism
explaining the maintenance of long-term species diversity in
ecological communities subject to disturbances (Wilson
1990; Roxburgh et al. 2004). The IDH allows related testable
hypotheses to be constructed concerning the mechanisms that
control how species diversity responds to disturbance. In this
context, “intermediacy”is defined in terms of repeatability,
severity, or spatial extent of disturbance (Shea et al. 2004).
The possible mechanisms behind the IDH have been further
explored in ecological literature (Roxburgh et al. 2004; Shea
et al. 2004). However, it is noteworthy that none of the re-
viewed discussion papers made explicit reference to the IDH.
It thus seems that there has been very little formal integration
of the IDH into NDE research, models, or policy. Neverthe-
less, IDH-based management recommendations have been
given, e.g., for rangeland ecosystems (Sasaki et al. 2009).
IDH may therefore be a fruitful avenue to explore in efforts
to define a theoretical framework for NDE research because
it allows formulation of testable hypotheses (Shea et al.
2004).
Parallel to these theoretical developments, strategic models
of how to apply NDE in actual forest management have been
proposed. Perhaps the most widely applied of these are the
“patch-corridor-matrix model”and “landscape continuum
model”, which are founded on the island biogeography (Mac-
Arthur and Wilson 1967) and metapopulation theories (Han-
ski 1999). The “ASIO model”, developed in Sweden (Rülcker
et al. 1994; Angelstam 1998), and the “multicohort model”,
developed in eastern Canada (Bergeron et al. 2002), represent
approaches that are specifically based on the disturbance dy-
namics of the boreal forest.
Although rarely explicitly mentioned, the thinking behind
these models is closely linked to the general NRV frame-
work. They therefore reflect the ecological conditions and
specific disturbance regime characteristics of the geographic
regions where they were developed. The ASIO model is per-
haps more appropriate for Fennoscandian landscapes where
low- to medium-severity fires dominate (Kuuluvainen and
Aakala 2011). Fennoscandian forest landscapes are inten-
sively managed and there is a desire to reinstate more natural
conditions. The multicohort model can be used in regions
where stand-replacing fires are frequent, such as in continen-
tal Canada and Siberia. The management situation in these
areas is quite different from those in Fennoscandia, as pris-
tine forest is being cut for the first time, and instead of resto-
ration, the motivation of applying NDE is to carry out timber
harvesting with minimum damage to pristine ecosystems.
It is noteworthy that the different strategic approaches to
NDE discussed can also be seen as complementary in a hier-
archical regional or landscape-level management framework
(Fig. 4). The patch-corridor-matrix and landscape continuum
models can be considered to provide a general framework in
hierarchical landscape planning with a permanent protection
network as its core element. The ASIO and multicohort mod-
els can be regarded as tools to create the desired matrix qual-
ities and a dynamic habitat network, including buffer zones
around and ecological corridors between protected areas
(Kuuluvainen et al. 2002; Bengtsson et al. 2003). In this
way, using NDE principles in forest management and restora-
tion, the connectivity properties of the landscape could possi-
bly be improved for many species groups. The overall aim of
Kuuluvainen and Grenfell 1197
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

this kind of landscape design would be to create synergy be-
tween the different tools in a hierarchical ecosystem manage-
ment scheme based on NDE (Fig. 4).
Surprisingly, none of the reviewed papers dealing with
strategic models referred to the IDH, despite the fact that
IDH directly addresses the mechanisms of disturbance im-
pacts on species diversity. In general, it seems that strategic
models of NDE have taken a practical approach to forest
management, while the links to ecological theory remain
scarce (also see Puettmann et al. 2009). This is unfortunate,
since strategic models combined with sound ecological
theory and research on underlying mechanisms could provide
a firmer foundation on which to develop novel forest man-
agement solutions based on NDE.
Knowledge basis and the concept of naturalness
General and local knowledge on unmanaged forest ecosys-
tems and their NRV is the key to successfully applying NDE.
Recently, the understanding of the intrinsic variability of bor-
eal forest structure and dynamics has greatly improved (Shor-
ohova et al. 2009; Kneeshaw et al. 2011; Kuuluvainen and
Aakala 2011); however, in many regions, significant local
knowledge deficiencies remain. For example, in Fennoscan-
dia, the geographic distribution of studies on natural forests
exhibits a pronounced bias towards the middle and northern
boreal zones, while data from southern boreal zone remain
scarce (Kuuluvainen and Aakala 2011). This is understand-
able, given the distribution of the remaining natural forests,
but is unfortunate, as the greatest challenges for biodiversity
conservation reside firmly in the southern boreal zone. Also,
there is an apparent lack of knowledge concerning the rarer
types of natural forest, a problem that can probably be gener-
alized across the entire circumboreal forest belt.
The question has been raised whether knowledge of past
ecological conditions can guide us in future forest manage-
ment under rapidly changing climatic conditions (Christensen
et al. 2007). It is clear, however, that a good understanding of
the processes and functioning of ecosystems is needed to pre-
dict their responses to changing environmental conditions.
Also, historical information is invaluable, as the currently ex-
isting forest ecosystems are those that have adapted to and
survived abrupt climate changes in the past. For example, it
has been suggested that the onset of the warm climatic period
and associated changes in vegetation, which occurred 4000–
8000 years ago, could be used as analogues of present cli-
mate warming and vegetation shifts (Heikkilä 2010). Accord-
ingly, Keane et al. (2009) argued that despite potential
shortcomings, the NRV will remain an inherently useful con-
cept; it thus follows that NDE must also retain a similar sta-
tus (Kuuluvainen 2009).
Rapid land use changes may, however, make expression of
the full NRV an unrealistic goal, and in instances where the
forest landscape is highly fragmented, other conservation
goals may therefore be more suitable. Rapid climatic and so-
cial changes further confound the problem of the appropriate-
ness of the NRV as a target, especially when extrapolated
over extensive time scales. However, in NDE, the NRV can
be regarded more as a reference than a monolithic goal. A
desirable future range of variability should therefore be a
flexible concept that is congizant of the NRV while also ac-
commodating knowledge advances and social changes using
the principles of adaptive management (Walters 1986). In
contrast, typical alternative forest conservation strategies
such as species-based or park-based approaches tend to be
more static, which means that they are less flexible to re-
spond to changes in the ecological and social environment
(Bengtsson et al. 2003; Thompson et al. 2009). Such static
approaches generally separate social and ecological systems,
whereas the future range of variability concept is based on
holistic socioecological systems. The flexibility of NDE-
based management, when applied in an adaptive manner,
suggests that it can remain useful despite rapid land use, so-
cial, and climate changes.
NDE versus conventional even-aged management
The reviewed studies were almost universally positive re-
garding the ecological benefits of NDE in comparison with
traditional even-aged management. However, doubt remains
about the impacts on timber supply and harvest limits. The
ecological studies considered economic impacts in a specula-
tive and qualitative manner, and ecological objectives were
only included in economic studies as a limit to production.
A more integrative approach, and one more congruous with
a“resilience thinking”systems theory approach, would be to
examine both economic and ecological effects together. This
relies on having a consistent means to quantify nontimber
ecosystem services. Furthermore, the studies generally fo-
cused on wildfire emulation at the landscape scale in either
Canada or the United States. The limited scope of the studies
therefore makes it difficult to draw general conclusions for
areas with contrasting disturbance regimes or for European
boreal regions. The local-scale effects of NDE need to be
more thoroughly examined, as many potential benefits will
be realized at this scale. Moreover, many slow ecological
changes, which could result in cumulative surprise shifts in
key ecosystem properties, rely on multiscale interaction and
processes that originate at the local scale. In summary,
Fig. 4. Different strategic approaches to natural disturbance emula-
tion (NDE) can also be seen as complementary hierarchical compo-
nents in a regional- or landscape-level forest ecosystem management
scheme. The patch-corridor-matrix model breaks the forest area
down into three zone types: permanently protected areas (the fine
filter), corridor and buffer areas, and the production forest matrix
(the coarse filter). In all of these zones, NDE principles can be im-
plemented using restoration, spatial planning, and strategic models
such as the ASIO and multicohort models (Figs. 2 and 3).
1198 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

although many advantages were shown over traditional man-
agement, examination of the multiscale impacts of NDE on
integrated socioeconomic systems is necessary to make a
more realistic appraisal (see Holling 2001).
Shortcomings of NDE studies
Some broad trends are obvious in the reviewed studies ad-
dressing NDE (Tables 1 and 2). Firstly, there appears to be a
relative excess of discussion and review type papers (16 out
of 29 papers), which provide little detail on the practical val-
idity of NDE efforts. This probably reflects the fact that the
idea of NDE is quite recent and the history of research is
thus also short (only commencing in the 1990s). Another rea-
son for the excess of review and discussion papers may be
that the extended spatial and temporal scales associated with
empirical experimentation on forest management and restora-
tion alternatives are methodically challenging (Harvey et al.
2002; Spence et al. 2002; Vanha-Majamaa et al. 2007; Peter-
son and Anderson 2009).This may also explain why a large
share of the studies comparing NDE and conventional ap-
proaches were modelling studies (Table 2).
Practically all of the reviewed studies came from North
America, particularly Canada. This probably reflects the fact
that NDE has been one of the main strategic approaches in a
significant national effort to develop sustainable forestry
(Burton et al. 2003). Although some theoretical work on
NDE has been done in Europe (Angelstam 1998; Kuuluvai-
nen 2002b, 2009; Shorohova et al. 2009), NDE research re-
sults from Eurasia were strikingly absent. There is therefore
a strong geographic bias towards North America in develop-
ing NDE approaches to boreal forest management. Even
within North America, studies on specific topics were subject
to a pronounced geographic bias. For example, quantitative
examinations of the effects of NDE on timber supply are lim-
ited to a series of studies on wildfire emulation in Alberta,
Canada. This is problematic because the disturbance regimes
are known to differ considerably both within the continent
and between the North American and Eurasian continents
(Kneeshaw et al. 2011). Experience from North America and
resultant management guidelines are therefore not necessarily
directly applicable in Scandinavia or Russia and vice versa.
Generalizations concerning the application of NDE may
therefore be deceptive.
The North American focus probably accounts for the bias
toward wildfire, the emulation of other disturbance factors re-
ceiving comparatively little attention (Table 2). This simpli-
fied approach is unviable, as it is increasingly recognized
that disturbance regimes are a product of the complex inter-
action of several coexisting biotic and abiotic disturbance
agents that relegates the status of fire to that of a single
agent, often having comparatively limited influence (Kuulu-
vainen 2002a, 2002b; Kneeshaw et al. 2011). A fundamental
knowledge deficit thus persists in both the co-occurrence and
relative importance of different disturbance factors in differ-
ent regions. In particular, we lack holistic empirical assess-
ments of all drivers of forested landscape dynamics.
Nevertheless, some proposed strategic models of NDE incor-
porate nonfire disturbances (Angelstam 1998; Seymour et al.
2002).
As could be expected, the study approach strongly corre-
lated with the spatial and temporal scale covered. Empirical
studies were limited to the stand scale and to short time peri-
ods, whereas simulation studies typically considered land-
scapes or larger areas and long time spans (Table 2). One of
the biggest challenges in studying the implications of NDE is
how to determine the meaningful spatial and temporal scales
that characterize the system dynamics (Habeeb et al. 2005). It
is obvious that the spatiotemporal scales commonly covered
by many studies are severely limited (Table 2). However, we
know that many important disturbance-related ecological
phenomena may become visible only at the landscape or re-
gional scale or over long time periods (at least tens of years)
(Franklin 1993).
Research challenges and future development
It is evident from the previous discussion that although a
significant amount of research and development has been
undertaken on the NDE approach, serious gaps still exist in
the knowledge base concerning the rigorous application of
NDE in forest management. The main research challenges
are suggested to be the following: (1) strengthening the
knowledge base on the NRV of forest ecosystems and evalu-
ating the validity of this information in the context of a
changing climate and intensified forest and land manage-
ment, (2) fostering multidisciplinary research that is better
connected to ecological theory and that aims at reconciling
integrative and analytical research approaches, and (3) better
integration of social aspects, adaptive management schemes,
and international collaboration in NDE research and initia-
tives.
(1) During the last 20 years, knowledge of the natural var-
iability of structure and function of the boreal forest has
vastly increased (for reviews, see Shorohova et al. 2009;
Kneeshaw et al. 2011; Kuuluvainen and Aakala 2011). How-
ever, because of the diversity of ecosystem types, major gaps
in local and regional understanding of forest NRV still re-
main. In exploited southern boreal forests, natural reference
landscapes may not exist anymore or they exist under the
threat of exploitation in the near future (Burnett et al. 2003).
This should be a strong incentive to intensify research efforts
to better understand the structure and functioning of the re-
maining unmanaged forest ecosystems across the entire bor-
eal zone. A combination of research strategies is often
needed, including the use of biological and anthropogenic ar-
chives and combining empirical research with modelling ap-
proaches (Kuuluvainen 2002b). Modelling and empirical
research should also be better integrated, allowing forecasts
of how forest ecosystems and their NRVs will respond to
changing climatic conditions and to ascertain the implications
for NDE-based management.
(2) Natural resource management has to take into account
the multiple dimensions of sustainability in regions and land-
scapes with a variety of goals for land use. For example, in
triad type land use solutions (Côté et al. 2010), NDE can be
used as one component in a complex land use situation, also
comprised of protected areas and areas with intensive bio-
mass production. In southern Scandinavia, protected areas
are scarce, private forest ownership dominates, and forest use
is intense. The question in this case is how NDE could con-
tribute to biodiversity restoration and conservation and what
the consequences would be in terms of economic output and
social values if implemented on a limited section of the forest
Kuuluvainen and Grenfell 1199
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

area (Fig. 4). This clearly demonstrates that questions about
how to implement NDE need to be considered in a multidis-
ciplinary context and covering spatiotemporal scales that are
large enough to be relevant from the management point of
view. This calls for large-scale multidisciplinary research and
adaptive management projects that also incorporate social di-
mensions (Spence et al. 2002; Peterson and Anderson 2009).
(3) The NDE approach has been criticized for being eco-
logically biased at the expense of socioeconomic considera-
tions (Klenk et al. 2009). Indeed, based on our literature
review, it seems that very little work has been done on the
social consequences of NDE, and no relevant studies directly
compared NDE with conventional management. The success
of the NDE approach will ultimately depend on social ac-
ceptability and the trade-offs between different ecosystem
services. There is therefore an urgent need to strengthen re-
search on the social aspects of NDE. For application of
NDE, social innovations may be as important as ecological
knowledge and management strategies.
Conclusions
NDE has been proposed as a general approach to ecologi-
cally sustainable forest management. The roots of the NDE
approach lie in the general knowledge and theory on distur-
bances as a central factor shaping the diversity of ecological
communities. Stemming from this ecological foundation and
subsequent theoretical work, a number of strategic applica-
tion-oriented models for NDE-based management have been
proposed. Unfortunately, these two lines of development, the
theoretical and the applied, have largely evolved in isolation
from each other. The strategic NDE models have taken a
practical approach to forest management with poorly defined
links to ecological theory. It seems evident that better inte-
gration of ecological theory and strategic management mod-
els could help to create more coherent and fruitful
approaches to research and management applications based
on NDE.
The reviewed NDE research literature showed a bias to-
ward review and discussion papers in comparison with papers
presenting actual results on the performance of NDE relative
to conventional even-aged management. The papers were
also strongly focused on northern North America and wild-
fire was the main disturbance agent emulated. Even within
northern North America, studies dealing with particular
topics were few and strongly restricted to specific regions.
Only theoretical studies on NDE in Eurasia have been pub-
lished. Within this limited scope, NDE was generally found
to have a positive effect on biodiversity in terms of forest
structure and species diversity when compared with conven-
tional even-aged management. Studies on timber supply and
social implications of NDE were few and precluded general-
izations.
We conclude that NDE provides a promising approach to
forest ecosystem management and a framework for associated
research. However, the ecological and economic performance
of NDE based management still remains poorly examined for
many boreal ecosystems, especially over extended spatial and
temporal scales. For future research, we propose a strategy
that fully takes into account the interdisciplinary nature of
forest management and that is cognizant of both social goals
and relevant ecological theory.
Acknowledgements
This work has greatly benefitted from longlasting collabo-
ration and discussions with researchers and forest managers
working on issues related to forest disturbance ecology and
sustainable management across the circumboreal zone. We
thank Ruut Rabinowitsch-Jokinen and Tuomas Aakala for
help in the literature review process. The research was partly
funded by the Finnish Ministry of Agriculture and Forestry
under the Forest Biodiversity Research Programme
(METSO).
References
Andison, D.W. 1998. Temporal patterns of age-class distributions on
foothills landscapes in Alberta. Ecography, 21(5): 543–550.
doi:10.1111/j.1600-0587.1998.tb00446.x.
Angelstam, P.K. 1998. Maintaining and restoring biodiversity in
European boreal forests by developing natural disturbance
regimes. J. Veg. Sci. 9(4): 593–602. doi:10.2307/3237275.
Armstrong, G.W., Cumming, S.G., and Adamowicz, W.L. 1999.
Timber supply implications of natural disturbance management.
For. Chron. 75(3): 497–504.
Armstrong, G.W., Adamowicz, W.L., Beck, J.A., Cumming, S.G.,
and Schmiegelow, F.K.A. 2003. Coarse filter ecosystem manage-
ment in a nonequilibrating forest. For. Sci. 49(2): 209–223.
Attiwill, P.M. 1994. The disturbance of forest ecosystems —the
ecological basis for conservative management. For. Ecol. Manage.
63(2–3): 247–300. doi:10.1016/0378-1127(94)90114-7.
Belleau, A., Bergeron, Y., Leduc, A., Gauthier, S., and Fall, A. 2007.
Using spatially explicit simulations to explore size distribution and
spacing of regenerating areas produced by wildfires: recommenda-
tions for designing harvest agglomerations for the Canadian boreal
forest. For. Chron. 83(1): 72–83.
Bengtsson, J., Angelstam, P., Elmqvist, T., Emanuelsson, U., Folke,
C., Ihse, M., Moberg, F., and Nyström, M. 2003. Reserves,
resilience and dynamics landscapes. Ambio, 32(6): 389–396.
PMID:14627367.
Bergeron, Y. 1991. The influence of island and mainland lakeshore
landscapes on boreal forest fire regimes. Ecology, 72(6): 1980–
1992. doi:10.2307/1941553.
Bergeron, Y., and Harvey, B. 1997. Basing silviculture on natural
ecosystem dynamics: an approach applied to the southern boreal
mixedwood forest of Quebec. For. Ecol. Manage. 92(1–3): 235–
242. doi:10.1016/S0378-1127(96)03924-2.
Bergeron, Y., and Leduc, A. 1998. Relationships between change in
fire frequency and mortality due to spruce budworm outbreak in
the southeastern Canadian boreal forest. J. Veg. Sci. 9(4): 492–
500. doi:10.2307/3237264.
Bergeron, Y., Harvey, B., Leduc, A., and Gauthier, S. 1999. Forest
management guidelines based on natural disturbance dynamics:
stand- and forest-level considerations. For. Chron. 75(1): 49–54.
Bergeron, Y., Leduc, A., Harvey, B.D., and Gauthier, S. 2002.
Natural fire regime: a guide for sustainable management of the
Canadian boreal forest. Silva Fenn. 36(1): 81–95.
Bergeron, Y., Flannigan, M., Gauthier, S., Leduc, A., and Lefort, P.
2004. Past, current and future fire frequency in the Canadian
boreal forest: implications for sustainable forest management.
Ambio, 33(6): 356–360. PMID:15387074.
Bergeron, Y., Drapeau, P., Gauthier, S., and Lecomte, N. 2007. Using
knowledge of natural disturbances to support sustainable forest
1200 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

management in the northern Clay Belt. For. Chron. 83(3): 326–
337.
Bouchard, M., Kneeshaw, D., and Bergeron, Y. 2005. Mortality and
stand renewal patterns following the last spruce budworm outbreak
in mixed forests of western Quebec. For. Ecol. Manage. 204(2–3):
297–313. doi:10.1016/j.foreco.2004.09.017.
Bouchard, M., Kneeshaw, D., and Bergeron, Y. 2006. Forest
dynamics after successive spruce budworm outbreaks in mixed-
wood forests. Ecology, 87(9): 2319–2329. doi:10.1890/0012-9658
(2006)87[2319:FDASSB]2.0.CO;2. PMID:16995632.
Bouchard, M., Kneeshaw, D., and Bergeron, Y. 2008. Ecosystem
management based on large-scale disturbance pulses: a case study
from sub-boreal forests of western Quebec (Canada). For. Ecol.
Manage. 256(10): 1734–1742. doi:10.1016/j.foreco.2008.05.044.
Burnett, C., Fall, A., Tomppo, E., and Kalliola, R. 2003. Monitoring
current status of and trends in boreal forest land use in Russian
Karelia. Conserv. Ecol. 7(2): article 8.
Burton, P.J., Messier, C., Smith, D.W., and Adamovwics, W.L.
(Editors). 2003. Towards sustainable management of the boreal
forest. NRC Research Press, Ottawa, Ont.
Burton, P.J., Messier, C., Adamowicz, W.L., and Kuuluvainen, T.
2006. Sustainable management of Canada’s boreal forests:
progress and prospects. Ecoscience, 13(2): 234–248. doi:10.
2980/i1195-6860-13-2-234.1.
Carlson, M., and Kurz, W.A. 2007. Approximating natural landscape
pattern using aggregated harvest. Can. J. For. Res. 37(10): 1846–
1853. doi:10.1139/X07-059.
Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X.,
Held, I., Jones, R., Kolli, R.K., Kwon, W.-T., Laprise, R., Magaña
Rueda, V., Mearns, L., Menéndez, C.G., Räisänen, J., Rinke, A.,
Sarr, A., and Whetton, P. 2007. Regional climate projections. In
Climate change 2007: the physical science basis. Edited by S.
Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B.
Averyt, M. Tignor, and H.L. Miller. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental
Panel on Climate Change. Cambridge University Press, Cam-
bridge, U.K.
Clark, J.S. 1989. Ecological disturbance as a renewal process: theory
and application to fire history. Oikos, 56(1): 17–30. doi:10.2307/
3566083.
Cogbill, C.V. 1985. Dynamics of the boreal forests of the Laurentian
Highlands, Canada. Can. J. For. Res. 15(1): 252–261. doi:10.1139/
x85-043.
Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs.
Science, 199(4335): 1302–1310. doi:10.1126/science.199.4335.
1302. PMID:17840770.
Côté, P., Tittler, R., Messier, C., Kneeshaw, D.D., Fall, A., and
Fortin, M.-J. 2010. Comparing different forest zoning options for
landscape-scale management of the boreal forest: possible benefits
of TRIAD. For. Ecol. Manage. 259(3): 418–427. doi:10.1016/j.
foreco.2009.10.038.
Cyr, D., Gauthier, S., Bergeron, Y., and Carcaillet, C. 2009. Forest
management is driving the eastern North American boreal forest
outside its natural range of variability. Front. Ecol. Environ. 7(10):
519–524. doi:10.1890/080088.
Dansereau, P., and Bergeron, Y. 1993. Fire history in the southern
boreal forest of northwestern Quebec. Can. J. For. Res. 23(1): 25–
32. doi:10.1139/x93-005.
Delong, S.C., and Tanner, D. 1996. Managing the pattern of forest
harvest: lessons from wildfire. Biodivers. Conserv. 5(10): 1191–
1205.
Diaci, J. (Editor). 2006. Nature-based forestry in central Europe:
alternatives to industrial forestry and strict preservation. Studia
Forestalia Slovenica, No. 126. Department of Forestry and
Renewable Forest Resources, Biotechnical Faculty, University of
Ljubljana, Slovenia.
Drever, C.R., Peterson, G., Messier, C., Bergeron, Y., and Flannigan,
M. 2006. Can forest management based on natural disturbances
maintain ecological resilience? Can. J. For. Res. 36(9): 2285–
2299. doi:10.1139/x06-132.
Esseen, P.-A., Ehnström, B., Ericson, L., and Sjöberg, K. 1997.
Boreal forests. Ecol. Bull. 46:16–47.
Euler, D.L., Henschel, C., and Clark, T. 2004. A conservation
perspective on emulating natural disturbance in the management of
boreal forests in Ontario. In Emulating natural forest landscape
disturbances: concepts and applications. Edited by A.H. Perera, L.
J. Buse, and M.G. Weber. Columbia University Press, N.Y.
pp. 191–199.
Forman, R.T.T. 1995. Land mosaics: the ecology of landscapes and
regions. Cambridge University Press, Cambridge, U.K.
Franklin, J.F. 1993. Preserving biodiversity: species, ecosystems, or
landscapes. Ecol. Appl. 3(2): 202–205. doi:10.2307/1941820.
Gauthier, S., Vaillancourt, M.-A., Leduc, A., DeGarndpre, L.,
Kneeshaw, D., Morin, H., Drapeau, P., and Bergeron, Y. 2009.
Ecosystem management in the boreal forest. Presses de l’Univer-
sité du Québec, Québec, Que.
Girardin, M.P., Tardif, J., Flannigan, M.D., and Bergeron, Y. 2004.
Multicentury reconstruction of the Canadian Drought Code from
eastern Canada and its relationship with paleoclimatic indices of
atmospheric circulation. Clim. Dyn. 23(2): 99–115. doi:10.1007/
s00382-004-0417-x.
Grenier, D.J., Bergeron, Y., Kneeshaw, D., and Gauthier, S. 2005.
Fire frequency for the transitional mixedwood forest of Temiskam-
ing, Quebec, Canada. Can. J. For. Res. 35(3): 656–666. doi:10.
1139/x05-005.
Gunderson, L., and Holling, C.S. (Editors). 2002. Panarchy: under-
standing transformations in human and natural systems. Island
Press, Washington, D.C.
Habeeb, R.L., Trebilco, J., Wotherspoon, T., and Johnson, C.R. 2005.
Determining natural scales of ecological systems. Ecol. Monogr.
75(4): 467–487. doi:10.1890/04-1415.
Hannam, K.D., Quideau, S.A., and Kishchuk, B.E. 2006. Forest floor
microbial communities in relation to stand composition and timber
harvesting in northern Alberta. Soil Biol. Biochem. 38(9): 2565–
2575. doi:10.1016/j.soilbio.2006.03.015.
Hanski, I. 1999. Metapopulation ecology. Oxford University Press,
Oxford, U.K.
Harrison, R.B., Schmiegelow, F.K.A., and Naidoo, R. 2005. Stand-
level response of breeding forest songbirds to multiple levels of
partial-cut harvest in four boreal forest types. Can. J. For. Res. 35
(7): 1553–1567. doi:10.1139/x05-076.
Harvey, B.D., Leduc, A., Gauthier, S., and Bergeron, Y. 2002. Stand-
landscape integration in natural disturbance-based management of
the southern boreal forest. For. Ecol. Manage. 155(1–3): 369–385.
doi:10.1016/S0378-1127(01)00573-4.
Hebert, D., Harvey, B., Wasel, S., Dzus, E.H., Donnelly, M., Robert,
J., and Chambers, F.H. 2003. Implementing sustainable forest
management: some case studies. In Towards sustainable manage-
ment of the boreal forest. Edited by P.J. Burton, C. Messier, D.W.
Smith, and W.L. Adamowicz. NRC Research Press, Ottawa, Ont.
pp. 893–953.
Heikkilä, M. 2010. Postglacial climate changes and vegetation
responses in northern Europe. Academic dissertation, Department
of Geology and Geosciences, University of Helsinki, Helsinki,
Finland.
Heinselman, M.L. 1973. Fire in the virgin forests of the Boundary
Waters Canoe Area, Minnesota. Quat. Res. 3(3): 329–382. doi:10.
1016/0033-5894(73)90003-3.
Kuuluvainen and Grenfell 1201
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

Hellberg, E., Niklasson, M., and Granström, A. 2004. Influence of
landscape structure on patterns of forest fires in boreal forest
landscapes in Sweden. Can. J. For. Res. 34(2): 332–338. doi:10.
1139/x03-175.
Hogberg, L.K., Patriquin, K.J., and Barclay, R.M.R. 2002. Use by
bats of patches of residual trees in logged areas of the Boreal
forest. Am. Midl. Nat. 148(2): 282–288. doi:10.1674/0003-0031
(2002)148[0282:UBBOPO]2.0.CO;2.
Holling, C.S. 1998. Two cultures of ecology. Conserv. Ecol. 2(2): 4.
Holling, C.S. 2001. Understanding the complexity of ecological,
ecological, and social systems. Ecosystems (N.Y.), 4(5): 390–405.
doi:10.1007/s10021-001-0101-5.
Hunter, M.L., Jr.. 1993. Natural fire regimes as spatial models for
managing boreal forests. Biol. Conserv. 65(2): 115–120. doi:10.
1016/0006-3207(93)90440-C.
Hunter, M.L., Jr., Jacobson, G.L., and Webb, T. 1988. Paleoecology
and the coarse filter approach in maintaining biological diversity.
Conserv. Biol. 2(4): 375–385. doi:10.1111/j.1523-1739.1988.
tb00202.x.
Johann, E. 2006. Historical development of nature-based forestry in
Central Europe. In Nature-based forestry in central Europe.
Alternatives to industrial forestry and strict preservation. Edited
by J. Diaci. Studia Forestalia Slovenica, No. 126. Department of
Forestry and Renewable Forest Resources, Biotechnical Faculty,
University of Ljubljana, Slovenia. pp. 1–17.
Johnson, E.A., and Van Wagner, C.E. 1985. The theory and use of
two fire history models. Can. J. For. Res. 15(1): 214–220. doi:10.
1139/x85-039.
Jonsson, B.G., Brumelis, G., and Kuuluvainen, T. 2011. Early
classical studies of forest ecology in Northern Europe. Scand. J.
For. Res. 26(Suppl. 10): 1–2. doi:10.1080/02827581.2011.
517946.
Kashian, D.M., Barnes, B.V., and Walker, W.S. 2003. Landscape
ecosystem of northern Lower Michigan and the occurrence and
management of the Kirtland’s warbler. For. Sci. 49: 140–159.
Keane, R.E., Hessburg, P.F., Landres, P.B., and Swanson, F.J. 2009.
The use of historical range and variability (HRV) in landscape
management. For. Ecol. Manage. 258(7): 1025–1037. doi:10.
1016/j.foreco.2009.05.035.
Klenk, N.L., Bull, G.Q., and MacLellan, J.I. 2009. The “emulation of
natural disturbance”(END) management approach in Canadian
forestry: a critical evaluation. For. Chron. 85: 440–445.
Kneeshaw, D., Bergeron, Y., and Kuuluvainen, T. 2011. Forest
ecosystem structure and disturbance dynamics across the circim-
boreal forest. In The SAGE handbook of biogeography. Edited by
A.C. Millington, M.B. Blumler, and U. Schickhoff. SAGE, Los
Angeles, Calif. pp. 263–280.
Kohm, K.A., and Franklin, J.F. (Editors). 1997 Creating a forestry for
the twenty-first century: the science of ecosystem management.
Island Press, Washington, D.C. pp. 263–280.
Korpilahti, E., and Kuuluvainen, T. (Editors). 2002. Disturbance
dynamics in boreal forests: defining the ecological basis of
restoration and management of biodiversity. Silva Fenn. 36(1).
Kuuluvainen, T. 1994. Gap disturbance, ground microtopography,
and the regeneration dynamics of boreal coniferous forests in
Finland: a review. Ann. Zool. Fenn. 31:35–51.
Kuuluvainen, T. 2002a. Disturbance dynamics in boreal forests:
defining the ecological basis of restoration and management of
biodiversity. Silva Fenn. 36:5–12.
Kuuluvainen, T. 2002b. Natural variability of forests as a reference for
restoring and managing biological diversity in boreal Fennoscan-
dia. Silva Fenn. 36(1): 97–125.
Kuuluvainen, T. 2009. Forest management and biodiversity con-
servation based on natural ecosystem dynamics in northern
Europe: the complexity challenge. Ambio, 38(6): 309–315.
doi:10.1579/08-A-490.1. PMID:19860154.
Kuuluvainen, T., and Aakala, T. 2011. Types of natural forest
dynamics in boreal Fennoscandia: a review and classification.
Silva Fenn. 45(5): 823–841.
Kuuluvainen, T., Aapala, K., Ahlroth, P., Kuusinen, M., Lindholm,
T., Sallantaus, T., Siitonen, J., and Tukia, H. 2002. Principles of
ecological restoration of boreal forested ecosystems: Finland as an
example. Silva Fenn. 36(1): 409–422.
Landres, P.B., Morgan, P., and Swanson, F.J. 1999. Overview of the
use of natural variability concepts in managing ecological systems.
Ecol. Appl. 9: 1179–1188.
Leibundgut, H. 1978. Die Waldflege. Verlag P. Haupt, Bern,
Stuttgart.
Leibundgut, H. 1981. Die Naturliche Wakdverjungung. Verlag P.
Haupt, Bern, Stuttgart.
Leopold, A. 1949. A Sand County almanac, with essays of
conservation from Round River. Ballantine, New York.
Lindenmayer, D., and Franklin, J.F. 2002. Conserving forest
biodiversity. A comprehensive multiscaled approach. Island Press,
Washington, D.C.
Long, J.N. 2009. Emulating natural disturbance regimes as a basis for
forest management: a North American view. For. Ecol. Manage.
257(9): 1868–1873. doi:10.1016/j.foreco.2008.12.019.
MacArthur, R.H., and Wilson, E.O. 1967. The theory of island
biogeography. Princeton University Press, Princeton, N.J.
McCarney, G.R., Armstrong, G.W., and Adamowicz, W.L. 2008.
Joint production of timber, carbon, and wildlife habitat in the
Canadian boreal plains. Can. J. For. Res. 38(6): 1478–1492.
doi:10.1139/X07-246.
McIntyre, S., and Hobbs, R. 1999. A framework for conceptualizing
human effects on landscapes and its relevance to management and
research models. Conserv. Biol. 13(6): 1282–1292. doi:10.1046/j.
1523-1739.1999.97509.x.
Morin, H., Laprise, D., and Bergeron, Y. 1993. Chronology of spruce
budworm outbreaks near Lake Duparquet, Abitibi region, Quebec.
Can. J. For. Res. 23(8): 1497–1506. doi:10.1139/x93-189.
Nielsen, S.E., Stenhouse, G.B., Beyer, H.L., Huettmann, F., and
Boyce, M.S. 2008. Can natural disturbance-based forestry rescue a
declining population of grizzly bears? Biol. Conserv. 141(9):
2193–2207. doi:10.1016/j.biocon.2008.06.020.
North, P.M., and Keeton, W.S. 2008. Emulating natural disturbance
regimes: an emerging approach for sustainable forest management.
In Patterns and processes in forest landscapes. Edited by R.
Lafortezza, J. Chen, G. Sanesi, and Th.R. Crow. Springer Science
+Business Media B.V., New York.
Perera, A.H., Buse, L.J., and Weber, M.G. (Editors). 2004. Emulating
natural forest landscape disturbances. Columbia University Press,
New York.
Peterson, C.E., and Anderson, P.D. 2009. Large-scale interdisciplin-
ary experiments inform current and future forestry management
options in the U.S. Pacific Northwest. For. Ecol. Manage. 258:
409–414.
Pickett, S.T.A., and White, P.S. 1986. The ecology of natural
disturbance and patch dynamics. Academic Press, New York.
Pommerening, A., and Murphy, S.T. 2004. A review of the history,
definitions and methods of continuous cover forestry with special
attention to afforestation and restocking. Forestry, 77(1): 27–44.
doi:10.1093/forestry/77.1.27.
Puettmann, K.J., Coates, K.D., and Messier, C. 2009. A critique of
silviculture: managing for complexity. Island Press, Washington,
D.C.
Pullin, A.S., and Stewart, G.B. 2006. Guidelines for systematic
review in conservation and environmental management. Conserv.
1202 Can. J. For. Res. Vol. 42, 2012
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.

Biol. 20(6): 1647–1656. doi:10.1111/j.1523-1739.2006.00485.x.
PMID:17181800.
Pullin, A.S., Knight, T.M., and Watkinson, A.R. 2009. Linking
reductionist science and holistic policy using systematic reviews:
unpacking environmental policy questions to construct an
evidence-based framework. J. Appl. Ecol. 46(5): 970–975.
doi:10.1111/j.1365-2664.2009.01704.x.
Rempel, R.S., Baker, J., Elkie, P.C., Gluck, M.J., Jackson, J.,
Kushneriuk, R.S., Moore, T., and Perera, A.H. 2007. Forest policy
scenario analysis: sensitivity of songbird community to changes in
forest cover amount and configuration. Avian Conserv. Ecol. 2(1):
5.
Roccaforte, J.P., Fule, P.Z., and Covington, W.W. 2008. Landscape-
scale changes in canopy fuels and potential fire behaviour
following ponderosa pine restoration treatments. Int. J. Wildland
Fire, 17(2): 293–303. doi:10.1071/WF06120.
Rowe, J.S., and Scotter, G.W. 1973. Fire in the boreal forest. Quat.
Res. 3(3): 444–464. doi:10.1016/0033-5894(73)90008-2.
Roxburgh, S.H., Shea, K., and Wilson, J.B. 2004. The intermediate
disturbance hypothesis: patch dynamics and mechanisms of
species coexistence. Ecology, 85(2): 359–371. doi:10.1890/03-
0266.
Rülcker, C., Angelstam, P., and Rosenberg, P. 1994. Naturlig
branddynamik kan styra naturvård och skogskötsel in boreal
skog. SkogForsk, 8.
Samways, M.J. 2007. Insect conservation: a synthetic management
approach. Annu. Rev. Entomol. 52(1): 465–487. doi:10.1146/
annurev.ento.52.110405.091317. PMID:16968205.
Sasaki, T., Okubo, S., Okayasu, T., Jamsran, U., Ohkuro, T., and
Takeuchi, K. 2009. Management applicability of the intermediate
disturbance hypothesis across Mongolian rangeland ecosystems.
Ecol. Appl. 19(2): 423–432. doi:10.1890/08-0144.1. PMID:
19323200.
Seymour, R.S., White, A.S., and deMaynadier, P.G. 2002. Natural
disturbance regimes in northeastern North America —evaluating
silvicultural systems using natural scales and frequencies. For.
Ecol. Manage. 155(1–3): 357–367. doi:10.1016/S0378-1127(01)
00572-2.
Shea, K., Roxburgh, S.H., and Rauschert, S.J. 2004. Moving from
pattern to process: coexistence mechanisms under intermediate
disturbance regimes. Ecol. Lett. 7(6): 491–508. doi:10.1111/j.
1461-0248.2004.00600.x.
Shinneman, D.J., Cornett, M.W., and Palik, B.J. 2010. Simulating
restoration strategies for a southern boreal forest landscape with
complex land ownership patterns. For. Ecol. Manage. 259(3): 446–
458. doi:10.1016/j.foreco.2009.10.042.
Shorohova, E., Kuuluvainen, T., Kangur, A., and Jogiste, K. 2009.
Natural stand structures, disturbance regimes and successional
dynamics in the Eurasian boreal forests: a review with special
reference to Russian studies. Ann. For. Sci. 66(2): 201. doi:10.
1051/forest/2008083.
Sirén, G. 1955. The development of spruce forest on raw humus sites
and its ecology. Acta For. Fenn. 62(4).
Sousa, W.P. 1984. The role of disturbance in natural communities.
Annu. Rev. Ecol. Syst. 15(1): 353–391. doi:10.1146/annurev.es.
15.110184.002033.
Spence, J.R. 2001. The new boreal forestry: adjusting timber
management to accommodate biodiversity. Trends Ecol. Evol. 16
(11): 591–593. doi:10.1016/S0169-5347(01)02335-7.
Spence, J.W., Volney, J.A., Sidders, D., Luchkow, S., Vinge, T.,
Oberle, F., Gilmore, D., Bielech, J.P., Wearmouth, P., Edwards, J.,
Bothwell, P., Shorthouse, D., Wilkinson, D., and Brais, S. 2002.
The EMEND experience. In Advances in Forest Management:
From Knowledge to Practice, Proceedings of SFMN Conference,
13–15 November, Edmonton, Alta. Edited by T Veeman et al.
SFMN Network, Edmonton, Alta. pp. 40–44.
Stuart-Smith, A.K., Hayes, J.P., and Schieck, J. 2006. The influence
of wildfire, logging and residual tree density on bird communities
in the northern Rocky Mountains. For. Ecol. Manage. 231(1–3):
1–17. doi:10.1016/j.foreco.2006.02.053.
Thompson, J.R., Duncan, S.L., and Johnson, K.N. 2009. Is there
potential for the historical range of variability to guide conserva-
tion given the social range of variability? Ecol. Soc. 14(1): 18.
Thorpe, H.C., and Thomas, S.C. 2007. Partial harvesting in the
Canadian boreal: success will depend on stand dynamic responses.
For. Chron. 83(3): 319–325.
Van Wagner, C.E. 1995. Analysis of fire history for Banff, Jasper, and
Kootenay National Parks. Report to Parks Canada, Banff, Alta.
Van Wilgenburg, S.L., and Hobson, K.A. 2008. Landscape-scale
disturbance and boreal forest birds: can large single-pass harvest
approximate fires? For. Ecol. Manage. 256(1–2): 136–146. doi:10.
1016/j.foreco.2008.04.017.
Vanha-Majamaa, I., Lilja, S., Ryömä, R., Kotiaho, J.S., Laaka-
Lindberg, S., Lindberg, H., Puttonen, P., Tamminen, P., Toivanen,
T., and Kuuluvainen, T. 2007. Rehabilitating boreal forest
structure and species composition in Finland through logging,
dead wood creation and fire: the EVO experiment. For. Ecol.
Manage. 250(1–2): 77–88. doi:10.1016/j.foreco.2007.03.012.
Walker, B., and Salt, D. 2006. Resilience thinking. Sustaining
ecosystems and people in a changing world. Island Press,
Washington, D.C.
Walters, C.J. 1986. Adaptive management of renewable resources.
McGraw Hill, New York.
Walters, C.J., and Holling, C.S. 1990. Large-scale management
experiments and learning by doing. Ecology, 71(6): 2060–2068.
doi:10.2307/1938620.
Wiens, J.A. 1997. The emerging role of patchiness in conservation
biology. In The ecological basis of conservation: heterogeneity,
ecosystems, and biodiversity. Edited by S.T.A. Pickett, R.S.
Ostfelt, M. Sachak, and G.E. Likens. Chapman & Hall, New York.
pp. 93–107.
Wilson, J.B. 1990. Mechanisms of species coexistence: twelve
explanations for Hutchinson’s“paradox of the plankton”: evidence
from New Zealand plant communities. N.Z. J. Ecol. 13:17–42.
Wilson, J.B. 1994. The ‘intermediate disturbance hypothesis’of
species coexistence is based on patch dynamics. N.Z. J. Ecol. 18
(2): 176–181.
Wintle, B.A., Bekessy, S.A., Venier, L.A., Pearce, J.L., and
Chisholm, R.A. 2005. Utility of dynamic-landscape metapopula-
tion models for sustainable forest management. Conserv. Biol. 19
(6): 1930–1943. doi:10.1111/j.1523-1739.2005.00276.x.
Zackrisson, O. 1977. The influence of forest fires on North Swedish
boreal forest. Oikos, 29(1): 22–32. doi:10.2307/3543289.
Zwolak, R. 2009. A meta-analysis of the effects of wildfire,
clearcutting, and partial harvest on the abundance of North
American small mammals. For. Ecol. Manage. 258(5): 539–545.
doi:10.1016/j.foreco.2009.05.033.
Kuuluvainen and Grenfell 1203
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13
For personal use only.