Effects of Habitat Fragmentation On Biod
Effects of Habitat Fragmentation On Biod
Effects of Habitat Fragmentation On Biod
34:487–515
doi: 10.1146/annurev.ecolsys.34.011802.132419
Copyright ° c 2003 by Annual Reviews. All rights reserved
First published online as a Review in Advance on August 14, 2003
Lenore Fahrig
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
Key Words habitat loss, landscape scale, habitat configuration, patch size, patch
isolation, extinction threshold, landscape complementation
■ Abstract The literature on effects of habitat fragmentation on biodiversity is
huge. It is also very diverse, with different authors measuring fragmentation in dif-
ferent ways and, as a consequence, drawing different conclusions regarding both the
magnitude and direction of its effects. Habitat fragmentation is usually defined as a
landscape-scale process involving both habitat loss and the breaking apart of habi-
tat. Results of empirical studies of habitat fragmentation are often difficult to inter-
pret because (a) many researchers measure fragmentation at the patch scale, not the
landscape scale and (b) most researchers measure fragmentation in ways that do not
distinguish between habitat loss and habitat fragmentation per se, i.e., the breaking
apart of habitat after controlling for habitat loss. Empirical studies to date suggest
that habitat loss has large, consistently negative effects on biodiversity. Habitat frag-
mentation per se has much weaker effects on biodiversity that are at least as likely
to be positive as negative. Therefore, to correctly interpret the influence of habitat
fragmentation on biodiversity, the effects of these two components of fragmentation
must be measured independently. More studies of the independent effects of habitat
loss and fragmentation per se are needed to determine the factors that lead to positive
versus negative effects of fragmentation per se. I suggest that the term “fragmen-
tation” should be reserved for the breaking apart of habitat, independent of habitat
loss.
INTRODUCTION
A recent search of the Cambridge Scientific Abstracts database revealed over 1600
articles containing the phrase “habitat fragmentation.” The task of reviewing this
literature is daunting not only because of its size but also because different authors
use different definitions of habitat fragmentation, and they measure fragmentation
in different ways and at different spatial scales.
This diversity of definitions of habitat fragmentation can be readily seen in
the titles of some articles. For example, “Impacts of habitat fragmentation and
1543-592X/03/1215-0487$14.00 487
488 FAHRIG
patch size. . .” (Collingham & Huntly 2000) suggests that habitat fragmenta-
tion and patch size are two different things. However, other authors actually use
patch size to measure habitat fragmentation (e.g., Golden & Crist 2000, Hovel &
Lipicus 2001). “The effects of forest fragmentation and isolation. . .” (Good-
man & Rakotodravony 2000) suggests that forest fragmentation and isolation are
different, in contrast to authors who use forest isolation as a measure of for-
est fragmentation (e.g., Mossman & Waser 2001, Rukke 2000). “Effect of land
cover, habitat fragmentation, and. . .” (Laakkonen et al. 2001) contrasts with many
authors who equate landscape fragmentation with land cover (e.g., Carlson &
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
Hartman 2001; Fuller 2001; Gibbs 1998, 2001; Golden & Crist 2000; Hargis
et al. 1999; Robinson et al. 1995; Summerville & Crist 2001; Virgós 2001). “The
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Abundance/ Richness/ Presence/ Fitness Genetic Species Extinction/ Individual Movement/ Population
Fragmentation (predictor) density diversity absence measures variability interactions turnover habitat use dispersal growth
variables (35) (28) (26) (15) (12) (10) (8) (5) (4) (3)
489
490 FAHRIG
Fragmentation as Process
Habitat fragmentation is often defined as a process during which “a large expanse
of habitat is transformed into a number of smaller patches of smaller total area,
isolated from each other by a matrix of habitats unlike the original” (Wilcove et al.
1986) (Figure 1). By this definition, a landscape can be qualitatively categorized
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
only two, i.e., one continuous landscape and one fragmented landscape. With such a
design, inferences about the effects of fragmentation are weak. Apparent effects of
fragmentation could easily be due to other differences between the landscapes. For
example, Mac Nally et al. (2000) found consistent vegetation differences between
fragments and reference sites and concluded that apparent effects of fragmentation
on birds could be due to preexisting habitat differences between the two landscapes.
Second, this characterization of habitat fragmentation is strictly qualitative, i.e.,
each landscape can be in only one of two states, continuous or fragmented. This
design does not permit one to study the relationship between the degree of habitat
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
reason is that when ecologists think of fragmentation, the word invokes more than
habitat removal: “fragmentation . . . not only causes loss of the amount of habi-
tat, but by creating small, isolated patches it also changes the properties of the
remaining habitat” (van den Berg et al. 2001).
Habitat can be removed from a landscape in many different ways, resulting in
many different spatial patterns (Figure 2). Do some patterns represent a higher
degree of fragmentation than others, and does this have implications for biodi-
versity? If the answer to either of these questions is “no,” then the concept of
fragmentation is redundant with habitat loss. The assertion that habitat fragmen-
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
tation means something more than habitat loss depends on the existence of effects
of fragmentation on biodiversity that can be attributed to changes in the pattern
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
of habitat that are independent of habitat loss. Therefore, many researchers define
habitat fragmentation as an aspect of habitat configuration.
Figure 2 Illustration of habitat loss resulting in some, but not all, of the other three
expected effects of habitat fragmentation on landscape pattern. Expected effects are
(a) an increase in the number of patches, (b) a decrease in mean patch size, and
(c) an increase in mean patch isolation (nearest neighbor distance). Actual changes are
indicated by arrows.
494 FAHRIG
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Figure 3 Illustration of the typical relationships between habitat amount and various mea-
sures of fragmentation. Individual data points correspond to individual landscapes. Based on
relationships in Bélisle et al. (2001), Boulinier et al. (2001), Drolet et al. (1999), Gustafson
(1998), Haines-Young & Chopping (1996), Hargis et al. (1998), Robinson et al. (1995),
Schumaker (1996), Trzcinski et al. (1999), and Wickham et al. (1999).
EFFECTS OF HABITAT FRAGMENTATION 495
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Figure 4 (A) Patch-scale study. Each observation represents the information from a single
patch. Only one landscape is studied, so sample size for landscape-scale inferences is one.
(B) Landscape-scale study. Each observation represents the information from a single land-
scape. Multiple landscapes, with different structures, are studied. Here, sample size for
landscape-scale inferences is four.
estimated that more than 57% of all fragmentation studies are at the patch scale.
Some researchers even refer to patch-scale measures as landscape features (e.g.,
Fernandez-Juricic 2000, Schweiger et al. 2000).
Figure 5 Both habitat loss and habitat fragmentation per se (independent of habitat
loss) result in smaller patches. Therefore, patch size itself is ambiguous as a mea-
sure of either habitat amount or habitat fragmentation per se. Note also that habitat
fragmentation per se leads to reduced patch isolation.
Figure 6 Landscape in southern Ontario (from Tischendorf 2001) showing that regions
where forest patches (black areas) are small typically correspond to regions where there is
little forest. Compare (A) and (B), where (A) has small patches and less than 5% forest and
(B) has larger patches and approximately 50% forest.
or “nearest-neighbor distance” (e.g., Delin & Andrén 1999, Haig et al. 2000, Hargis
et al. 1999). Patches with small nearest-neighbor distances are typically situated in
landscapes containing more habitat than are patches with large nearest-neighbor
distances (Figure 7), so in most situations this measure of isolation is related to
habitat amount in the landscape. Another common measure of patch isolation is
the inverse of the amount of habitat within some distance of the patch in question
(e.g., Kinnunen et al. 1996, Magura et al. 2001, Miyashita et al. 1998). In other
words, patch isolation is measured as habitat amount at the landscape scale. All
other measures of patch isolation are a combination of distances to other patches
and sizes of those patches (or the populations they contain) in the surrounding
landscape (reviewed in Bender et al. 2003). As such they are all measures of the
amount of habitat in the surrounding landscape.
498 FAHRIG
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Figure 7 Illustration of the relationship between patch isolation and amount of habitat in
the landscape immediately surrounding the patch. Gray areas are forest. Isolated patches
(black patches labeled “I”) are situated in landscapes (circles) containing less forest than are
less isolated patches (black patches labeled “N”).
ON BIODIVERSITY
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
In this section I review the empirical evidence for effects of habitat fragmentation
on biodiversity. This review is not limited to the 100 papers summarized in Table
1. The fragmentation literature can be distilled into two major effects: the gener-
ally strong negative effect of habitat loss on biodiversity, and the much weaker,
positive or negative effect of fragmentation per se on biodiversity. Because the
effect of fragmentation per se is weaker than the effect of habitat loss, to detect the
effect of fragmentation per se, the effect of habitat loss must be experimentally or
statistically controlled.
and turnover rate of forest birds in 214 landscapes. Because mean patch size had a
0.94 correlation with forest amount in their study, this result most likely represents
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Patch isolation effects Patch isolation is a measure of the lack of habitat in the
landscape surrounding the patch (Figure 7). Therefore, the many studies that have
shown negative effects of patch isolation on species richness or presence/absence
represent further evidence for the strong negative effect of landscape-scale habitat
loss on biodiversity (e.g., McCoy & Mushinsky 1999, Rukke 2000, Virgós 2001).
Bender et al. (2003) and Tischendorf et al. (2003) conducted simulation analyses
to determine which patch isolation measures are most strongly related to movement
of animals between patches. They found that the “buffer” measures, i.e., amount
of habitat within a given buffer around the patch, were best. This suggests a strong
effect of habitat amount on interpatch movement. It also suggests, again, that
effects of patch isolation and landscape-scale habitat amount are equivalent.
Patch size effects Individual species have minimum patch size requirements (e.g.,
Diaz et al. 2000). Therefore, smaller patches generally contain fewer species than
larger patches (Debinski & Holt 2000), and the set of species on smaller patches
is often a more-or-less predictable subset of the species on larger patches (e.g.,
Ganzhorn & Eisenbeiß 2001, Kolozsvary & Swihart 1999, Vallan 2000). Similarly,
the amount of habitat on a landscape required for species occurrence there differs
among species (Gibbs 1998, Vance et al. 2003), so landscapes with less habitat
should contain a subset of the species found in landscapes with more habitat.
Despite this apparent correspondence between patch- and landscape-scale ef-
fects, the landscape-scale interpretation of patch size effects depends on the land-
scape context of the patch. For example, Donovan et al. (1995) found that forest
birds had lower reproductive rates in small patches than in large patches. If small
patches occur in areas with less forest, the reduced reproductive rate may not be
the result of patch size, but may result from larger populations of nest predators
and brood parasites that occur in landscapes with more open habitat (Hartley &
Hunter 1998, Robinson et al. 1995, Schmiegelow & Mönkkönen 2002).
that species in the landscape. However, several theoretical studies suggest that the
relationship is not proportional; they predict a threshold habitat level below which
the population cannot sustain itself, termed the extinction threshold (Bascompte
& Solé 1996, Boswell et al. 1998, Fahrig 2001, Flather & Bevers 2002, Hill &
Caswell 1999, Lande 1987, With & King 1999; Figure 8). There have been very
few direct empirical tests of the extinction threshold hypothesis (but see Jansson
& Angelstam 1999).
Note that the predicted occurrence of the extinction threshold results from habi-
tat loss, not habitat fragmentation per se. Theoretical studies suggest that habitat
fragmentation per se can affect where the extinction threshold occurs on the habi-
tat amount axis. Also, the effects of habitat fragmentation per se are predicted to
increase below some level of habitat loss (see The 20–30% Threshold, below).
However, the occurrence of the extinction threshold is a response to habitat loss,
not fragmentation per se. This has led to some ambiguity in interpretation of empir-
ical literature. For example, Virgós (2001) found that patch isolation affects bad-
ger density only for patches in landscapes with <20% forest cover. As explained
above, patch isolation is typically an index of habitat amount at the landscape scale.
Therefore, this result probably suggests a threshold effect of forest loss on badger
density. This conclusion is different from that of the author, who interpreted the
isolation effect as an effect of habitat configuration. The interpretation is ambigu-
ous because the relationship between habitat amount and patch isolation was not
statistically controlled in this study. Similarly, Andrén (1994) reviewed patch size
and patch isolation effects on population density and concluded that these effects
502 FAHRIG
increase below a threshold amount of habitat in the landscape. Because patch size
and isolation can be indicators of habitat amount at a landscape scale (see Patch
size: An Ambiguous Measure of Fragmentation and Patch Isolation: A Measure
of Habitat Amount, above), this result could be interpreted as an intensification
of the effects of habitat loss at low habitat levels, i.e., it supports the extinction
threshold hypothesis (Figure 8). This result has also been viewed as evidence for
configuration effects below a threshold habitat level (e.g., Flather & Bevers 2002,
Villard et al. 1999). Again, the interpretation is ambiguous because the relation-
ships between patch size and isolation and amount of habitat surrounding each
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
TABLE 2 Summary of empirical studies that examined the effects of habitat fragmentation per se, i.e., controlling for effects of habitat
amount on biodiversity
Direction of effect(s) of
Relative effects of habitat loss versus fragmentation per se on
Study Taxa and response variable(s) habitat fragmentation per se biodiversity
503
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
504
TABLE 2 (Continued)
FAHRIG
Direction of effect(s) of
Relative effects of habitat loss versus fragmentation per se on
Study Taxa and response variable(s) habitat fragmentation per se biodiversity
Hovel & Lipcius 2001 Blue crab: juvenile survival, n.a.a 1 positive (juvenile
adult (predator) density survival),
1 negative (adult
density)
Tscharntke et al. 2002 Butterflies: species richness, n.a.a 2 positive (total
endangered species richness and endangered
species richness)
Flather et al. 1999 Forest birds: abundance Amount ≫ fragmentation No effect
Studies in experimental landscapes:
Collins & Barrett 1997 Meadow vole: density n.a.a Positive
Wolff et al. 1997 Gray-tailed vole: abundance, Fragmentation > amount Positive
density, reproductive rate,
recruitment
Collinge & Forman 1998 Grassland insects: abundance, Not stated Positive
species richness
Caley et al. 2001 8 coral commensals: species Amount ≫ fragmentation 1 positive
richness and abundance
With et al. 2002 Clover insects: spatial aggregation Amount ≫ fragmentation n.a.b
a
Habitat amount was held constant; only fragmentation was varied.
b
Response variable was spatial distribution of the insects.
EFFECTS OF HABITAT FRAGMENTATION 505
The overall result from these studies is that habitat loss has a much larger
effect than habitat fragmentation per se on biodiversity measures (Table 2). When
fragmentation per se did have an effect, it was at least as likely to be positive
as negative (Table 2). Given the relatively small number of studies and the large
variation in conditions among studies, it is not possible to tease apart the factors
that lead to positive versus negative effects of fragmentation per se. However,
the positive effects of fragmentation can not be explained as merely responses
by “weedy,” habitat generalist species. For example, the results reported from
McGarigal & McComb (1995) are specifically limited to late-seral forest species,
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
species.
THE 20–30% THRESHOLD Some theoretical studies suggest that the effects of frag-
mentation per se should become apparent only at low levels of habitat amount,
below approximately 20–30% habitat on the landscape (Fahrig 1998, Flather &
Bevers 2002). To date, there is no convincing empirical evidence for this prediction.
If the threshold does occur, it should result in a statistical interaction effect be-
tween habitat amount and habitat fragmentation per se; such an interaction would
indicate that the effect of fragmentation per se depends on the amount of habitat
in the landscape. Trzcinski et al. (1999) tested for this interaction effect but found
no evidence for it. The hypothesis that fragmentation effects increase below a
threshold of habitat amount has not yet been adequately tested.
sented the distance from a patch to its neighbor(s), not the distance to a mainland.
Because of its roots in island biogeography, isolation was viewed as represent-
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
There are at least four additional possible reasons for positive effects of habitat
fragmentation per se on biodiversity. First, Bowman et al. (2002) argued that, for
many species, immigration rate is a function of the linear dimension of a habitat
patch rather than the area of the patch. For these species, overall immigration rate
should be higher when the landscape is comprised of a larger number of smaller
patches (higher fragmentation per se) than when it is comprised of a smaller num-
ber of larger patches. In situations where immigration is an important determinant
of population density, this could result in a positive effect of fragmentation per se
on density.
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
se. In fact, the effects of fragmentation per se are absent or too small to be detected
in most empirical tests to date. This is in contrast to several theoretical predictions
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
(Burkey 1999, Hill & Caswell 1999, Urban & Keitt 2001, With & King 1999) and
has important implications for conservation. It suggests that conservation efforts
should focus on habitat preservation and restoration. It also suggests that research
in support of particular conservation problems should focus on determining the
amount of habitat required for conservation of the species of concern. The fact
that effects of fragmentation per se are usually small and at least as likely to be
positive as negative suggests that conservation actions that attempt to minimize
fragmentation (for a given habitat amount) may often be ineffectual.
Note, however, that this conclusion is preliminary because there are still only a
small number of relevant empirical studies. To my knowledge there are, to date, no
studies in tropical regions of the effects of forest fragmentation per se (controlling
for habitat loss). Laurance et al. (2002) concluded that in Brazilian tropical forest
there are strong negative effects of forest edge on several taxa. These effects
are apparently much stronger than negative edge effects in temperate systems
(Kremsater & Bunnell 1999). Negative edge effects could translate into a negative
effect of fragmentation per se at the landscape scale because fragmentation per
se increases the amount of edge on the landscape. This suggests that effects of
fragmentation per se may be greater in tropical systems than in temperate systems.
This prediction remains to be tested.
Third, ambiguous empirical results could lead to errors in modeling studies. For
example, Donovan & Lamberson (2001) constructed a model to look at the effects
of habitat fragmentation on population growth rate. They held amount of habitat
constant and varied mean patch size. For input parameters they used empirical work
suggesting that reproductive success increases with increasing patch size. However,
as they point out, in these empirical studies patch size was highly correlated with
habitat amount in the surrounding landscape. It is not known whether reproductive
success increases with increasing patch size when habitat amount in the landscape
is held constant. It could be that reproductive success increases with amount of
habitat on the landscape, independent of habitat fragmentation per se. If this is
true, the results of the simulation may be misleading.
These conclusions are based on the relatively small, but growing, number of
empirical studies that separate the effects of habitat loss and fragmentation per se.
So far these studies have been conducted on a limited set of taxa primarily within
EFFECTS OF HABITAT FRAGMENTATION 509
North America. More research is needed to determine how general the conclusions
are (Harrison & Bruna 1999).
ACKNOWLEDGMENTS
I thank Jeff Bowman, Julie Brennan, Dan Bert, Tormod Burkey, Neil Charbonneau,
Kathryn Freemark, Audrey Grez, Jeff Houlahan, Jochen Jaeger, Maxim Larrivée,
Lutz Tischendorf, Rebecca Tittler, Paul Zorn, and members of the Landscape Ecol-
ogy Laboratory at Carleton for comments on an earlier version of this manuscript.
This work was supported by a grant from the Natural Sciences and Engineering
Research Council of Canada.
510 FAHRIG
LITERATURE CITED
Andrén H. 1994. Effects of habitat fragmen- hierarchical study of habitat fragmentation:
tation on birds and mammals in landscapes responses at the individual, patch, and land-
with different proportions of suitable habitat: scape scale. Landsc. Ecol. 14:381–89
a review. Oikos 71:355–66 Bowman J, Cappuccino N, Fahrig L. 2002.
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
Atkinson WD, Shorrocks B. 1981. Competition Patch size and population density: the effect
on a divided and ephemeral resource: a sim- of immigration behavior. Conserv. Ecol. 6:9.
ulation model. J. Anim. Ecol. 50:461–71 http://www.consecol.org/vol6/iss1/art9
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Bascompte J, Solé RV. 1996. Habitat frag- Brennan JM, Bender DJ, Contreras TA, Fahrig
mentation and extinction thresholds in spa- L. 2002. Focal patch landscape studies for
tially explicit models. J. Anim. Ecol. 65:465– wildlife management: optimizing sampling
73 effort across scales. In Integrating Land-
Bascompte J, Possingham H, Roughgarden J. scape Ecology into Natural Resource Man-
2002. Patchy populations in stochastic envi- agement, ed. J Liu, WW Taylor, pp. 68–
ronments: critical number of patches for per- 91. Cambridge, MA: Cambridge Univ. Press.
sistence. Am. Nat. 159:128–37 480 pp.
Bélisle M, Desrochers A, Fortin M-J. 2001. In- Burkey TV. 1999. Extinction in fragmented
fluence of forest cover on the movements of habitats predicted from stochastic birth-death
forest birds: a homing experiment. Ecology processes with density dependence. J. Theor.
82:1893–904 Biol. 199:395–406
Bender DJ, Tischendorf L, Fahrig L. 2003. Caley MJ, Buckley KA, Jones GP. 2001. Sepa-
Evaluation of patch isolation metrics for pre- rating ecological effects of habitat fragmen-
dicting animal movement in binary land- tation, degradation, and loss on coral com-
scapes. Landsc. Ecol. 18:17–39 mensals. Ecology 82:3435–48
Bergin TM, Best LB, Freemark KE, Koehler Carlson A, Hartman G. 2001. Tropical forest
KJ. 2000. Effects of landscape structure on fragmentation and nest predation—an exper-
nest predation in roadsides of a midwest- imental study in an Eastern Arc montane for-
ern agroecosystem: a multiscale analysis. est, Tanzania. Biodivers. Conserv. 10:1077–
Landsc. Ecol. 15:131–43 85
Best LB, Bergin TM, Freemark KE. 2001. In- Cascante A, Quesada M, Lobo JJ, Fuchs EA.
fluence of landscape composition on bird use 2002. Effects of dry tropical forest fragmen-
of rowcrop fields. J. Wildl. Manage. 65:442– tation on the reproductive success and ge-
49 netic structure of the tree Samanea saman.
Boswell GP, Britton NF, Franks NR. 1998. Conserv. Biol. 16:137–47
Habitat fragmentation, percolation theory Chalfoun AD, Thompson FR, Ratnaswamy
and the conservation of a keystone species. MJ. 2002. Nest predators and fragmentation:
Proc. R. Soc. London Ser. B 265:1921–25 a review and meta-analysis. Conserv. Biol.
Boulinier T, Nichols JD, Hines JE, Sauer JR, 16:306–18
Flather CH, Pollock KH. 2001. Forest frag- Chesson PL. 1985. Coexistence of competitors
mentation and bird community dynamics: in- in spatially and temporally varying environ-
ference at regional scales. Ecology 82:1159– ments: a look at the combined effects of dif-
69 ferent sorts of variability. Theor. Popul. Biol.
Bowers MA, Dooley JL. 1999. A controlled, 28:263–87
EFFECTS OF HABITAT FRAGMENTATION 511
Collinge SK, Forman RTT. 1998. A conceptual and fragmentation on species extinction. J.
model of land conversion processes: predic- Wildl. Manage. 61:603–10
tions and evidence from a microlandscape Fahrig L. 1998. When does fragmentation of
experiment with grassland insects. Oikos breeding habitat affect population survival?
82:66–84 Ecol. Model. 105:273–92
Collingham YC, Huntley B. 2000. Impacts of Fahrig L. 2001. How much habitat is enough?
habitat fragmentation and patch size upon Biol. Conserv. 100:65–74
migration rates. Ecol. Appl. 10:131–44 Fahrig L. 2002. Effect of habitat fragmentation
Collins RJ, Barrett GW. 1997. Effects of habi- on the extinction threshold: a synthesis. Ecol.
tat fragmentation on meadow vole (Micro- Appl. 12:346–53
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
and ants: patch area or edge? Oikos 90:525– landscape pattern on American martens. J.
38 Appl. Ecol. 36:157–72
Goodman SM, Rakotondravony D. 2000. The Harrison S, Bruna E. 1999. Habitat fragmenta-
effects of forest fragmentation and isolation tion and large-scale conservation: What do
on insectivorous small mammals (Lipoty- we know for sure? Ecography 22:225–32
phla) on the Central High Plateau of Mada- Hartley MJ, Hunter ML. 1998. A meta-analysis
gascar. J. Zool. 250:193–200 of forest cover, edge effects, and artificial
Groppe K, Steinger T, Schmid B, Baur B, Boller nest predation rates. Conserv. Biol. 12:465–
T. 2001. Effects of habitat fragmentation on 69
choke disease (Epichloe bromicola) in the Hastings A. 1977. Spatial heterogeneity and
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
grass Bromus erectus. J. Ecol. 89:247–55 the stability of predator-prey systems. Theor.
Gurd DB, Nudds TD, Rivard DH. 2001. Conser- Popul. Biol. 12:37–48
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
vation of mammals in Eastern North Ameri- Henein K, Wegner J, Merriam G. 1998. Popu-
can wildlife reserves: How small is too small? lation effects of landscape model manipula-
Conserv. Biol. 15:1355–63 tion on two behaviourally different woodland
Gustafson EJ. 1998. Quantifying landscape small mammals. Oikos 81:168–86
spatial pattern: What is the state of the art? Hill MF, Caswell H. 1999. Habitat fragmenta-
Ecosystems 1:143–56 tion and extinction thresholds on fractal land-
Guthery FS, Green MC, Masters RE, DeMaso scapes. Ecol. Lett. 2:121–27
SJ, Wilson HM, Steubing FB. 2001. Land Hinsley SA, Bellamy PE, Newton I, Sparks TH.
cover and bobwhite abundance on Oklahoma 1995. Habitat and landscape factors influ-
farms and ranches. J. Wildl. Manage. 65:838– encing the presence of individual breeding
49 bird species in woodland fragments. J. Avian.
Hager HA. 1998. Area-sensitivity of rep- Biol. 26:94–104
tiles and amphibians: Are there indica- Hovel KA, Lipcius RN. 2001. Habitat fragmen-
tor species for habitat fragmentation? Eco- tation in a seagrass landscape: Patch size and
Science 5:139–47 complexity control blue crab survival. Ecol-
Haig AR, Matthes U, Larson DW. 2000. Ef- ogy 82:1814–29
fects of natural habitat fragmentation on the Huffaker CB. 1958. Experimental studies on
species richness, diversity, and composition predation: dispersion factors and predator-
of cliff vegetation. Can. J. Bot. 78:786–97 prey oscillations. Hilgardia 27:343–83
Haila Y. 2002. A conceptual genealogy of frag- Ims RA, Andreassen HP. 1999. Effects of exper-
mentation research: from island biogeogra- imental habitat fragmentation and connectiv-
phy to landscape ecology. Ecol. Appl. 12: ity on root vole demography. J. Anim. Ecol.
321–34 68:839–52
Haines-Young R, Chopping M. 1996. Quanti- Jansson G, Angelstam P. 1999. Threshold lev-
fying landscape structure: a review of land- els of habitat composition for the presence of
scape indices and their application to forested the long-tailed tit (Aegithalos caudatus) in a
landscapes. Prog. Phys. Geogr. 20:418–45 boreal landscape. Landsc. Ecol. 14:283–90
Hanski I, Moilanen A, Gyllenberg M. 1996. Kareiva P. 1990. Population dynamics in spa-
Minimum viable metapopulation size. Am. tially complex environments: theory and
Nat. 147:527–41 data. Philos. Trans. R. Soc. London Ser. B
Hargis CD, Bissonette JA, David JL. 1998. The 330:175–90
behavior of landscape metrics commonly Kinnunen H, Jarvelainen K, Pakkala T, Tiainen
used in the study of habitat fragmentation. J. 1996. The effect of isolation on the occur-
Landsc. Ecol. 13:167–86 rence of farmland carabids in a fragmented
Hargis CD, Bissonette JA, Turner DL. 1999. landscape. Ann. Zool. Fenn. 33:165–71
The influence of forest fragmentation and Kolozsvary MB, Swihart RK. 1999. Habitat
EFFECTS OF HABITAT FRAGMENTATION 513
fragmentation and the distribution of am- Mac Nally R, Brown GW. 2001. Reptiles and
phibians: patch and landscape correlates in habitat fragmentation in the box-ironbark
farmland. Can. J. Zool. 77:1288–99 forests of central Victoria, Australia: predic-
Komonen A, Penttilae R, Lindgren M, Hanski I. tions, compositional change and faunal nest-
2000. Forest fragmentation truncates a food edness. Oecologia 128:116–25
chain based on an old-growth forest bracket Mac Nally R, Bennett AF, Horrocks G. 2000.
fungus. Oikos 90:119–26 Forecasting the impacts of habitat fragmen-
Kremsater L, Bunnell FL. 1999. Edge effects: tation. Evaluation of species-specific predic-
theory, evidence and implications to man- tions of the impact of habitat fragmentation
agement of western North American forests. on birds in the box-ironbark forests of cen-
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
In Forest Fragmentation: Wildlife and Man- tral Victoria, Australia. Biol. Conserv. 95:7–
agement Implications, ed. JA Rochelle, LA 29
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
Morato EF. 2001. Effects of forest fragmen- Sánchez-Zapata JA, Calvo JF. 1999. Rocks and
tation on solitary wasps and bees in Cen- trees: habitat response of Tawny Owls Strix
tral Amazonia. II. Vertical stratification. Rev. aluco in semiarid landscapes. Ornis Fenn.
Brasileira Zool. 18:737–47 76:79–87
Mossman CA, Waser PM. 2001. Effects of Schmiegelow FKA, Mönkkönen M. 2002.
habitat fragmentation on population genetic Habitat loss and fragmentation in dynamic
structure in the white-footed mouse (Per- landscapes: avian perspectives from the bo-
omyscus leucopus). Can. J. Zool. 79:285–95 real forest. Ecol. Appl. 12:375–89
Niemelä J. 2001. Carabid beetles (Coleoptera: Schumaker NH. 1996. Using landscape in-
Carabidae) and habitat fragmentation: A re- dices to predict habitat connectivity. Ecology
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
coon habitat use at two spatial scales. J. Wildl. Pierotti R, Gaines MS. 2000. The interac-
Manage. 61:102–12 tion of habitat fragmentation, plant, and small
Pither J, Taylor PD. 1998. An experimental as- mammal succession in an old field. Ecol.
sessment of landscape connectivity. Oikos Monogr. 70:383–400
83:166–74 Shmida A, Ellner S. 1984. Coexistence of plant
Pope SE, Fahrig L, Merriam HG. 2000. Land- species with similar niches. Vegetatio 58:29–
scape complementation and metapopulation 55
effects on leopard frog populations. Ecology Slatkin M. 1974. Competition and regional co-
81:2498–508 existence. Ecology 55:128–34
Reddingius J, den Boer PJ. 1970. Simulation Steffan-Dewenter I, Münzenberg U, Bürger
experiments illustrating stabilization of ani- C, Thies C, Tscharntke T. 2002. Scale-
mal numbers by spreading of risk. Oecologia dependent effects of landscape context on
5:240–84 three pollinator guilds. Ecology 83:1421–
Renjifo LM. 1999. Composition changes in a 32
Subandean avifauna after long-term forest Summerville KS, Crist TO. 2001. Effects of
fragmentation. Conserv. Biol. 13:1124–39 experimental habitat fragmentation on patch
Robinson SK, Thompson FR, Donovan TM, use by butterflies and skippers (Lepidoptera).
Whitehead DR, Faaborg J. 1995. Regional Ecology 82:1360–70
forest fragmentation and the nesting success Swenson JJ, Franklin J. 2000. The effects of
of migratory birds. Science 267:1987–90 future urban development on habitat frag-
Roff DA. 1974a. Spatial heterogeneity and mentation in the Santa Monica Mountains.
the persistence of populations. Oecologia Landsc. Ecol. 15:713–30
15:245–58 Taylor PD, Merriam G. 1995. Habitat fragmen-
Roff DA. 1974b. The analysis of a population tation and parasitism of a forest damselfly.
model demonstrating the importance of dis- Landsc. Ecol. 11:181–89
persal in a heterogeneous environment. Oe- Tischendorf L. 2001. Can landscape indices
cologia 15:259–75 predict ecological processes consistently?
Rosenberg KV, Lowe JD, Dhondt AA. 1999. Landsc. Ecol. 16:235–54
Effects of forest fragmentation on breeding Tischendorf L, Fahrig L. 2000. On the usage
tanagers: a continental perspective. Conserv. and measurement of landscape connectivity.
Biol. 13:568–83 Oikos 90:7–19
Rukke BA. 2000. Effects of habitat fragmen- Tischendorf L, Bender DJ, Fahrig L. 2003.
tation: increased isolation and reduced habi- Evaluation of patch isolation metrics in mo-
tat size reduces the incidence of dead wood saic landscapes for specialist vs. generalist
fungi beetles in a fragmented forest land- dispersers. Landsc. Ecol. 18:41–50
scape. Ecography 23:492–502 Trzcinski MK, Fahrig L, Merriam G. 1999.
EFFECTS OF HABITAT FRAGMENTATION 515
Independent effects of forest cover and frag- Walters JR, Ford HA, Cooper CB. 1999. The
mentation on the distribution of forest breed- ecological basis of sensitivity of brown
ing birds. Ecol. Appl. 9:586–93 treecreepers to habitat fragmentation: a pre-
Tscharntke T, Steffan-Dewenter I, Kruess liminary assessment. Biol. Conserv. 90:13–
A, Thies C. 2002. Contribution of small 20
habitat fragments to conservation of insect Wettstein W, Schmid B. 1999. Conservation of
communities of grassland-cropland land- arthropod diversity in montane wetlands: Ef-
scapes. Ecol. Appl. 12:354–63 fect of altitude, habitat quality and habitat
Urban D, Keitt T. 2001. Landscape connectiv- fragmentation on butterflies and grasshop-
ity: a graph-theoretic perspective. Ecology pers. J. Appl. Ecol. 36:363–73
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
tion on amphibian diversity in the nature re- mentation: implications for restoration op-
serve of Ambohitantely, highland Madagas- portunities at regional scales. Landsc. Ecol.
car. Biol. Conserv. 96:31–43 14:137–45
Vance MD, Fahrig L, Flather CH. 2003. Rela- Wilcove DS, McLellan CH, Dobson AP. 1986.
tionship between minimum habitat require- Habitat fragmentation in the temperate zone.
ments and annual reproductive rates in forest In Conservation Biology, ed. ME Soulé, pp.
breeding birds. Ecology. In press 237–56. Sunderland, MA: Sinauer
van den Berg LJL, Bullock JM, Clarke RT, With KA, Crist TO. 1995. Critical thresholds
Langston RHW, Rose RJ. 2001. Territory se- in species’ responses to landscape structure.
lection by the Dartford warbler (Sylvia un- Ecology 76:2446–59
data) in Dorset, England: the role of vegeta- With KA, Gardner RH, Turner MG. 1997.
tion type, habitat fragmentation and popula- Landscape connectivity and population dis-
tion size. Biol. Conserv. 101:217–28 tributions in heterogeneous environments.
Vandermeer JH. 1973. On the regional stabi- Oikos 78:151–69
lization of locally unstable predator-prey re- With KA, King AW. 1999. Dispersal success on
lationships. J. Theor. Biol. 41:161–70 fractal landscapes: a consequence of lacunar-
Venier L, Fahrig L. 1996. Habitat availability ity thresholds. Landsc. Ecol. 14:73–82
causes the species abundance-distrubution With KA, Pavuk DM, Worchuck JL, Oates
relationship. Oikos 76:564–70 RK, Fisher JL. 2002. Threshold effects of
Villard M-A, Trzcinski MK, Merriam G. 1999. landscape structure on biological control in
Fragmentation effects on forest birds: rela- agroecosystems. Ecol. Appl. 12:52–65
tive influence of woodland cover and con- Wolff JO, Schauber EM, Edge WD. 1997. Ef-
figuration on landscape occupancy. Conserv. fects of habitat loss and fragmentation on
Biol. 13:774–83 the behavior and demography of gray-tailed
Virgós E. 2001. Role of isolation and habitat voles. Conserv. Biol. 11:945–56
quality in shaping species abundance: a test Young CH, Jarvis PJ. 2001. Measuring urban
with badgers (Meles meles L.) in a gradient habitat fragmentation: an example from the
of forest fragmentation. J. Biogeogr. 28:381– Black Country, UK. Landsc. Ecol. 16:643–
89 58
Annual Review of Ecology, Evolution, and Systematics
Volume 34, 2003
CONTENTS
EFFECTS OF INTRODUCED BEES ON NATIVE ECOSYSTEMS,
Annu. Rev. Ecol. Evol. Syst. 2003.34:487-515. Downloaded from arjournals.annualreviews.org
Dave Goulson 1
AVIAN SEXUAL DICHROMATISM IN RELATION TO PHYLOGENY AND
by Ontario Council of Universities Libraries on 05/02/08. For personal use only.
v
vi CONTENTS
INDEXES
Subject Index 691
Cumulative Index of Contributing Authors, Volumes 30–34 705
Cumulative Index of Chapter Titles, Volumes 30–34 708
ERRATA
An online log of corrections to Annual Review of Ecology,
Evolution, and Systematics chapters may be found at
http://ecolsys.annualreviews.org/errata.shtml