Download Positive interactions expand habitat use and the realized niches of

Reports
Ecology, 96(10), 2015, pp. 2575–2582
Ó 2015 by the Ecological Society of America
Positive interactions expand habitat use and the realized niches of
sympatric species
SINEAD M. CROTTY1
AND
MARK D. BERTNESS
Department of Ecology and Evolutionary Biology, Brown University, Providence, Rhode Island 02912 USA
Abstract. Niche theory, the oldest, most established community assembly model, predicts
that in sympatry, the realized niche will contract due to negative interspecific interactions, but
fails to recognize the effects of positive interactions on community assembly. The stress
gradient hypothesis predicts that positive interactions expand realized niches in stressful
habitats. We tested the predictions of the stress gradient hypothesis in a cobble beach model
system across both physical and biological stress gradients. We transplanted seven common
littoral species within, adjacent to, and below Spartina alterniflora cordgrass stands in control,
cage control, predator exclusion cage, shade, and shaded predator exclusion cage treatments
to test the hypothesis that cordgrass expands intertidal organism habitats. On cobble beaches,
cordgrass ameliorates physical and predation stresses, expanding the distribution and realized
niches of species to habitats in which they cannot live without facilitation, suggesting that
niche theory and species distribution models should be amended to accommodate the role of
positive interactions in community assembly.
Key words: cobble beach; facilitation; foundation species; Narragansett Bay, Rhode Island, USA; niche
theory; Spartina alterniflora; species distribution model; stress gradient hypothesis.
INTRODUCTION
Niche theory dictates that species distributions are
largely shaped by interspecific competition as well as by
physical and resource constraints (Grinnell 1917, Elton
1927). Textbook niche theory operationally defines the
fundamental niche as the abiotic conditions and habitats
that a species can occupy in allopatry, and the realized
niche as the conditions and habitats that a species
occupies in sympatry. In theory, the niche is defined
abstractly by habitat and resource variables as an ndimensional hypervolume (Hutchinson 1957). However,
in practice, the niche is commonly operationally defined
and typically measured correlatively and/or experimentally as habitat overlap with and without neighbors (e.g.,
MacArthur 1958, Connell 1961), the original Grinnellian concept of the niche as a species address (Trainor
and Schmitz 2014).
Recent treatments of niche theory have attempted to
reduce its dependence on competition, but still consider
the realized niche to be by definition smaller than the
fundamental niche (Rodriguez-Cabal et al. 2012).
Manuscript received 6 February 2015; revised 13 May 2015;
accepted 19 May 2015. Corresponding Editor: B. J. Grewell.
1 E-mail: [email protected]
Biogeographers and species distribution modelers have
more aggressively contested the veracity of niche theory,
questioning its value (Vannette and Fukami 2014),
broadening its definition and application (Kearney 2006,
Soberón and Nakamura 2009, Kylafis and Loreau
2011), and proposing to abandon it altogether (Chase
and Leibold 2003).
Positive interactions are particularly problematic for
niche theory (Bruno and Bertness 2001). Foundation
species, for example, provide group benefits, ameliorate
physical and biotic stresses, expand species distributions,
and increase the size of fundamental and realized niches
(Bruno et al. 2003). Positive interspecific interactions
exist on a continuum where the net interspecific
interaction may be positive or negative depending on
environmental conditions (Stachowicz 2001). The contextual nature of these interactions makes their incorporation into advanced conceptual definitions or species
distribution models (SDMs) especially challenging.
Some studies have acknowledged the role of facilitators in community assembly and have incorporated
them into definitions of the fundamental niche (Sax et
al. 2013). In the context of SDMs, this would require
that any habitat suitable for a facilitator be incorporated into the fundamental niche of its co-occurring
2575
Reports
2576
SINEAD M. CROTTY AND MARK D. BERTNESS
species, but would ignore the areas where the net
interspecific interaction is negative. SDMs generated
using this definition would overestimate the fundamental niche of the focal species. Critics argue that
distinctions between certain types of sympatric interactions are artificial and detract from the usefulness of
niche models (Araújo and Guisan 2006). Alternatively,
maintaining the historical definition of the realized
niche as the culmination of interspecific interactions
(Soberón and Nakamura 2009), along with careful
incorporation of positive interactions, may improve
these models without sacrificing simplicity.
In contrast to niche theory, the stress gradient
hypothesis (SGH; Bertness and Callaway 1994) predicts
that positive intra- and interspecific species interactions
are common in physically and biologically stressful
habitats, driven by habitat amelioration and associational defenses. The SGH has been widely tested by
plant ecologists, typically with single species along
physical stress gradients (He et al. 2013). Consumer
pressure gradients and associational defenses have
received little attention, as has the generality of the
SGH with animals, even though it was originally
proposed based on sessile animal interactions (Bertness
1989).
The SGH predicts that at physical and biological
extremes the distribution and abundance of a species will
expand in sympatry due to positive interactions (Bertness and Callaway 1994). Incorporating facilitations
into niche theory leads to the paradoxical prediction
that a species’ realized niche will be larger than its
fundamental niche in high-stress environments (Bruno
et al. 2003). Thus, incorporating the SGH into niche
theory leads to novel, testable predictions.
We conducted experimental studies on New England
cobble beaches to test the prediction that positive
interactions in stressful habitats expand realized niches.
We hypothesize that Spartina alterniflora cordgrass, the
foundation species (Dayton 1975) that builds and
maintains New England salt marshes, increases the
realized niches of marine fauna and flora by ameliorating limiting physical and biological stresses. We
operationally define realized niche space as habitat
occupied in sympatry, as is common in empirical niche
studies of extant and extinct assemblages (e.g., Benson et
al. 2014, Price et al. 2014). The realized niche is therefore
either a portion of the fundamental niche and reflects
competitive and/or consumer exclusion, is larger than
the fundamental niche and reflects habitat amelioration,
or is potentially equal to the fundamental niche and
reflects associational defenses (see Bruno et al. 2003,
Rodriguez-Cabal et al. 2012, Stachowicz 2012). We
expand the generality of the SGH to an entire intertidal
assemblage and simultaneously consider niche relationships between plants and animals along both physical
and biological stress gradients to test the relationship
between SGH and niche theory.
MATERIALS
Ecology, Vol. 96, No. 10
AND
METHODS
Species abundance
To examine the relationship between the abundance
and distribution of the numerically dominant fauna and
flora of cobble beaches and cordgrass, we quantified
species abundance in cordgrass stands (high vegetated
zone), cobble habitats at the same elevation as cordgrass
stands (high bare zone), and cobble habitats at
elevations 1 m horizontally below where cordgrass
occurs (low bare zone). We quantified cover of
organisms at 2-m intervals in the high vegetated, high
bare, and low bare zones (N ¼ 10 sites) along random
50-m transects. Points were sampled with a 50 3 50 cm
quadrat subdivided into 100 5 3 5 cm points (N ¼ 8
replicates/site). Transect data (pooled organism percent
cover/quadrat) were transformed as necessary to meet
the assumptions of parametric statistics and analyzed
among zones with ANOVA; results of this and all other
analyses were considered significant at P , 0.05.
Transplant experiment
To elucidate mechanisms driving species distributions
on cobble beaches, we ran a factorial experiment
manipulating physical stress with shades and large
predators with cages in the high vegetated, high bare,
and low bare zones at Providence Point in the
Narragansett Bay National Estuarine Research Reserve,
Rhode Island, USA. We installed 0.5 3 0.5 m plots at
.3-m intervals with five treatments: controls, predator
exclusion cages (0.5 3 0.5 3 0.5 m of 1 3 1 cm plasticcoated wire mesh), cage controls (exclusion cages
lacking two sides), shades (cage controls covered with
two layers of 0.5 3 0.5 cm plastic mesh), and shaded
cages (exclusion cages covered with two layers of 0.5 3
0.5 cm plastic mesh). Treatments were replicated eight
times per zone.
Surface photosynthetically active radiation (PAR)
was reduced by 76%, 75%, 14%, and 11% in shaded
cage, shade, cage, and cage control treatments, respectively. Cages eliminated access by pelagic predators, but
not benthic predators such as green crabs (Carcinus
maenus) or oyster drills (Urosalpinx cinerea) that
crawled under edges because of irregular cobble surface.
We did not manipulate the impact of cobble movement
disturbance, as our experiment was performed in the
summer when cobble disturbance is minimal (Bruno
2000). Instead, we experimentally examined summer
(May–August) distributions relative to thermal and
consumer stress gradients.
To quantify physical stresses, evaporative water loss,
rock temperature, and water flow were measured.
Evaporative water loss, a proxy for desiccation, was
measured with 5 3 5 3 1.5 cm sponges in each replicate
placed on 1-cm mesh hardware cloth platforms. Sponges
were saturated with fresh water, weighed, and deployed
as the midday tide was receding. Sponges were
reweighed every 30 min and total water loss was
October 2015
FACILITATION EXPANDS REALIZED NICHES
chi-square tests with Bonferroni-corrected P values for
pairwise comparisons.
RESULTS
Species abundance
Cobble beach organisms had 19 times higher percent
cover in high vegetated zones (100%) than high bare
zones (5%), and more than 2.5 times higher percent
cover than in low bare zones (39%). Cordgrass stands on
cobble beaches had 13 associated species, while low bare
and high bare habitats had seven and three associated
species, respectively.
Physical stress
Physical parameters differed markedly between vegetated and bare zones, generally revealing that cordgrass
ameliorated physical stresses on cobble beaches (Fig. 1).
During low tide exposure, the low bare zone (1-h
exposure) and the high vegetated zone (4-h exposure)
experienced less total desiccation stress (evaporative
water loss) than the high bare zone (4-h exposure; Tukey
HSD, P , 0.01). At similar elevations, cordgrass cut
evaporative water loss threefold. There were no significant (P , 0.05) treatment effects in the low bare or high
vegetated zones. In the high bare zone, controls
experienced higher desiccation stress than shaded
treatments (Tukey HSD, P , 0.01), with other
treatments experiencing intermediate levels (Fig. 1a).
Mean maximum daily temperature was lower in the high
vegetated and low bare than in the high bare zone
(Tukey HSD, P , 0.01). There were no treatment effects
in the high vegetated zone. In the low bare zone,
unshaded treatments had higher temperatures than both
shaded treatments (Tukey HSD, P , 0.01). In the high
bare zone, controls had higher temperatures than cages
and cage controls, which had higher temperatures than
shaded treatments (Tukey HSD, P , 0.01; Fig. 1b).
Chalk dissolution, a time-integrated measure of flow,
was highest in low bare, intermediate in high bare, and
lowest in the high vegetated zone (Tukey HSD, P ,
0.01). There were no treatment effects in any zone (Fig.
1c).
Transplant experiment
Common cobble beach organisms transplanted to
experimental plots performed better in cordgrass stands
than on cobble without cordgrass (see Plate 1). In
general, when transplanted to cobble at the same
elevation without cordgrass (high bare zone), heat and
desiccation mortality was common, but when transplanted to elevations below cobble beach cordgrass (low
bare zone), mortality due to predation was more
common.
More Enteromorpha survived in the high vegetated
(70% alive) than the low bare (0%; v2 ¼ 43.08, P ,
0.001) or high bare zones (5%; v2 ¼ 36.05, P , 0.001;
Fig. 2a). Fucus survival was also highest in the high
vegetated (70%) as compared with the low bare (20%; v2
Reports
calculated for each plot. Surface air temperature was
measured every 5 min for 8 weeks with data loggers
attached to cobbles with epoxy (N ¼ 2 loggers/
treatment/zone). Flow was measured by erosion of
magnesium calcite chalk blocks glued to galvanized
hardware cloth pinned to the surface in each plot.
Physical parameter data were transformed as necessary
and analyzed with a zone 3 treatment ANOVA.
To test the hypothesis that cordgrass facilitates
intertidal fauna and flora, we transplanted seven
common littoral organisms into all replicates and scored
their mortality and/or survivorship from May to August
2014. We transplanted two algal species, Enteromorpha
linza and Fucus vesiculosus, the barnacle Semibalanus
balanoides, two mussel species, Mytilus edulis and
Geukensia demissa, the herbivorous snail Littorina
littorea, and the Asian shore crab Hemigrapsus sanguineus (all species will be referred to by generic name
hereafter).
Cobbles of standardized size with high Enteromorpha
cover were collected from a common source and
transplanted to each plot. Cobbles were photographed
initially and after 48 h. Fucus was attached to large
cobbles with epoxy and transplanted to all plots. Fucus
was photographed initially and then weekly for 7 weeks.
Given the complete loss of both species in the low and
high bare zones, algal mortality was reported as a binary
variable and analyzed with chi-square tests with
Bonferroni-corrected P values for pairwise comparisons.
Semibalanus was transplanted at standardized densities into all replicates on cobbles of consistent size. Each
cobble was photographed initially and then weekly for 8
weeks to follow percentage survival. Endpoint survivorship was arcsine square root transformed to meet
parametric assumptions and analyzed with a zone 3
treatment ANOVA.
Mussels were transplanted in predator-permeable
aquaculture mesh bags of five individuals each (Mytilus,
20–40 mm; Geukensia, 40–70 mm) pinned to the surface
in each plot. After 4 weeks, they were scored for
mortality and cause of death. Mortality was observed as
shells drilled by Urosalpinx or as intact empty shells as a
result of heat and desiccation stress. Mussel mortality
among zones and experimental treatments was analyzed
using chi-square tests with Bonferroni-corrected P
values for pairwise comparisons.
Littorina and Hemigrapsus were transplanted on 5–7
cm tethers attached to substrate (N ¼ 4/plot). Juvenile
Littorina ,15 mm long were tethered in each plot with
access to shelter and left in the field for 4 weeks, after
which they were scored for mortality and cause of death
(predation by crabs or fish or by desiccation). Hemigrapsus were similarly tethered and scored for mortality
after 5 d when more than 50% mortality had occurred.
Predation mortality (carapace fragments or clipped
tethers) and desiccation mortality (intact crab carapaces
on tethers) were easily distinguished. Species mortality
counts among zones and treatments were analyzed using
2577
Reports
2578
SINEAD M. CROTTY AND MARK D. BERTNESS
Ecology, Vol. 96, No. 10
FIG. 1. Physical parameters in experimental plots; plots were control (C), control cage (CgC), predator exclusion cage (Cg),
shade (Sd), and shaded predator exclusion cage (Sd Cg). (a) Evaporative water loss (exposure duration reflects low tidal exposure),
(b) daily maximum surface rock temperature, and (c) water flow per day measured by the dissolution of chalk blocks (percentage
basis); all means are shown þ SE. Statistically significant differences (P , 0.05) were determined by post hoc testing and are
indicated by different lowercase letters above bars in zones where differences were found.
¼ 20.20, P , 0.001) and high bare zones (5%; v2 ¼ 36.05,
P , 0.001; Fig. 2b). There were no treatment effects in
any zone for either algal species.
Semibalanus survival was highest in the high vegetated
compared to the high and low bare zones (Tukey HSD,
P , 0.01; Fig. 2c). In the high vegetated zone, controls
had higher mortality than both shaded treatments
(Tukey HSD, P , 0.01), with no other treatment
effects. Low bare zone unshaded treatments had higher
mortality than shaded treatments (Tukey HSD, P ,
0.01). In the high bare zone, controls had the highest
mortality, cages and cage controls had intermediate
mortality, and shaded treatments suffered the lowest
mortality (Tukey HSD, P , 0.01).
The high vegetated zone had less Mytilus predation
mortality (0%) than the low bare zone (49%; v2 ¼ 118.24,
P , 0.001) and less desiccation mortality (23%) than the
high bare zone (52%; v2 ¼ 29.75, P , 0.001; Fig.
2d). Over 65% of Mytilus mortality in the low bare zone
was caused by Urosalpinx, as evidenced by drilled shells.
There were no treatment effects in the high vegetated or
the low bare zones, as Urosalpinx was not excluded by
cages in the low bare zone. High bare zone controls had
higher desiccation mortality than shaded treatments
(shades: v2 ¼ 16.31; shaded cages: v2 ¼ 16.98, P , 0.001)
but did not differ from cages and cage controls.
Geukensia suffered less predation mortality (0%) in
the high vegetated than in the low bare zone (13%; v2 ¼
38.90, P , 0.001) and less desiccation mortality (5%)
than the high bare zone (13%; v2 ¼ 7.25, P ¼ 0.007; Fig.
2e). Over 96% of Geukensia mortality in the low bare
zone was caused by Urosalpinx predation. There were no
significant treatment effects within the high vegetated or
low bare zones. In the high bare zone, desiccation
mortality was significantly higher in controls than in
shaded cages (v2 ¼ 11.5467, P ¼ 0.001), with no other
treatment effects.
The high vegetated zone had less Littorina predation
mortality (1%) than the low bare zone (12%; v2 ¼
15.07, P , 0.001) and less desiccation mortality (3%)
than the high bare zone (19%; v2 ¼ 20.04, P , 0.001;
Fig. 2f ). There were no treatment effects within the
high vegetated or low bare zones. High bare zone
controls had higher desiccation mortality than cage
October 2015
FACILITATION EXPANDS REALIZED NICHES
2579
Reports
FIG. 2. Control results across high vegetated (HV), high bare (HB), and low bare (LB) zones for mortality (broken down by
cause) of (a) Enteromorpha linza, (b) Fucus vesiculosus, (c) Semibalanus balanoides (shown þ SE), (d) Mytilus edulis, (e) Geukensia
demissa, and (f ) Littorina littorea. Hemigrapsus sanguineus mortality results are shown across all treatments in the (g) high
vegetated, (h) high bare, and (i) low bare zones. Statistically significant differences (P , 0.05) in mortality were determined by post
hoc testing and are indicated by different lowercase letters above bars.
controls (v2 ¼ 15.32, P , 0.001), cages (v2 ¼ 15.97, P ,
0.001), shades (v2 ¼ 21.96, P , 0.001), and shaded
cages (v2 ¼ 26.53, P , 0.001), with no other treatment
effects.
The high vegetated zone had less Hemigrapsus
predation mortality (37%) than the low bare (65%; v2
¼ 22.44, P , 0.001) and less desiccation mortality (6%)
than the high bare zone (57%; v2 ¼ 105.85, P , 0.001).
There were no treatment effects in the high vegetated
zone (Fig. 2g). In the high bare zone, shaded cages had
lower desiccation mortality than controls (v2 ¼ 12.03, P
¼ 0.001), cage controls (v2 ¼ 12.62, P , 0.001), and cages
(v2 ¼ 12.66, P , 0.001), and did not differ from shaded
treatments (Fig. 2h). There was a significant predator
exclusion effect in the low bare zone. Controls had
higher predation mortality than cages (v2 ¼ 12.57, P ,
0.001) and shaded cages (v2 ¼ 11.98, P ¼ 0.001) and did
not differ from cage controls or shades (Fig. 2i ).
DISCUSSION
Our study simultaneously examines facilitation along
physical and biotic stress gradients for an entire
community of plants and animals, rather than taking
the single species, single stress gradient approach that is
vulnerable to idiosyncratic, vague results (He and
Bertness 2014). Our results reveal that at high elevations, fauna and flora are restricted to cordgrass patches
largely due to buffering of thermal and desiccation stress
by cordgrass. Shading at high elevations without
cordgrass limited the mortality of cordgrass-dependent
species. At lower elevations, predation generally limited
organisms to cordgrass habitats protected from biotic
stress.
Reports
2580
SINEAD M. CROTTY AND MARK D. BERTNESS
FIG. 3. Conceptual model incorporating positive interactions into niche theory. In contrast to traditional niche theory,
this model predicts that the realized niche will be larger than the
fundamental niche in physically stressful habitats through
habitat amelioration, and will be identical to the fundamental
niche in biologically stressful habitats through associational
defenses. The areas enclosed by dashed lines represent the
fundamental niche space. Gray areas outlined by solid black
lines represent realized niche space along physical and
biological stress gradients.
Shaded and caged treatments rarely increased
organism survival in the high vegetated zone. Only
one of seven total species experienced enhanced
survival in shaded treatments and this effect was
smallest in the high vegetated zone. In contrast,
shading effects were pervasive in the high bare zone
where desiccation stress is highest; five of the seven
total species experienced enhanced survival in shaded
treatments. Similarly, in the low bare zone, predator
exclusion treatments enhanced organism survival when
caging effectively excluded predators. These results
highlight the mechanisms by which cordgrass facilitates
organism survival.
Our results support the SGH; in physically and
biologically stressful habitats, facilitations driven by
habitat amelioration and associational defenses increase
habitat use by dependent organisms. This supports the
hypothesis that facilitators can expand the distribution
of associated species, increasing the size of realized
niches relative to fundamental niches. This contradicts a
major assumption of niche theory and supports the
hypothesis that positive interactions expand species
distributions by ameliorating physical and biological
stresses.
Niche theory
Niche theory has been a central model of community
assembly for nearly a century, even though it is based on
the notion that interspecific competition is the dominant
driving force in spatial community assembly (Strong et
al. 1984). It has been adapted as decades of experimental
studies have improved our understanding of the roles of
Ecology, Vol. 96, No. 10
predation, disturbance, propagule supply, dispersal
limitation, and mutualism in community organization
(Morin 1999).
Despite these adaptations, modelers have commonly
identified problems with the current state of niche-based
SDMs and have proposed that the development of
simple, concrete definitions of all niche theory concepts
(Araújo and Guisan 2006) and inclusion of biotic
interactions (Wisz et al. 2013) may significantly improve
their use for conservation and management. The most
recent models have attempted to incorporate spatial
heterogeneity of trophic interactions and successfully
incorporate consumer–resource and predator–prey interactions into SDMs (Trainor and Schmitz 2014). We
propose a similar addition of spatially heterogeneous
positive interactions using our stress gradient framework
(Fig. 3).
Stress gradient hypothesis
The SGH was proposed at a time when ecologists
thought that positive interactions were rare (Connell
and Slatyer 1977, Connell 1983). It was offered as a
general pedagogical model for plants and animals,
predicting that positive interactions would increase
under stressful physical conditions due to habitat
amelioration and stressful biotic conditions due to
associational defenses. Plant ecologists were attracted
to the SGH but focused on the physical stress
predictions of the model (e.g., Brooker et al. 2008).
Little research focused on animals, whole communities,
or associational defenses, so the SGH remained poorly
tested.
Later studies elaborated on the conceptual and
practical implications of positive interactions in plant
and animal communities (Bruno and Bertness 2001) and
the contradictory predictions of the SGH and niche
theory (Bruno et al. 2003). In spite of this broadening
focus of the SGH, many plant ecologists instead focused
on exceptions to the SGH (Maestre et al. 2005,
Holmgren and Scheffer 2010). A recent meta-analysis
revealed, however, that 20 years of studies on vascular
plants found the SGH to be globally robust (He et al.
2013).
Incorporating the SGH into niche theory
Incorporating positive interactions into niche theory
is supported by our data and the inability of niche
theory to predict assembly of all but the simplest
communities. While niche theory may predict community assembly in low-stress habitats, we predict that
realized niches will be larger than or identical to
fundamental niches in high-stress habitats due to habitat
amelioration and associational defenses. We present a
simple graphical model that can be tested in other
habitats, species, and whole communities (Fig. 3). The
x-axis is the inverse relationship between physical and
biological stress documented within and among communities and is a basic assumption of other stress-based
October 2015
FACILITATION EXPANDS REALIZED NICHES
2581
community models (Menge and Sutherland 1987). At
low physical stress, high predation pressure leads to
associational defenses (e.g., Hay 1986). With high
predation pressure, competitive exclusion is limited
(Paine 1974), and associational defenses allow facilitation-dependent species to live throughout the fundamental niche space. At intermediate stresses, the
fundamental niche is larger than the realized niche,
reflecting the negative net effects of interspecific
competition and predation pressure. In contrast, at high
physical stresses, neighbor habitat amelioration in
sympatry leads to expansion of the realized niche,
potentially making it larger than the fundamental niche.
Expansion of the realized niche in relation to the
fundamental niche has been proposed in the context of
facilitation (Bruno et al. 2003), but never before in the
context of a stress gradient.
This simple didactic model reflects our experimental
results on common cobble beach organisms. When
cordgrass is not present at the higher elevations on
cobble beaches, physical stress leads to thermal and
desiccation death. Habitat amelioration by cordgrass
expands the habitat or niche space occupied by other
cordgrass-dependent species. At lower elevations, heavy
predation pressure limits species survival and forces
organisms to live at higher elevations associated with
cordgrass for predation refuge. At intermediate stress
levels, neither associational defenses nor foundation
species stress amelioration occurs, and the realized niche
is smaller than the fundamental niche as predicted by
current niche theory (Chase and Leibold 2003). Our
model is consistent with previous studies on the
dynamics of positive interactions and models of the
hierarchical organization of biogenic communities built
and maintained by foundation species (Bruno and
Bertness 2001, Altieri et al. 2010). Thus we provide a
simple new framework for incorporating facultative
mechanisms into niche theory and SDMs. As climate
changes and distributions shift, a mechanistic understanding of the components of a species niche will be
vital to protect individual species and entire ecological
communities. Current niche-based models should reflect
conceptual advancements and generate improved predictions going forward.
ACKNOWLEDGMENTS
We thank E. Suglia, H. Chen, and S. Hagerty for field
support, J. Witman for statistical advice, D. Sax, Q. He, and J.
Reports
PLATE. 1. Transplant experiment within high vegetated and low bare zones at Providence Point on Prudence Island, Rhode
Island, USA. Photo credit: M. D. Bertness.
2582
SINEAD M. CROTTY AND MARK D. BERTNESS
Bruno for manuscript comments, and K. Raposa for logistical
support. The Robert P. Brown Professorship endowment to M.
Bertness supported our work.
Reports
LITERATURE CITED
Altieri, A. H., B. K. van Wesenbeeck, M. D. Bertness, and
B. R. Silliman. 2010. Facilitation cascade drives positive
relationship between native biodiversity and invasion success.
Ecology 91:1269–75.
Araújo, M. B., and A. Guisan. 2006. Five (or so) challenges for
species distribution modeling. Journal of Biogeography 33:
1677–88.
Benson, R. B., N. E. Campione, M. T. Carrano, P. D. Mannion
C. Sullivam, P. Upchurch, and D. C. Evans. 2014. Rates of
dinosaur body mass evolution indicate 170 million years of
sustained ecological innovation on the avian stem lineage.
PLoS Biology 12(6):e1001896.
Bertness, M. D. 1989. Interspecific competition and facilitation
in a northern acorn barnacle population. Ecology 70:257–
268.
Bertness, M. D., and R. Callaway. 1994. Positive interactions in
communities. Trends in Ecology and Evolution 9:191–193.
Brooker, R. W., et al. 2008. Facilitation in plant communities:
the past, the present, and the future. Journal of Ecology 96:
18–34.
Bruno, J., J. J. Stachowitz, and M. D. Bertness. 2003. Inclusion
of facilitation into ecological theory. Trends in Ecology and
Evolution 18:119–125.
Bruno, J. F. 2000. Facilitation of cobble beach plant
communities through habitat modification by Spartina
alterniflora. Ecology 81:1179–1192.
Bruno, J. F., and M. D. Bertness. 2001. Habitat modification
and facilitation in benthic marine communities. Pages 201–
218 in M. D. Bertness, et al., editors. Marine community
ecology. Sinauer, Sunderland, Massachusetts, USA.
Chase, J. M., and M. A. Leibold. 2003. Ecological niches:
linking classical and contemporary approaches. University of
Chicago Press, Chicago, Illinois, USA.
Connell, J. H. 1961. The influence of interspecific competition
and other factors on the distribution of the barnacle
Cthamalus stellatus. Ecology 42:710–723.
Connell, J. H. 1983. On the prevalence and relative importance
of interspecific competition: evidence from field experiments.
American Naturalist 122:661–696.
Connell, J. H., and R. O. Slatyer. 1977. Mechanisms of
succession in natural communities and their role in community stability and organization. American Naturalist 111:
1119–1144.
Dayton, P. K. 1975. Experimental evaluation of ecological
dominance in a rocky intertidal algal community. Ecological
Monographs 45:137–159.
Elton, C. 1927. Animal ecology. Sidgwick and Jackson,
London, UK.
Grinnell, J. 1917. The niche relationships of the California
Thrasher. Auk 34:427–433.
Hay, M. E. 1986. Associational plant defenses and the
maintenance of species diversity: turning competitors into
accomplices. American Naturalist 128:617–641.
Ecology, Vol. 96, No. 10
He, Q., M. D. Bertness, and A. H. Altieri. 2013. Global shifts
towards positive species interactions with increasing environmental stress. Ecology Letters 16:695–706.
He, Q., and M. D. Bertness. 2014. Extreme stresses, niches and
positive species interactions along stress gradients. Ecology
95:1437–1443.
Holmgren, M., and M. Scheffer. 2010. Strong facilitation in
mild environments: the stress gradient hypothesis revisited.
Journal of Ecology 98:1269–1275.
Hutchinson, G. E. 1957. Concluding remarks. Cold Spring
Harbor Symposium on Quantitative Biology 22:415–427.
Kearney, M. 2006. Habitat, environment and niche: what are
we modeling? Oikos 115:186–191.
Kylafis, G., and M. Loreau. 2011. Niche construction in the
light of niche theory. Ecology Letters 14:82–90.
MacArthur, R. H. 1958. Population ecology of some warblers
of northeastern coniferous forests. Ecology 39:599–619.
Maestre, F. T., F. Valladares, and J. F. Reynolds. 2005. Is the
change of plant–plant interactions with abiotic stress
predictable? A meta-analysis of field results in arid environments. Journal of Ecology 93:748–57.
Menge, B. A., and J. P. Sutherland. 1987. Community
regulation: variation in disturbance, competition and predation in relation to environmental stress and recruitment.
American Naturalist 130:730–757.
Morin, P. J. 1999. Community ecology. Blackwell Science,
Malden, Massachusetts, USA.
Paine, R. T. 1974. Intertidal community structure. Experimental studies on the relationship between a dominant competitor and its principal predator. Oecologia 15:93–120.
Price, T. D., et al. 2014. Niche filling slows the diversification of
Himalayan songbirds. Nature 509:222–225.
Rodriguez-Cabal, M. A., M. N. Barrios-Garcia, and M. A.
Nuñez. 2012. Positive interactions in ecology: filling the
fundamental niche. Ideas in Ecology and Evolution 5:37–41.
Sax, D. F., R. Early, and J. Bellemare. 2013. Niche syndromes,
species extinction risks, and management under climate
change. Trends in Ecology and Evolution 28:517–523.
Soberón, J., and M. Nakamura. 2009. Niches and distributional
areas: concepts, methods, and assumptions. Proceedings of
the National Academy of Sciences USA 106:19644–19650.
Stachowicz, J. J. 2001. Mutualism, facilitation, and the
structure of ecological communities. BioScience 51:235–246.
Stachowicz, J. J. 2012. Niche expansion by positive interactions: realizing the fundamentals. A comment on RodriguezCabal et al. Ideas in Ecology and Evolution 5:42–43.
Strong, D. R., D. S. Simberloff, L. G. Abele, and A. B. Thistle,
editors. 1984. Ecological communities: conceptual issues and
the evidence. Princeton University Press, Princeton, New
Jersey, USA.
Trainor, A. M., and O. J. Schmitz. 2014. Infusing considerations of trophic dependencies into species distribution
modelling. Ecology Letters 17:1507–1517.
Vannette, R. L., and T. Fukami. 2014. Historical contingency
in species interactions: towards niche based predictions.
Ecology Letters 17:115–124.
Wisz, M. S., et al. 2013. The role of biotic interactions in
shaping distributions and realised assemblages of species:
implications for species distribution modelling. Biological
Reviews 88:15–30.