Ecology Letters, (2008) 11: 969–979 doi: 10.1111/j.1461-0248.2008.01210.

REVIEW AND
SYNTHESIS Toward a better understanding of the regional causes
of local community richness

Abstract
Susan Harrison* and Howard Despite widespread acknowledgement that local ecological communities are profoundly
Cornell shaped by regional-scale influences, including evolutionary and biogeographic processes,
Department of Environmental this perspective has yet to be widely incorporated into ecological research. Drawing on
Science and Policy, University of recent research, we propose four steps towards making regional influences a stronger
California, Davis, CA 95616, USA
part of research on the richness of local communities: (1) identifying the regional-scale
*Correspondence: E-mail:
causes of variation in species richness in the systems ecologists study; (2) testing for
[email protected]
effects of regional richness on local richness, using improved observational and
experimental analyses to overcome earlier problems; (3) simultaneously analysing
environmental influences on regional and local species richness as well as the influence
of regional richness on local richness and (4) considering the potential reciprocal effects
of local processes on regional richness. In conclusion, we suggest some ways that similar
approaches could be applied to other aspects of community structure beyond species
richness.

Keywords
Biogeography, history, large-scale ecology, local richness, regional richness, species
diversity, species richness.

Ecology Letters (2008) 11: 969–979

communities to immigration from a regional pool (Stohl-

INTRODUCTION
gren et al. 2008), while the advent of community phyloge-
In the past several decades, ecologists have become aware netics has provided a powerful new tool for studying
that species assemblages are governed not only by local regional and evolutionary influences upon local community
interactions among coexisting species, but also by large-scale patterns (Brooks & McLennan 2002; Webb et al. 2002).
biogeographic, historical and evolutionary processes (Rick- Many ecologists are critical, nonetheless, of studies of
lefs & Schluter 1993; Ricklefs 2004, 2007). One particular regional influences on local community richness (e.g.
idea that arose from this large-scale outlook, namely that Hillebrand & Blenckner 2002; Hillebrand 2005; Shurin &
regional propagule supply is an important reason for Srivastava 2005). One issue is conceptual; some find it hard
variation in the species richness of natural communities, to accept the idea of a regional pool of species that governs
was first given prominence by MacArthur & WilsonÕs (1967) the richness of local communities within the region and that
Theory of Island Biogeography. This idea gained further impetus is relatively little affected by local ecological processes.
from many studies showing that the number of species Other concerns are more technical; for example, local and
within small localities increased either linearly or asymptot- regional richness analyses have been criticized for statistical
ically with the number of species in the larger surrounding artefacts and problems of interpretation. Even among those
regions (local–regional richness or local and regional who enthusiastically accept the general idea of regional
richness analyses; e.g. Cornell 1985; Ricklefs 1987; Cornell influences on communities, there is still a widespread
& Lawton 1992; Shurin & Srivastava 2005). The observation tendency to focus on the local scale where powerful tools
that native and exotic species richness are nearly always such as manipulative experiments are available, whereas it
positively correlated at moderate-to-large spatial extents has has been less clear how to make progress on understanding
helped reinforce the concept of the openness of local regional influences.

 2008 Blackwell Publishing Ltd/CNRS

970 S. Harrison and H. Cornell Review and Synthesis

Here we draw upon recent research to try to provide a including rates of diversification, niche conservatism and
series of suggestions for improving our understanding of differential dispersal (for a more detailed overview, see
regional influences on local communities. The proposed Harrison & Cornell 2007).
steps include: first, advancing our understanding of the To test whether regions differ in species richness because
regional-scale causes of variation in species richness; second, of different rates of diversification, i.e. speciation minus
using improved techniques to test the effects of regional extinction, the simplest approach is a sister clade compar-
richness on local richness; third, integrating external ison, in which one asks (for example) whether clades at
environmental influences on local and regional richness lower latitudes are consistently larger than their sister taxa at
into analyses of the local and regional richness relationship higher latitudes. A more complex approach is to analyse
and, fourth, considering the reciprocal effects of local lineage-through-time plots to compare rates of diversifica-
processes upon regional richness. In conclusion, we tion among clades. Using these and related phylogeny-based
consider how to extend similar regionally oriented methods, some recent studies have found evidence that
approaches to studying other aspects of community higher diversification rates can explain the greater species
structure beyond species richness. richness of regions at lower latitudes, as well as differences
among other types of regions and among taxa within the
same region (e.g. Mittelbach et al. 2007; Ricklefs 2007;
UNDERSTANDING LARGE-SCALE CAUSES OF
Wiens 2007).
VARIATION IN SPECIES RICHNESS
Alternatively, some regions may have higher species
Under a strictly ecological worldview, in which local richness due to niche conservatism (Wiens & Donoghue
interactions and local abiotic conditions are of prime 2004), i.e. intrinsic and extrinsic limitations on the ability to
importance, the properties of regions – such as their species adapt and disperse, which tend to confine some taxa to the
richness, or the relationship of productivity to richness geographic or climatic zones in which they originated. As
across multiple regions – can seemingly only be built up applied to the latitudinal richness gradient or the produc-
from the properties of the localities they contain. How is it tivity–richness relationship, for example, the niche conser-
possible, then, to define a Ôregional species poolÕ or a vatism hypothesis suggests that there are more species in
Ôregional influenceÕ on local richness or other community warm and mesic regions because the majority of modern
patterns? One answer is that a region has some attributes higher taxa originated in warm and mesic environments, and
that are largely independent of its localities, such as its total many of these taxa cannot adapt fully to cooler or drier
area, internal environmental heterogeneity and biogeo- conditions (Wiens & Donoghue 2004; Harrison & Grace
graphic history (as, for example, an old region can contain 2007; Hawkins et al. 2007; Wiens 2007).
young localities). It is possible for such attributes to affect A third class of explanations for variation in regional
regional species richness, and, through dispersal, for regional richness is the differential dispersal of taxa among geo-
richness to influence local richness. In one early example, graphic regions. Dispersal among regions takes place at time
Cornell (1985) found that the geographic range sizes of oaks scales intermediate between those of speciation and
predicted the regional (whole oak species) richness of their extinction on one hand and competition and predation on
gall wasp faunas, and that regional richness in turn predicted the other, and thus may form a critical link between the
local (per-tree) richness of gall wasp communities on evolutionary and ecological determinants of community
different oak species. The effect of regional area upon local structure (Ricklefs 2004). While it is a difficult class of
richness has recently been termed the Ôecho patternÕ of processes to study, Roy & Goldberg (2007) have devised a
large-scale species–area relationships (Rosenzweig & Ziv model-fitting approach for testing the role of differential
1999). More recently, Partel (2002) showed that the dispersal in explaining regional patterns in the richness and
relationship of soil pH to local plant richness was positive age distributions of taxa, and Moore & Donoghue (2007)
in large geographic regions within which soil pH was have recently developed a rigorous method for identifying
generally alkaline, but negative in regions within which pH key dispersal events in lineages.
was generally acidic, and attributed this difference to the A few studies have attempted to use evolutionary
larger pool of species adapted to the prevailing pH in each methods to explain patterns in local as well as regional
region. richness. For example, McPeek and colleagues (e.g. McPeek
The recent rapid growth in phylogeny-based techniques, & Brown 2000) concluded that there were more species of
combined with the increasing interest of ecologists in Enallagma damselflies in lakes with than without fish because
evolutionary influences, has offered hope for a more explicit of the particular history of diversification and post-glacial
understanding of the causes of regional-scale species colonization of this group in the eastern USA, which has
richness. We briefly review some major evolutionary and produced a regional species pool with fewer dragonfly-
historical explanations for variation in regional richness, adapted than fish-adapted Enallagma. Dragonflies are top

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Regional causes of richness 971

predators in fishless lakes, and only some lineages within interest (Cornell & Lawton 1992; Srivastava 1999). Curvi-
Enallagma have evolved the behaviours necessary to resist linearity may also occur if the locality is defined at the
dragonfly predation; Other phylogenetic analyses of local excessively small scale of a few individuals, where the
assembly patterns within clades are reviewed by Brooks & numbers of individuals set an upper limit on richness
McLennan (2002). regardless of increases in regional richness (Loreau 2000). At
Narrow taxonomic scope is a major limitation on these the other extreme, linear relationships at very large local
approaches to understanding regional species richness. In scales may occur because large localities subsume multiple
general, it has not yet been possible to reconcile the rigor of habitats, substantial beta diversity and therefore a large
phylogenetic evidence with the taxonomic breadth of proportion of regional diversity, producing autocorrelation
complete ecological communities; in other words, all at the two scales (Huston 1999; Loreau 2000; Hillebrand &
potentially interacting species as opposed to only the Blenckner 2002).
members of a single clade. Clearly, ecologists will continue Underestimation can be corrected by the use of indices
to need a broad array of approaches to understanding such as FisherÕs alpha or the Chao estimator (Gotelli &
regional influences, including correlative studies and Colwell 2001), which use the frequency of rare species to
descriptive biogeography as well as phylogenetic analyses estimate true local richness. Overestimates of regional
(e.g. Qian et al. 2007). richness can be avoided if accurate natural history informa-
tion makes it possible to include in the species pool only
those species that are able to occupy the habitat of interest
IMPROVING ANALYSES OF THE LOCAL AND
(e.g. Zobel et al. 1998). Autocorrelation can be avoided
REGIONAL RICHNESS RELATIONSHIP
when truly independent datasets are available to measure
The majority of attempts to understand regional contribu- local and regional richness (Srivastava 1999), and ⁄ or by
tions to local richness in natural communities have used choosing scales for sampling local richness that are small
regressions of local on regional richness, or local and enough to minimize internal environmental heterogeneity.
regional richness analyses, although a minority have used Random placement null models are another useful way to
manipulative experiments. Positive local and regional deal with the autocorrelation and underestimation problems
richness relationships have been found widely (Cornell & (Belmaker et al. 2008). One way to avoid arbitrary defini-
Lawton 1992; Shurin & Srivastava 2005), including in tions of local and regional scales is to examine the
hyperdiverse communities (Karlson et al. 2004; Witman et al. consistency of the relationship across different spatial scales
2004) and across a broad range of spatial scales (Hillebrand (Cornell & Karlson 1997; Hillebrand & Blenckner 2002;
& Blenckner 2002; Shurin & Srivastava 2005; Freestone & Shurin & Srivastava 2005). Such an approach increases the
Harrison 2006; Cornell et al. 2008). It was originally argued power to detect curvilinear patterns, which are often missed
that positive linear local and regional richness relationships when the local scale is too large relative to the regional scale
implied regional control over local richness, whereas (Shurin & Srivastava 2005).
asymptotic relationships implied that local richness was These solutions to sampling bias were employed in a
constrained by local ecological interactions (Ricklefs 1987). field study of coral species assemblages along a regional
However, local and regional richness analyses have been biodiversity gradient in the west-central Pacific Ocean
much criticized, in part because of sampling and statistical (Karlson et al. 2004; Cornell et al. 2008). Data were
artefacts, and in part because of theoretical problems with collected on species-relative abundances, and separate
their interpretation. We consider each of the most critical regional richness estimates for flat, crest and slope habitats
problems and discuss possible solutions and limitations, and were used. Five regions were sampled (large, widely spaced
then discuss the advantages and disadvantages of experi- island groups) along the gradient, at three local scales,
mental approaches. comprising 10 m transects 100–2 m apart, sites 103–4 m
One statistical issue is pseudosaturation, or the tendency apart on 15 islands and islands 104–6 m apart within each
for curvilinear local and regional richness relationships to region. Slopes of the regressions of log-LSR on log-RSR
arise if local richness is underestimated or the regional pool showed a weak tendency to decrease with decreasing scale
is overestimated. Underestimation of local richness results using the raw data, suggesting a slight but non-significant
from disproportionate under-sampling of rare species in tendency towards curvilinearity at the smallest (single
species-rich regions (Caley & Schluter 1997). Overestimates transect) spatial scale (Fig. 1). However, when the Chao-1
of the regional pool occur because local richness is usually estimator was substituted for raw richness, the trend
measured in a single habitat, whereas regional richness is disappeared, suggesting that it may have been caused by
often taken from taxonomic compendia that combine all undersampling; all slopes were linear, suggesting local
habitats within a biogeographic region. The supposed pool communities in this case are open to enrichment from the
thus includes species that do not occur in the habitat of regional species pool.

 2008 Blackwell Publishing Ltd/CNRS

972 S. Harrison and H. Cornell Review and Synthesis

tions. Observational local and regional richness analyses

should ideally be complemented by experiments, as we
discuss below (e.g. Fox et al. 2000; Shurin et al. 2000).
Theoretical questions have also been raised about the
interpretation of local and regional richness relationships
from a metacommunity perspective, in which local and
regional richness may have reciprocal influences; we will
consider these in Local influences on regional richness: the
metacommunity perspective section.
Manipulative experiments addressing the effects of
regional processes on local richness have generally involved
altering the size of the regional species pool, such as by adding
seeds from nearby sites into experimental plots and deter-
mining the effect on the richness and composition of the local
community (Turnbull et al. 2000). In a now-classic experi-
ment, Tilman (1997) added seeds of 0–54 native species to
plots containing varying numbers of resident species. The
effects of the additions on the local community were
dramatic. The number of species that became established
continued to increase with the number of species added,
indicating strong effects of the regional pool on local
richness. Many of the newly established species have
persisted in the plots for >10 years (D. Tilman, pers. comm.).
Nevertheless, richer plots were more resistant to establish-
ment, suggesting that local processes also played a role in
limiting the richness of the community. A subsequent study
demonstrated that species in the same functional group as the
residents were less likely to become established, probably
through competitive inhibition (Fargione et al. 2003).
Even stronger resistance to establishment was observed
in an experimental manipulation of the regional pool of
zooplankton species occupying fishless ponds in northern
Michigan (Shurin et al. 2000). When an average of 12.9
species were experimentally added to the ponds, some
species were able to colonize successfully, but over 91%
Figure 1 Log-local vs. log-regional species richness of corals at went extinct and there were no observable effects of the
three spatial scales along a regional biodiversity gradient in the additions on diversity in the ponds. As in the Tilman
west-central Pacific Ocean that includes the worldÕs most diverse (1997) experiment, successful establishment correlated
coral assemblages. The local scales comprised transects 100–2 m negatively with the diversity of resident species, suggesting
apart, sites 103–4 m apart and islands 104–6 m apart. Local richness that species interactions played an important role in
was averaged over all samples at each scale, and separate preventing establishment, despite the strong linear rela-
calculations of local and regional richness were made for each tionships between local and regional richness in the
habitat. All habitats fall on the same regression line, and none of zooplankton system. However, when resident species
the regression slopes differs significantly from 1, indicating linearity
biomass was experimentally reduced, successful establish-
of the relationship at all scales (Cornell et al. 2008).
ment by the added species increased dramatically (Shurin
et al. 2000).
We conclude that it is possible to overcome the statistical In an aquatic microcosm experiment, Fox et al. 2000
obstacles to testing for significant effects of regional on found a saturating relationship between regional (initial) and
local richness. We suggest that positive local and regional local (final) species richness. They attributed this relation-
richness relationships provide some evidence for the ship to the fact that the richest regional pools, but not the
openness of local communities to regional enrichment, less-rich pools, apparently included some species that were
although the degree of linearity or nonlinearity by itself may highly vulnerable to exclusion via local species interactions.
not be enough to determine the strength of local interac- This result highlights the importance of ensuring that

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Regional causes of richness 973

regional species richness is not confounded with regional

species composition. Regional Local
Regional species pools of moss-inhabiting microarthro- climate climate
pods were altered by placing moss fragments from different
sites around a defaunated local moss patch and observing
the process of colonization for 16 months (Starzomski et al.
2008). Regional richness had no effects on local richness, a Regional Local
pattern that the authors attribute to pseudosaturation. richness richness
However, the composition of the regional pool did affect
the richness and composition of the local community, as did
assembly time and seasonality. Other experimental studies
confirm the effects of the composition of the regional pool Geologic
(Fox et al. 2000; Fukami 2004a,b). history Disturbance

The experiments described above generally support the

joint importance of local processes and regional pools in
limiting local richness, and show that empirically, strong Figure 2 Conceptual framework for a multiscale structural equa-
interactions and openness to regional enrichment are not tion model of the correlates of species richness (Harrison et al.
mutually exclusive. One resolution to this apparent paradox 2006).
is provided by TilmanÕs (2004) stochastic niche theory,
which proposes that propagules can establish successfully in local and regional scales, rather than reflecting cause and
a saturated community as long as their arrival coincides with effect?
an appropriately timed disturbance. Another possibility is One possible approach to such problems is to include
that local richness may be maintained not only by measures of the environment in statistical models of the
competitive relations but by a continuous influx of local and regional richness relationship, using structural
propagules from other parts of a spatially heterogeneous equation modelling (SEM; Grace 2006). In such a model
region, as in mass-effects metacommunity models (Leibold (Fig. 2), some environmental factors may affect local
et al. 2004). Some experimental evidence from herbaceous richness directly, others may act directly on regional richness
plant communities supports this possibility (Mouquet et al. and affect local richness only indirectly, and still others may
2004). have direct (and possibly different) effects on richness at
both scales. This approach has the potential to shed clearer
light on the scales at which the environment affects
INTEGRATING ENVIRONMENTAL INFLUENCES ON
richness, and thereby to aid in the search for plausible
REGIONAL AND LOCAL SPECIES RICHNESS
mechanisms.
While some influences on species richness appear inherently In a study of plants on serpentine soils in California
regional in scope, such as regional area, heterogeneity, age (Harrison et al. 2006), the conceptual model assumed that
and evolutionary history, and other influences appear regional (80–5000 km2) species richness positively affected
obviously local in their action, such as competition and local (1000 m2) species richness. Climate (e.g. rainfall,
other biotic interactions, other aspects of the environment temperature, productivity) and spatial structure (area and
present more of a conceptual and methodological problem. isolation of serpentine patches) were hypothesized to affect
Gradients in temperature, precipitation, productivity and both regional and local richness directly, although poten-
latitude, for example, have all been shown to correlate with tially in different ways; these variables were measured at
species richness at both local and regional scales, although both regional scale and local scales. Historical variables (e.g.
the form and strength of these relationships may often differ age of exposure of serpentine within a region) were
across scales (e.g. Mittelbach et al. 2001, Chase & Leibold expected to act directly only on regional richness, while
2003; Hillebrand 2004). When regional and local richness conditions in the study plots (e.g. soil fertility, animal
are correlated with one another, but both are also influenced disturbance) were expected to act directly only on local
by abiotic environmental gradients, what can we conclude richness. SEM was used to evaluate the significance and
about regional and other influences on local richness? Are strength of the direct and indirect paths in this model. The
the environmental gradients in question best interpreted as strongest predictor in the model was the remotely sensed
acting locally, through ecological interactions, or regionally, measure of productivity known as NDVI (normalized
through evolutionary or biogeographic mechanisms? Could difference vegetation index), which, when measured at the
an observed local and regional richness relationship be an regional scale, explained 50% of the variation in regional
artefact of environmental conditions that are shared at the richness. Whether measured regionally or locally, however,

 2008 Blackwell Publishing Ltd/CNRS

974 S. Harrison and H. Cornell Review and Synthesis

et al. 2001), and other ecological phenomena. We suggest

Mean + Regional + Woody that it may be possible to test such hypotheses using
annual productivity
rainfall and litter
(NDVI) statistical models similar to the one described above, in
+ cover
+/–
which the influence of regional richness on local richness is
+
a core structural assumption. Such models could also
Regional Local
richness richness
address concerns that have been raised about circularity in
local and regional richness analyses; in a model with multiple
environmental predictors, SEM could be used to test
Other Other whether regional richness causes (can uniquely explain
regional local
factors factors variation in) local richness, whether local richness causes
(can uniquely explain variation in) regional richness, or
whether the local and regional richness relationship disap-
Figure 3 Results of a multiscale structural equation model of the
correlates of species richness (Harrison et al. 2006). Width of
pears once the shared responses of local and regional
arrows indicates the relative strengths of pathways. A Ô+Õ indicates richness to environmental influences are taken into account.
positive effect and a Ô+ ⁄ )Õ indicates a unimodal effect. So far, there have been few studies of the kind suggested in
this section, although Grace et al. (2007) used a similar
approach to investigate whether local species richness drives
productivity exerted only indirect and weak effects on local local community biomass or vice versa. The major limitation is
richness. Regional richness significantly predicted both total the requirement for data on regional and local environments
and residual variation in local richness (Fig. 3). as well as regional and local richness. Also, while SEM is the
Productivity is among the most widely studied correlates only statistical technique that allows any variable – in this case,
of local species richness (e.g. Grace 1999; Mittelbach et al. regional richness – to both act on some variables and be acted
2001), although cross-study evidence suggests that its on by others, it has been criticized for its inability to test for
relationship to richness is stronger and more consistently complex nonlinear relationships, and Wootton (1994) has
positive at large regional scales (Hawkins et al. 2003; Currie argued persuasively that the predictions of SEM models
et al. 2004), as is the latitudinal diversity gradient (Hillebrand should be tested by experiments.
2004). The study described above is the first to show that
productivity acts directly on regional richness and affects
LOCAL INFLUENCES ON REGIONAL RICHNESS:
local richness only indirectly. This may cast some doubt
THE METACOMMUNITY PERSPECTIVE
upon ecological mechanisms for the productivity–richness
relationship, such as that greater productivity enhances the In large part due to the influential work of Ricklefs and
ability of competing species to coexist locally, and instead colleagues (Ricklefs 1987, 2004, 2007; Ricklefs & Schluter
suggests the relationship may reflect history or evolution. 1993), ecologists have tended to equate embracing the
Consistent with this, later analyses showed that the positive regional scale with acknowledging the importance of
productivity–richness pattern is driven by the dominance of evolution and biogeography in ecology. Yet it is inescapably
the regional species pool by species with evolutionary true that regions are made up of localities, and that local
affinities to high-productivity conditions (Harrison & Grace ecological processes and dispersal among localities must
2007). Similarly, Hawkins et al. (2007) found the regional reciprocally contribute to regional patterns. Moreover, just
productivity–richness relationship was stronger in bird as ecologists have tended to focus on very small spatial
lineages of tropical than temperate derivation, and Partel scales, evolutionists interested in questions of species
et al. (2007) found the classic unimodal local productivity– richness have tended to focus on single clades at very
richness relationship prevailed only in the temperate zone, broad, often continental or global geographic scales.
where fewer species are adapted to high-productivity Perhaps the greatest challenge in the study of regional
conditions than in the tropics. These studies suggest that influences on local communities is the need for more
productivity may be a regional-scale influence on richness, analyses at the scale of regions much smaller than continents
and that evolutionary niche conservatism may be a strongly yet much larger than collections of field plots – scales of
contributing mechanism. perhaps 101–106 km2. This has been suggested as roughly
Hypotheses based upon the existence of a larger regional the scale at which the species pools are assembled that are
species pool that influences local-scale richness have among the most significant drivers of local community
previously been proposed to explain the relationships composition (Ricklefs 2004). Within such regions, species
between diversity and productivity (Grace 1999; Huston richness is likely to be strongly affected by meso-scale
1999), diversity and pH (Partel 2002), diversity and influences such as dispersal, local extinction and spatial
invasibility (Wardle 2001), diversity and stability (Loreau heterogeneity, which form a link between the very large-

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Regional causes of richness 975

scale processes of speciation and (global) extinction on the may arise even in the presence of strong competition,
one hand, and the local-scale processes of competition and depending on the relative rates of dispersal (Caswell &
predation on the other. Cohen 1993; Mouquet & Loreau 2003; Fukami 2004a),
Metacommunity theory, in which regions are portrayed disturbance (Caswell & Cohen 1993; Mouquet et al. 2003;
as collections of local communities linked by dispersal, Tilman 2004; Hillebrand 2005) and ⁄ or predation (Shurin &
provide an important framework for considering the Allen 2001). It has also been suggested that curvilinear
potential reciprocal interactions between local and regional relationships can arise even without competition (He et al.
richness and other aspects of community structure. One of 2005; Fox & Srivastava 2006). However, by providing the
the fundamental questions asked by metacommunity basis for improved null models, metacommunity theory may
models, for example, is under what conditions locally be useful in refining local and regional richness analyses just
incompatible competitors can coexist at the regional scale. as it has been in identifying their weaknesses (e.g. Mouquet
In a synthesis, Leibold et al. (2004) identified four types of & Loreau 2003; Cadotte 2006; Hugueny et al. 2007).
metacommunity models according to the roles played by For example, Hugueny et al. (2007) adapted a simple
dispersal and environmental heterogeneity: species-sorting, metacommunity model by Hastings (1987) to create a null
mass-effect, neutral and patch dynamics models. These model for the system of three Daphnia species in rockpools
models generally predict that the regional coexistence of on Baltic Sea islands, where colonization and extinction had
strong competitors is possible at low-to-intermediate been measured in hundreds of rockpools for multiple years
dispersal rates, whereas when dispersal is sufficiently high, (Bengtsson 1989; Pajunen & Pajunen 2003). The observed
local competitive exclusion may propagate to the regional relationship of local (rockpool) to regional (island) Daphnia
scale and reduce regional richness (e.g. Caswell & Cohen species richness was asymptotic. The model by Hugueny
1993; Mouquet & Loreau 2003). The species-sorting and et al. (2007) assumed that all Daphnia and all rockpools were
mass-effects metacommunity models provide an additional identical, Daphnia were either present or absent in rock-
reason why regional richness may be relatively resistant to pools, and competition among the three species would cause
local processes; these models emphasize regional environ- a linear increase in local extinction with the number of
mental heterogeneity and niche differences, which expand species per pool. Colonization of pools by each species was
the possibilities for locally incompatible species to coexist modelled in two ways: as a function of the frequency of
stably at the regional scale (e.g. Mouquet & Loreau 2003). occupied pools, or as a constant rate, reflecting uncertainty
When competition or other interactions are strong, rates about whether this system shows Levins-like (all popula-
of dispersal among localities are high and spatial refuges are tions small and transient) or mainland-island (some popu-
absent, metacommunity theory predicts that local dynamics lations large and permanent) structure. The model was
can feed back to reduce regional (metacommunity) richness. parameterized and tested with independent datasets. The
Empirical studies have demonstrated the interacting effects empirical local and regional richness relationship could only
of local community dynamics, among-community dispersal be fit by the model with competition and a Levins-like
and regional species composition in aquatic microcosms and metacommunity structure (Fig. 4); this model also fits the
mesocosms; when connectivity among patches is manipu- observed average local species richness per island better
lated, the results generally support the predictions that local than either a no-competition model or a mainland–island
and regional richness are maximized at low-to-intermediate model (Hugueny et al. 2007).
dispersal (e.g. Holyoak & Lawler 1996; Forbes & Chase Extending metacommunity ideas to large-scale natural
2002; Cadotte 2006). In systems with unmanipulated spatial systems remains a major empirical challenge. The studies by
structure, such as zooplankton in natural ponds or inquilines Shurin & Allen (2001) and Hugueny et al. (2007) illustrate
in pitcher plants, it has usually been easier to test the local- the utility of making predictions about local and regional
scale than regional-scale predictions of metacommunity richness based on detailed metacommunity models, as a
models (e.g. Shurin & Allen 2001; Cottenie et al. 2003; Miller basis for improved inferences about underlying processes.
& Kneitel 2005). However, Shurin & Allen (2001) showed However, it must be acknowledged that relevant parameters
that patterns of predator and prey zooplankton richness such as the rates of extinction and colonization of localities
across multiple natural ponds met the predictions of a will be very difficult to measure in most systems, especially
system-specific metacommunity model. when species have cryptic life-history stages.
Metacommunity models have led to an important Finally, we reiterate that in the absence of the detailed
conceptual critique of local and regional richness analyses, information required to test metacommunity theory, the
by showing that it is possible for either linear or curvilinear question of whether regional richness is strongly affected
local and regional richness relationships to arise under a by local processes could be treated as a statistical problem.
variety of assumptions (Shurin & Srivastava 2005). Most One possible approach is to test whether local richness
significantly, linear local and regional richness relationships predicts residual variation in regional richness or vice versa,

 2008 Blackwell Publishing Ltd/CNRS

976 S. Harrison and H. Cornell Review and Synthesis

Noncompetitive
their species composition vary significantly among discrete
regions, suggesting species pool effects; third, how do
Mainland–island
1.0 environmental influences relate to the species composition
and impacts of predators at both regional and local scales
Levins
Local richness

and, fourth, are local community dynamics plausible

explanations for the regional presence or absence of
0.5 particular predators? In keeping with a very large-scale
view, Losos (1996) concluded that the strong differences in
predator assemblages between North America, East Africa
and Southeast Asia resulted from the intercontinental
0.0
biogeographic history of the Carnivora. But explanations
0 1 2 3 4
based upon the environment, local dynamics or dispersal
Regional richness among localities are possible in other cases, particularly at
smaller scales.
Figure 4 Observed (points) and predicted (lines) relationships Regional influences on some other aspects of species
between local and regional species richness for the rock pool composition have been suggested in several of the examples
Daphnia system. Predictions are based on the Levins-like and we have described. For example, the changing relative
mainland–island metacommunity models with competition. For abundances of species of different functional or biogeo-
comparison, the predicted relationship without competition (iden- graphic groups within local communities, across gradients of
tical in both Levins-like and mainland–island models) is shown.
latitude, productivity or soil pH, may be the result of
Vertical bars represent the 95% confidence region around the
predictions of the Levins-like model (see Hugueny et al. 2007 for
variation in the composition of regional species pools (e.g.
details). Grace 1999; Harrison & Grace 2007; Partel et al. 2007).
Metacommunity dynamics have been shown to influence
the regional presence or absence of dominant predators or
as outlined in the preceding section. Another approach competitors, with substantial effects on the structure as well
has been to analyse the correlation between native and as the richness of local communities. For example, a strong
exotic species richness at multiple spatial scales; for competitor led to increased community similarity in well-
example, Stohlgren et al. (2008) used a comprehensive connected microcosm metacommunities (e.g. Forbes &
analysis of plant invasions to argue that the addition of Chase 2002), and the local abundances of competing pitcher
new species to communities seldom if ever results in the plant inquilines were affected by manipulating levels of
regional-scale extinction of resident species. Based on dispersal (Miller & Kneitel 2005).
metacommunity theory, plausible explanations for his One of the greatest recent advances in understanding
results include weak interactions, spatial heterogeneity large-scale influences on local community structure has
and ⁄ or low-to-moderate connectivity among communities come from the new field of community phylogenetics,
within regions. which compares the degree of taxonomic relatedness among
members of a local community to their expected relatedness
based on a regional species pool (e.g. Webb et al. 2002).
CONCLUSIONS: BEYOND SPECIES RICHNESS?
Although the focus is on phylogenetic species composition,
In light of the well-recognized need for ecologists to adopt a there are parallels to studies of local and regional richness:
larger spatial and temporal perspective, we have tried to first, because both regional and local community data are
articulate some concrete ways in which progress can be required, and second, because historical vs. ecological
made in understanding the influence of processes at the explanations are being contrasted. The usual expectations
regional scale. We have tried to identify areas of confusion are that coexisting species will be less related than expected
and limitations of methodology or data, as well as to report based on the regional pool (i.e. overdispersed) if competi-
new and emerging insights. We have restricted our tion predominates, while they will be more related than
discussion to species richness, as it is perhaps the most expected based on the regional pool (i.e. underdispersed) if
accessible aspect of community structure, but we think our regional species pools constitute a dominant limitation on
suggested approaches can be extended to other aspects as local community membership. Also similarly to analyses of
well. For example, consider the question of why the top– richness, the scale at which regions and localities are defined
down influences of predators are greater in some commu- may influence the results (Cavender-Bares et al. 2006;
nities than others. Paralleling the approach we have taken, Swenson et al. 2006). As the regional scale increases, local
one could ask, first, what is the regional biogeographic communities tend to switch from over- to underdispersion,
history of the predator lineages in question; second, does and it has been suggested that this switch can be used to

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Regional causes of richness 977

indicate the scale at which regional influences become Caswell, H. & Cohen, J.E. (1993). Local and regional regulation of
dominant controls over local community membership species-area relations: a patch-occupancy model. In: Species
(Swenson et al. 2006). However, the interpretation of Diversity in Ecological Communities: Historical and Geographical Per-
spectives (eds Ricklefs, R.E. & Schluter, D.). University of Chicago
scale-dependence in community phylogenetic structure is
Press, Chicago, IL, pp. 99–107.
still in its early stages (Cavender-Bares et al. 2006). Cavender-Bares, J., Keen, A. & Miles, B. (2006). Phylogenetic
Hardy & Senterre (2007) have recently proposed an structure of Floridian plant communities depends on taxonomic
analytical framework in which the phylogenetic structure of and spatial scale. Ecology, 87, S109–S122.
a community can be partitioned into within-locality (alpha) Chase, J.M. & Leibold, M.A. (2003). Spatial scale dictates the
and among-locality (beta) components, exactly analogous to productivity-biodiversity relationship. Nature, 416, 427–430.
the traditional multiscale partitioning of species richness. Cornell, H.V. (1985). Local and regional richness of cynipine gall
wasps on California oaks. Ecology, 66, 1247–1260.
This new approach offers a potentially powerful way to ask
Cornell, H.V. & Karlson, R.H. (1997). Local and regional processes
whether there are parallel patterns in these two aspects of as controls of species richness. In: Spatial Ecology: The Role of Space
the local and regional geographic structure of communities. in Population Dynamics and Interspecific Interactions (eds Tilman, D. &
At this stage, community phylogenetic analyses dealing with Kareiva, P.). Princeton University Press, Princeton, NJ, pp. 250–
entire communities (as opposed to particular clades) are 268.
limited by their requirement for complete phylogenies, and Cornell, H.V. & Lawton, J.H. (1992). Species interactions, local and
the kinds of sampling biases that may arise in such analyses regional processes, and limits to the richness of ecological
communities: theoretical perspective. J. Anim. Ecol., 61, 1–12.
are just beginning to be explored (Hardy & Senterre 2007).
Cornell, H.V., Karlson, R.H. & Hughes, T.P. (2008). Local-regional
No single study has yet combined more than one of the species richness relationships are linear at very small to large
steps we suggest here. Data limitations remain perhaps the scales in west-central Pacific corals. Coral Reefs, 27, 145–
most formidable obstacles. Nonetheless, we conclude with 151.
the optimistic suggestion that the ever-increasing availability Cottenie, K., Michels, E., Nuytten, N. & De Meester, L. (2003).
of large environmental, biogeographical and phylogenetic Zooplankton metacommunity structure: regional vs. local pro-
databases, combined with the new analytic tools ranging cesses in highly connected ponds. Ecology, 84, 991–1000.
Currie, D.J., Mittelbach, G.G., Cornell, H.V., Field, R., Guégan,
from SEM and geostatistics to community phylogenetics,
J.-F., Hawkins, B.A. et al. (2004). Predictions and tests of
offers the possibility that creative ecologists and evolution- climate-based hypotheses of broad-scale variation in taxonomic
ists will fulfil the hope expressed in classic works such as richness. Ecol. Lett., 7, 1121–1134.
Geographical Ecology by MacArthur (1972) and Species Diversity Fargione, J., Brown, C.S. & Tilman, D. (2003). Community
in Ecological Communities edited by Ricklefs & Schluter (1993): assembly and invasion: an experimental test of neutral versus
a fuller understanding of the large-scale regional processes niche processes. Proc. Natl Acad. Sci. USA, 100, 8916–8920.
that shape local ecological communities. Forbes, E.A. & Chase, J.M. (2002). The role of habitat connectivity
and landscape geometry in experimental zooplankton meta-
communities. Oikos, 96, 433–440.
ACKNOWLEDGEMENTS Fox, J.W. & Srivastava, D.S. (2006). Predicting local-regional
richness relationships using island biogeography models. Oikos,
We thank T. Fukami, J. Chase, M. Holyoak and two 133, 376–382.
anonymous referees for insights that greatly improved an Fox, J.W., McGrady-Steed, J. & Petchey, O.L. (2000). Testing for
earlier version of the manuscript. local species saturation with nonindependent regional species
pools. Ecol. Lett., 3, 198–206.
Freestone, A.L. & Harrison, S. (2006). Regional enrichment of
REFERENCES local assemblages is robust to variation in local productivity,
abiotic gradients, and heterogeneity. Ecol. Lett., 9, 95–102.
Belmaker, J., Ziv, Y., Shashar, N. & Connolly, S. (2008). Regional Fukami, T. (2004a). Assembly history interacts with ecosystem size
variation in the hierarchical partitioning of alpha and beta to influence species diversity. Ecology, 85, 3234–3242.
diversity in coral-dwelling fishes. Ecology, in press. Fukami, T. (2004b). Community assembly along a species pool
Bengtsson, J. (1989). Interspecific competition increases local gradient: implications for multiple-scale patterns of species
extinction rate in a metapopulation system. Nature, 340, 713– diversity. Popul. Ecol., 46, 137–147.
715. Gotelli, N.J. & Colwell, R.K. (2001). Quantifying biodiversity:
Brooks, D.R. & McLennan, D.A. (2002). The Nature of Diversity: An procedures and pitfalls in the measurement and comparison of
Evolutionary Voyage of Discovery. University of Chicago Press, species richness. Ecol. Lett., 4, 379–391.
Chicago, IL, USA. Grace, J.B. (1999). The factors controlling species density in her-
Cadotte, M.W. (2006). Metacommunity influences on community baceous plant communities: an assessment. Perspect. Plant Ecol.
richness at multiple spatial scales: a microcosm experiment. Evol. Syst., 2, 1–28.
Ecology, 87, 1008–1016. Grace, J.B. (2006). Structural Equation Modeling and Natural Systems.
Caley, M.J. & Schluter, D. (1997). The relationship between local Cambridge University Press, Cambridge.
and regional diversity. Ecology, 78, 70–80.

 2008 Blackwell Publishing Ltd/CNRS

978 S. Harrison and H. Cornell Review and Synthesis

Grace, J.B., Anderson, T.M., Smith, M.D., Seabloom, E., Andel- MacArthur, R.H. & Wilson, E.O. (1967). The Theory of Island Bio-
man, S.J., Meche, G. et al. (2007). Does species diversity limit geography. Princeton University Press, Princeton, NJ.
productivity in natural grasslands? Ecol. Lett., 10, 680–689. McPeek, M. & Brown, J. (2000). Building a regional species pool:
Hardy, O.J. & Senterre, B. (2007). Characterizing the phylogenetic diversification of the Enallagma damselflies in eastern North
structure of communities by an additive partitioning of phylo- America. Ecology, 81, 904–920.
genetic diversity. J. Ecol., 95, 493–506. Miller, T.E. & Kneitel, J.M. (2005). Inquiline communities in
Harrison, S. & Cornell, H.V. (2007). Merging evolutionary and pitcher plants as a prototypical metacommunity. In: Metacom-
ecological explanations for geographic gradients in species munities: Spatial Dynamics and Ecological Communities (eds Holyoak,
richness. Am. Nat., 170, S1–S4. M., Leibold, M.A. & Holt, R.). University of Chicago Press,
Harrison, S. & Grace, J.B. (2007). Biogeographic affinity helps Chicago, IL, pp. 122–145.
explain productivity-richness relationships at regional and local Mittelbach, G.G., Steiner, C.F., Scheiner, S.M., Gross, K.L., Rey-
scales. Am. Nat., 170, S5–S15. nolds, H.L., Waide, R.B. et al. (2001). What is the observed
Harrison, S., Safford, H.D., Grace, J.B., Viers, J.H. & Davies, K.F. relationship between species richness and productivity? Ecology,
(2006). Regional and local species richness in an insular envi- 82, 2381–2396.
ronment: serpentine plants in California. Ecol. Monogr., 76, Mittelbach, G., Schemske, D., Cornell, H., Allen, A., Brown, J., Bush,
41–56. M. et al. (2007). Evolution and the latitudinal diversity gradient:
Hastings, A. (1987). Can competition be detected using species speciation, extinction, and biogeography. Ecol. Lett., 10, 315–331.
co-occurrence data? Ecology, 68, 117–123. Moore, B.R. & Donoghue, M.J. (2007). Correlates of diversification
Hawkins, B.A., Field, R., Cornell, H.V., Currie, D.J., Guegan, J.-F., in the plant clade Dipsacales: geographic movement and evo-
Kaufman, D.M. et al. (2003). Energy, water and broad-scale lutionary innovation. Am. Nat., 170, S28–S55.
patterns of species richness. Ecology, 84, 3105–3117. Mouquet, N. & Loreau, M. (2003). Community patterns in source-
Hawkins, B.A., Diniz-Filho, J.A.F., Jaramillo, C.A. & Soeller, S.A. sink metacommunities. Am. Nat., 162, 544–557.
(2007). Climate, niche conservatism, and the global bird diversity Mouquet, N., Munguia, P., Kneitel, J.M. & Miller, T.E. (2003).
gradient. Am. Nat., 170, S16–S27. Community assembly time and the relationship between local
He, F., Gaston, K.J., Conner, E.F. & Srivastava, D.S. (2005). The and regional species richness. Oikos, 103, 618–626.
local-regional relationship: immigration, extinction, and scale. Mouquet, N., Ledley, P., Meriget, J. & Loreau, M. (2004). Immi-
Ecology, 86, 360–365. gration and local competition on herbaceous plant communities:
Hillebrand, H. (2004). On the generality of the latitudinal diversity a three-year seed-sowing experiment. Oikos, 104, 77–90.
gradient. Am. Nat., 163, 192–211. Pajunen, V.I. & Pajunen, I. (2003). Long-term dynamics in rock
Hillebrand, H. (2005). Regressions of local on regional diversity do pool Daphnia metapopulations. Ecography, 26, 731–738.
not reflect the importance of local interactions or saturation of Partel, M. (2002). Local plant diversity patterns and evolutionary
local diversity. Oikos, 110, 195–198. history at the regional scale. Ecology, 83, 2361–2366.
Hillebrand, H. & Blenckner, T. (2002). Regional and local impact Partel, M., Lauriisto, L. & Zobel, M. (2007). Contrasting plant
on species diversity-from pattern to process. Oecologia, 132, 479– productivity-diversity relationships across latitude: the role of
491. evolutionary history. Ecology, 88, 1091–1097.
Holyoak, M.A. & Lawler, S.P. (1996). Persistence of an extinction- Qian, H., White, P.S. & Song, J.-S. (2007). Effects of regional vs.
prone predator-prey interaction through metapopulation ecological factors on plant species richness: an intercontinental
dynamics. Ecology, 77, 1867–1879. analysis. Ecology, 8, 1440–1453.
Hugueny, B., Cornell, H.V. & Harrison, S. (2007). Metacommunity Ricklefs, R.E. (1987). Community diversity: the relative roles of
models predict the local-regional richness relationship in a nat- regional and local processes. Science, 235, 167–171.
ural system. Ecology, 88, 1696–1706. Ricklefs, R.E. (2004). A comprehensive framework for global
Huston, M.A. (1999). Local processes and regional patterns: patterns in biodiversity. Ecol. Lett., 7, 1–15.
appropriate scales for understanding variation in the diversity of Ricklefs, R.E. (2007). History and diversity: explorations at the
plants and animals. Oikos, 86, 393–401. intersection of ecology and evolution. Am. Nat., 170, S56–S70.
Karlson, R.H., Cornell, H.V. & Hughes, T.P. (2004). Coral com- Ricklefs, R.E. & Schluter, D. (1993). Species Diversity in Ecological
munities are regionally enriched along an oceanic biodiversity Communities: Historical and Geographical Perspectives. University of
gradient. Nature, 429, 867–870. Chicago Press, Chicago, IL, pp. 243–252.
Leibold, M.A., Holyoak, M., Mouquet, N., Amarasekare, P., Chase, Rosenzweig, M.L. & Ziv, Y. (1999). The echo pattern of species
J.M., Hoopes, M.F. et al. (2004). The metacommunity concept: a diversity: pattern and processes. Ecography, 22, 614–628.
framework for multi-scale community ecology. Ecol. Lett., 7, Roy, K. & Goldberg, E.E. (2007). Origination, extinction, and
601–613. dispersal: integrative models for understanding present-day
Loreau, M. (2000). Are communities saturated? On the role of a, b, diversity gradients. Am. Nat., 170, S71–S85.
and c diversity. Ecol. Lett., 3, 73–76. Shurin, J.B. & Allen, E.G. (2001). Effects of competition, preda-
Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., tion, and dispersal on species richness at local and regional
Hector, A. et al. (2001). Biodiversity and ecosystem functioning: scales. Am. Nat., 158, 624–637.
current knowledge and future challenges. Science, 294, 804–808. Shurin, J.B. & Srivastava, D.S. (2005). New perspectives on local
Losos, J. (1996). Phylogenetic perspectives on community ecology. and regional diversity: beyond saturation. In: Metacommunities:
Ecology, 77, 1344–1354. Spatial Dynamics and Ecological Communities (eds Holyoak, M.,
MacArthur, R.H. (1972). Geographical Ecology: Patterns in the Distri- Leibold, M.A. & Holt, R.). University of Chicago Press, Chicago,
bution of Species. Princeton University Press, Princeton, NJ. IL, pp. 399–417.

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Regional causes of richness 979

Shurin, J.B., Havel, J.E., Leibold, M.A. & Pinel-Alloul, B. (2000). Webb, C.O., Ackerly, D.D., McPeek, M.A. & Donoghue, M.J.
Local and regional zooplankton species richness: a scale-inde- (2002). Phylogenies and community ecology. Ann. Rev. Ecol.
pendent test for saturation. Ecology, 11, 3062–3073. Evol. Syst., 33, 475–505.
Srivastava, D.S. (1999). Using local-regional richness plots to Wiens, J.J. (2007). Global patterns of diversification and species
test for saturation: potentials and pitfalls. J. Anim. Ecol., 68, richness in amphibians. Am. Nat., 170, S86–S106.
1–16. Wiens, J.J. & Donoghue, M.J. (2004). Historical biogeography,
Starzomski, B.M.., Parker, R.L. & Srivastava, D.S. (2008). Does ecology and species richness. Trends Ecol. Evol., 19, 639–644.
regional species richness determine local species richness? An Witman, J.D., Etter, R.J. & Smith, F. (2004). The relationship
experimental test of saturation theory. Ecology, in press. between regional and local species diversity in marine benthic
Stohlgren, T.J., Barnett, D.T., Jarnevich, C.S., Flather, C. & Kar- communities: a global perspective. Proc. Natl Acad. Sci. USA, 101,
tesz, C. (2008). The myth of plant species saturation. Ecol. Lett., 15664–15669.
11, 315–326. Wootton, J.T. (1994). Predicting direct and indirect effects: an
Swenson, N.G., Enquist, B.J., Pither, J., Thompson, J. & Zimm- integrated approach using experiments and path analysis. Ecology,
erman, J.K. (2006). The problem and promise of scale depen- 75, 151–165.
dency in community phylogenetics. Ecology, 87, 2418–2424. Zobel, M., van der Maarelm, E. & Dupre, C. (1998). Species pool:
Tilman, D. (1997). Community invasibility, recruitment limitation, the concept, its determination and significance for community
and grassland biodiversity. Ecology, 78, 81–92. structure. Appl. Veg. Sci., 1, 55–66.
Tilman, D. (2004). Niche tradeoffs, neutrality, and community
structure: a stochastic theory of resource competition, invasion,
and community assembly. Proc. Natl Acad. Sci. USA, 101, 10854–
10861. Editor, Tadashi Fukami
Turnbull, L.A., Crawley, M.J. & Rees, M. (2000). Are plant popu- Manuscript received 6 January 2008
lations seed limited? A review of seed sowing experiments.
Oikos, 88, 225–238.
First decision made 9 February 2008
Wardle, D.A. (2001). Experimental demonstration that plant Manuscript accepted 1 May 2008
diversity reduces invasibility: evidence of a biological mechanism
or a consequence of sampling effect? Oikos, 95, 161–170.

 2008 Blackwell Publishing Ltd/CNRS