Notas de opinión

 

Ecological niche shifts and environmental space anisotropy: a cautionary note

 

Desplazamientos en el nicho y la anisotropía del espacio ambiental: una nota precautoria

 

Jorge Soberón^* and A. Townsend Peterson

 

Biodiversity Institute and Department of Ecology and Evolutionary Biology, University of Kansas. Dyche Hall, 1345 Jayhawk Boulevard. Lawrence, KS 66045 USA. ^*jsoberon@ku.edu.

 

Recibido: 15 marzo 2011;
aceptado: 11 junio 2011

 

Abstract

The anisotropic structure of climatic space may cause significant (and to a large extent unappreciated) non–evolutionary niche shifts. This can be seen mostly in the context of spatial transferability of ecological niche models. We explore this effect using a virtual species in the United States. We created a simple virtual species by postulating its fundamental niche as an ellipse in a two–dimensional realistic climatic space. The climatic combinations defined by the ellipse were projected in the geography of the United States and 2 regions of equal area were selected. The structure of niche in the 2 areas is compared. It is shown that the 2 regions have differently positioned subsets of the environmental space, which creates "shifts" in the realized niches despite the fact that no evolution and no biotic interactions are present. The most parsimonious hypothesis when ecological niche modeling reveals shifts in the realized niche is that environmental space is heterogeneous. Without considering differences in the structure of environmental space no speculation about niche evolution or the role of competitors should be attempted.

Key words: fundamental niche, realized niche, existing niche, ecological niche modeling, species distribution modeling, environmental space.

 

Resumen

La estructura anisotrópica del espacio climático puede causar desplazamientos significativos no evolutivos en los nichos de las especies. Este efecto poco apreciado en la literatura se manifiesta con gran claridad cuando se realizan transferencias espaciales de modelos de nicho ecológico. Se explora este efecto utilizando una especie virtual en los Estados Unidos. Se creó una especie virtual simplificada postulando su nicho fundamental en forma de una elipse en un espacio realista de 2 dimensiones. Las combinaciones climáticas definidas por la elipse se proyectaron en la geografía de los Estados Unidos y se seleccionaron 2 regiones de igual superficie espacial. Se compara la estructura del nicho en las 2 regiones, mostrando que estas 2 regiones espaciales presentan subconjuntos distintos del espacio de variables ambientales, lo cual induce "desplazamientos" en los nichos realizados, a despecho de que no existió evolución del nicho fundamental y de que no hay competidores ni otras interacciones bióticas presentes. Al modelar nichos ecológicos y transferirlos espacialmente, si se revelan desplazamientos en los nichos realizados, la hipótesis más parsimoniosa para explicar los desplazamientos es que el espacio ambiental tiene diferente estructura en las diferentes localidades. Sin considerar la estructura del espacio ambiental no debe especularse sobre evolución del nicho, ni sobre posibles efectos bióticos sobre él.

Palabras clave: nicho fundamental, nicho realizado, nicho existente, modelación de nichos ecológicos, modelación de distribuciones de especies, espacio medioambiental.

 

Introduction

Ecological niche modeling (ENM) is increasingly being used to predict potential ranges of invasive species (Soberón et al., 2001; Peterson, 2003; Thuiller et al., 2005; Ficetola et al., 2007). An important assumption in making such exercises feasible is that of conservatism in the ecological niche characteristics (Peterson et al., 1999; Ackerly, 2003; Wiens and Graham, 2005; Peterson, 2011): in essence, the idea that the position and shape of the fundamental ecological niche evolve slowly relative to the invasion process. If the fundamental niche of a species is not conserved, a species would be able to invade novel regions of the world presenting environmental conditions very different from the original ones, and predictivity of the phenomenon would be nil. Similar reasoning applies to many other applications of the niche modeling idea: forecasting effects of global climate change (Thomas et al., 2004), hindcasting historical distributions (Waltari et al., 2007), etc.

Several recent studies have concluded that particular species invasions are associated with niche shifts (Broennimann et al., 2007; Fitzpatrick et al., 2007; Medley, 2010). These shifts are evidenced by displacements in the position of clouds of invaded–range occurrence points in environmental space relative to the environmental space of the native range. Niche shifts have been attributed to evolution of the fundamental niche, or to changes in its expression owing to presence of competitors, predators, or pathogens (Randin et al., 2006; Broennimann et al., 2007). However, another possible cause is simply that the structure of the environmental space differs between the 2 regions where the species is being studied (Elith and Graham, 2009). This last possibility is not always considered carefully, although the problem was recognized in the early niche measurement literature (Colwell and Futuyma, 1971; Green, 1971; Carnes and Slade, 1982). Nonetheless, this point has deep implications for niche modeling and for analysis of niche conservatism and shifts.

To clarify the issue, we begin by establishing some terminology. Using terminology and symbols may appear unnecessary, if not downright pedantic, but in the end it helps significantly to prevent confusion when referring to complex relationships between concepts. Ecological niche modeling moves in a duality (Colwell and Rangel, 2009) of a geographic space (G) and the space (E) of environmental combinations of scenopoetic variables (Hutchinson, 1978), as illustrated in Figure 1. G, the geographic region relevant to the problem at hand is divided into a grid, and for every cell g in G, we have a vector of environmental variables. Geographic information systems allow mapping from portions of G to E and vice versa (Guisan and Zimmermann, 2000). Typically, E–spaces are anisotropic (in the sense that its aspect changes with direction: see scatterplot in Fig. 1) and very irregular in shape, mostly at large geographic extents, a quality not frequently mentioned in the niche modeling literature.

The fundamental niche, N_F (Hutchinson, 1957), is the multivariate range of physiological tolerances to variables like temperature, water availability, etc., within which a species will have positive intrinsic population growth rates (Birch, 1953; Hooper et al., 2008). It is feasible to transform such biophysical, microclimatic ranges into broader climatic terms (Kearney and Porter, 2009), allowing one to express niches in terms of coarse–grained environmental variables. Jackson and Overpeck (2000) identified the idea of a "potential niche", which we denote by N^*_F. This is the intersection of the fundamental niche with the available environmental space (Green, 1971): N^*_F = N_F E. In other words, N^*_F is the set of climatic features actually existing on the relevant landscape and time, that coincides with the physiological requirements of the species. Notice that E and therefore N^*_F depend on the focal region under study: the same fundamental niche in a changed reference region may produce a different N^*_F. As illustrated in Figure 1, geographically, the environments in N^*_F are found in the actual occupied range (blue regions denoted by G₀), and also in the remaining regions in G that potentially could be invaded because they are also suitable (denoted by G_I) but still inaccessible. The realized niche, NR, is the subset of E in the occupied area (G₀) of the species. In symbols

Where the simbol η (Go) represents the environments in Go. This simple set relation expresses that the environments constituting the realized niche are equal to the environments in the actual, occupied area of distribution, and in turn these are a subset of those existing environments in the area of reference that intersect with the fundamental niche. This is a very important equation that may be regarded as the fundamental hypothesis of niche modeling, and verbally it has been known since Hutchinson (1957). He proposed that N_R would be smaller than N_F owing only to the effect of negative biotic interactions (Colwell and Futuyma, 1971; Austin, 1999). More recent authors have stressed that NR should be a subset of N_F, because, first, what has geographic existence is N^*_F, which almost certainly is smaller than N_F, and second, because N^*_F can be reduced by interactions like competition, or not expressed fully by lack of dispersal capacities of the species (Jackson and Overpeck, 2000; Pulliam, 2000; Svenning and Skov, 2004; Soberón and Peterson, 2005; Pearson, 2007; Colwell and Rangel, 2009). Operationally, N_F may only be measured experimentally, or based on first principles of physiology and biophysics (Kearney and Porter, 2009), taking into account the fact that adaptation to local conditions may create subpopulations with slightly different ranges of tolerances (Labra et al., 2009), and that strictly speaking, one cannot ever measure the fundamental niche, but only assess the probability that an environment is part of it (Godsoe, 2009). On the other hand, N_R can only be estimated from unbiased observations of both presences and absences, since the results of interactions are difficult to anticipate. The difference between N_F and N_R, taking into account the actual structure of E, is major, so failing to consider it confuses interpretation considerably.

For the sake of simplicity and illustration, we assume that N_F can be represented using convex shapes that contain environments favorable for the species (Maguire, 1973). Notice, however, that neither N^*_F nor N_R should be expected to be convex in shape, since N^*_F might have been distorted by the shape of E, and N_R by movement limitations or biological interactions. The 3 sets N_F, N^*_F and N_R can be measured in terms of their position, shape and size (Jackson and Overpeck, 2000; Pearman et al., 2007; Rangel et al., 2007); what matters most from the perspective of niche conservatism is whether the position of NF within E is changing, and at what speed (Soberón and Nakamura, 2009).

However, as we said before, measuring the N_F can only be done, generally speaking, mechanistically or experimentally (Kearney, 2006; Soberón, 2007; Kearney and Porter, 2009). The widely used correlative methods estimate vaguely defined objects: an area of distribution somewhere between the actual and the potential (Jiménez–Valverde et al., 2008), and thus the corresponding environmental subsets lie between N^*_F and N_R, as expressed in equation (1). Unfortunately when presence–only data is used, without extra information it is impossible to assess where exactly the position of the estimate lies (Soberón, 2010). If, however, distributions are modeled using correlative methods that include unbiased data documenting absences, most likely one gets an estimation of the actual occupied area of the species, GO, with associated N_R= η(G₀). At the other extreme, lacking data on absences and resorting only to envelope methods, it is likely that estimates shift towards the potential area of distribution (G₀G_I), and therefore the environmental space associated with η(G₀G_I) shifts more towards N*F (Jiménez–Valverde et al., 2008). In any case, correlative methods clearly do not provide immediate evidence about shifts in the NF, but only, indirectly, about N^*_F and N_R.

Observing displacements in N_R estimates for the native and invaded ranges of Centaurea maculosa, Broennimann et al. (2007) suggested 2 explanations: (i) genuine evolutionary change in N_F, or (ii) actions of ecological factors like lack of suitable habitat or presence of competitors are modifying the N_R. In a paper discussing transferability of predictions of species distributions, Randin et al. (2006) also explained niche shifts by differences in land–use history, phenotypic plasticity, and the ranges of environmental predictors being different. Dormann et al. (2010), studying climatic niche conservation in European mammals, were also aware of the possibility that differences in niche space may affect results, and developed a test to check whether "climate niches" (without specification of whether it is the fundamental, the realized or an intermediate) of sister species are affected by the underlying structure of E–space. Their test appears to be valid as regards to realized niches, but perhaps not so if used as evidence of fundamental niche conservatism.

Finally, in a very recent paper, Godsoe (2010) used virtual species with identical environmental requirements (i.e., fundamental niches) to explore how gradients and dispersal limitations affect the capacity of different methods to detect differences in both fundamental and realized niches. His results further illustrate the difficulties of making inferences about the fundamental niches on the basis of observed distributional data.

These results are very recent, and the implications of anisotropy of E–space for studying fundamental niche conservatism have not been appreciated widely. The structure of existing environmental combinations may vary radically from 1 region of G to another, and through time as well, so species with conserved N_F may have N_R significantly shifted in different regions, just because the environmental combinations available in 2 regions of the planet are different. This point is key in interpreting niche conservatism, as we will show below.

 

Materials and methods

We illustrate the effect of differences in existing conditions on N_R by constructing a virtual species with a known hypothetical fundamental niche, in a world with no biotic interactions, and unlimited ability to expand within a given region. To construct the E space, we start with the 19 bioclimatic variables in WorldClim (Hijmans et al., 2005) for the Western Hemisphere, at a spatial resolution of 5'. To reduce dimensionality and standardize variables, we performed a z–transformation of the data and then a PCA analysis. The first 5 components explain almost 98% of the variance, but we used just the first 2 axes (which explain around 90%) in order to simplify the figures. We assumed a 2–dimensional ellipsoid as the fundamental niche for the hypothetical species using the following equation:

where x are vectors of 2 coordinates, µ= (–0.295, 0.4407) is the centroid of N_F, and

is a covariance matrix with eigenvectors and eigenvalues determining the directions of the axes of the N^*_F and its area (Legendre and Legendre, 1998). The above parameters define the ellipse illustrated in Figures 2 and 3. Its intersection with the 2–dimensional E space (N^*_F) is projected to geographic space as a potential distributional area. For the sake of illustration, we selected 2 equal–sized regions in the United States (Figs. 3A, B), each containing environments within N_F.

To compare results we followed Broennimann et al. (2007) and analyzed the data using between–classes inertia analysis (Doledec et al., 2000) with the ade4 package of R which gave an inter–inertia ratio of 0.273, with a p < 10^–5, obtained by Monte Carlo simulation with 10 000 replicates. Inertia analysis explores the position and shape of clouds of points in environmental space by studying their distance to the centroid of the entire available space, and the degree of scatter on a vector linking the global centroid to the centroid of one particular cloud of points. Other methods to compare niche shifts exist (Warren et al., 2008), but we wanted to use one that has already been used to document niche shifts in an evolutionary context.

 

Results

The occurrence points falling in N^*_F of the species in the northern and southern regions in the map, respectively, are shown as blue and red in Figure (3), respectively. Despite wide overall overlap, the climatic combinations corresponding to the 2 ranges are different, and a species capable of reaching every grid cell would have different N^*_F (and thus realized niche) in the northern and the southern regions. This result holds true even assuming: (i) constant NF, (ii) no competitors or interactors of any kind, and (iii) full dispersal equilibrium within each subregion.

We plot the N^*_F of the 2 regions (Fig. 4), showing substantial niche shift between them. This result parallels strikingly the niche shifts observed by Broennimann et al. (2007) when analyzing invasion by Centaurea, which they explained in evolutionary or ecological terms. Our example, however, mimics realized niche shifts resulting only from the anisotropy of E–space, which may be frequently be a more parsimonious explanation for niche shifts.

 

Discussion

The conclusion is straightforward: documenting shifts in an estimated realized niche, which is what correlative methods most likely do, yields little information about the evolution of the fundamental niche, since shifts in N_R result from combinations of evolutionary, ecological and geographic effects. Without ancillary information, documenting shifts in estimates of N_R does not inform about changes in the physiologically– defined N_R, nor about ecological changes in N_F. The correct null hypothesis when transferring niche predictions spatially should be lack of change in the centroids of the N_R conditioned to the anisotropies in E; in other words, appropriate comparisons will include considerations of the structure and composition of the E space in each region. Such tests have been implemented in the background similarity tests of Warren et al. (2008), which consider niche differences relative to a specific area of interest, and see also Dormann et al. (2010) and Godsoe (2010). Notice, however, that documenting stasis in the N_R probably implies stasis in N_F as well.

Beginning with Broennimann (2007) and Fitzpatrick et al. (2007), numerous studies have addressed the question of niche shifts during the invasion process (Pearman et al., 2007; Broennimann and Guisan, 2008; Rödder and Lötters, 2009; Dormann et al., 2010; Medley, 2010). In general, they have concluded that niche shifts are occurring, an idea with which we concur if the understanding is that they refer to the realized niche. As should be clear from the example we have presented herein, the question of which niche is shifting is crucial. It is clear that the same fundamental niche N_F, expressed in different regions with correspondingly distinct environmental spaces E' and E'' will produce different existing fundamental niches and therefore, in all likelihood, different NR, even in the absence of competitors and evolutionary processes.

The idea that niche comparisons should be performed in relation to available ecological space is not new. The literature of niche measures very often makes explicit reference to available ecological space (Colwell and Futuyma, 1971; Green, 1971; Doledec et al., 2000; Elith and Burgman, 2002; Hirzel et al., 2002; Calenge et al., 2008; Warren et al., 2008; Dormann et al., 2010). It is important that niche modelers appreciate more fully the fact that environmental spaces are highly heterogeneous and that attempts to understand changes in either N_F or N_R, must therefore take into account the structure of the available environmental space E. The problem of the definition of the available niche space is a deep one, which is the subject of a separate treatment (Barve et al., 2011).

 

Acknowledgements

We thank our colleagues in the KU Niche Modeling Group for helpful discussions, and to David Ackerly, Olivier Broennimann and 2 anonymous referees for their comments. This work was partially financed by a Microsoft Research grant.

 

Literature cited

Ackerly, D. D. 2003. Community assembly, niche conservatism, and adaptive evolution in changing environments. International Journal of Plant Science 164:S165–S184.         [ Links ]

Austin, M. 1999. A silent clash of paradigms: some inconsistencies in community ecology. Oikos 86:170–178.         [ Links ]

Barve, N., V. Barve, A. Jiménez–Valverde, A. Lira–Noriega, S. P. Maher, A. T. Peterson, J. Soberón and F. Villalobos. 2011. The crucial role of the accessible area in ecological niche modeling and species distribution modeling. Ecological Modelling 222:1810–1819.         [ Links ]

Birch, L. C. 1953. Experimental background to the study of the distribution and abundance of insects: I. The influence of temperature, moisture and food on the innate capacity for increase of three grain beetles. Ecology 34:698–711.         [ Links ]

Broennimann, O. and A. Guisan. 2008. Predicting current and future biological invasions: both native and invaded ranges matter. Biology Letters 4:585–589.         [ Links ]

Broennimann, O., U. A. Treier, H. Muller–Schaerer, W. Thuiller, A. T. Peterson and A. Guisan. 2007. Evidence of climatic niche shift during biological invasion. Ecology Letters 10:701–709.         [ Links ]

Calenge, C., G. Darmon, M. Basille, A. Loison and J. M. Jullien. 2008. The factorial decomposition of the Mahalanobis distances in habitat selection studies. Ecology 89:555–566.         [ Links ]

Carnes, B. A. and N. Slade. 1982. Some comments on niche analysis in canonical space. Ecology 63:888–893.         [ Links ]

Colwell, R. K. and D. Futuyma. 1971. On the measurement of niche breadth and overlap. Ecology 52:567–576.         [ Links ]

Colwell, R. K. and T. F. Rangel. 2009. Hutchinson's duality: the once and future niche. Proceedings of the National Academy of Sciences USA 106:19644–19650.         [ Links ]

Doledec, S., D. Chessel and C. Gimaret–Carpentier. 2000. Niche separation in community analysis: A new method. Ecology 81:2914–2927.         [ Links ]

Dormann, C. F., B. Gruber, M. Winter and D. Hermann. 2010. Evolution of climate niches in European mammals? Biology Letters 6:229–232.         [ Links ]

Elith, J. and M. Burgman. 2002. Predictions and their validation: Rare plants in the central highlands, Victoria. In Predicting species occurrences: issues of scale and accuracy, J. M. Scott, P. J. Heglund and M. L. Morrison (eds.). Island Press, Washington, D.C. p. 303–313.         [ Links ]

Elith, J. and C. Graham. 2009. Do they? How do they? WHY do they differ? On finding reasons for differing performances of species distribution models. Ecography 32:66–77.         [ Links ]

Ficetola, G. F., W. Thuiller and C. Miaud. 2007. Prediction and validation of the potential global distribution of a problematic alien invasive species –– The American bullfrog. Diversity and Distributions 13:476–485.         [ Links ]

Fitzpatrick, M. C., J. E. Weltzin, N. Sanders and R. R. Dunn. 2007. The biogeography of prediction error: why does the introduced range of the fire ant over–predict its native range? Global Ecology and Biogeography 16:24–33.         [ Links ]

Godsoe, W. 2009. I can't define the niche but I know it when I see it: a formal link between statistical theory and the ecological niche. Oikos 119:53–60.         [ Links ]

Godsoe, W. 2010. Regional variation exaggerates ecological divergence in niche models. Systematic Biology 59:298–306.         [ Links ]

Green, R. H. 1971. A multivariate statistical approach to the Hutchinsonian niche: bivalve molluscs of central Canada. Ecology 52:544–556.         [ Links ]

Guisan, A. and N. Zimmermann. 2000. Predictive habitat distribution models in ecology. Ecological Modelling 135:147–186.         [ Links ]

Hijmans, R. J., S. Cameron, J. Parra, P. G. Jones and A. Jarvis. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25:1965–1978.         [ Links ]

Hirzel, A. H., J. Hausser, D. Chessel and N. Perrin. 2002. Ecological–niche factor analysis: how to compute habitat–suitability maps without absence data? Ecology 83:2027–2036.         [ Links ]

Hooper, H., R. Connon, A. Callaghan, G. Fryer, S. Yarwood–Buchanan, J. Biggs, S. Maund, T. Hutchinson and R. M. Sibly. 2008. The ecological niche of Daphnia magna characterized using population growth rate. Ecology 89:1015–1022.         [ Links ]

Hutchinson, G. E. 1957. Concluding remarks. Cold Spring harbor Symposia on Quantitative Biology 22:415–427.         [ Links ]

Hutchinson, G. E. 1978. An Introduction to Population Ecology. Yale University Press, New Haven. 260 p.         [ Links ]

Jackson, S. T. and J. T. Overpeck. 2000. Responses of plant populations and communities to environmental changes of the late Quaternary. Paleobiology 26 (Supplement):194–220.         [ Links ]

Jiménez–Valverde, A., J. M. Lobo and J. Hortal. 2008. Not as good as they seem: the importance of concept in species distribution modelling. Diversity and Distributions 14:885–890.         [ Links ]

Kearney, M. 2006. Habitat, environment and niche: what are we modelling? Oikos 115:186–191.         [ Links ]

Kearney, M. and W. P. Porter. 2009. Mechanistic niche modelling: combining physiological and spatial data to predict species' ranges. Ecology Letters 12:334–350.         [ Links ]

Labra, A., J. Pienaar and T. F. Hansen. 2009. Evolution of thermal physiology in Liolaemus lizards: adaptation, phylogenetic inertia, and niche tracking. The American Naturalist 174:204–220.         [ Links ]

Legendre, P. and L. Legendre. 1998. Numerical Ecology. Secon Edition edition. Elsevier Scientific Publishing Company, Amsterdam. 853 p.         [ Links ]

Maguire, B. 1973. Niche response structure and the analytical potentials of its relationship to the habitat. American Naturalist 107:213–246.         [ Links ]

Medley, K. A. 2010. Niche shifts during the global invasion of the Asian tiger mosquito, Aedes albopictus Skuse (Culicidae), revelaed by reciprocal distribution models. Global Ecology and Biogeography 19:122–133.         [ Links ]

Pearman, P. B., A. Guisan, O. Broennimann and C. Randin. 2007. Niche dynamics in space and time. Trends in Ecology & Evolution 23:149–158.         [ Links ]

Pearson, R. G. 2007. Species' distribution modeling for conservation educators and practitioners. Synthesis. American Museum of Natural History, New York. 50 p.         [ Links ]

Peterson, A. T. 2003. Predicting the geography of species' invasions via ecological niche modeling. Quarterly Review of Biology 78:419–433.         [ Links ]

Peterson, A. T. 2011. Ecological niche conservatism: a time–structured review of evidence. Journal of Biogeography In press.         [ Links ]

Peterson, A. T., J. Soberón and V. Sánchez–Cordero. 1999. Conservatism of ecological niches in evolutionary time. Science 285:1265–1267.         [ Links ]

Pulliam, R. 2000. On the relationship between niche and distribution. Ecology Letters 3:349–361.         [ Links ]

Randin, C. F., T. Dirnbock, S. Dullinger, N. Zimmermann, M. Zappa and A. Guisan. 2006. Are niche–based models transferable in space? Journal of Biogeography 33:1689–1703.         [ Links ]

Rangel, T. F., J. A. Diniz–Filho and R. K. Colwell. 2007. Species richness and evolutionary niche dynamics: a spatial pattern–oriented simulation experiment. American Naturalist 170:602–616.         [ Links ]

Rödder, D. and S. Lötters. 2009. Niche shift versus niche conservatism? Climatic characteristics of the native and invasive ranges of the Mediterranean house gecko (Hemidactylus turcicus). Global Ecology and Biogeography 18:674–687.         [ Links ]

Soberón, J. 2007. Grinnellian and Eltonian niches and geographic distributions of species. Ecology Letters 10:1115–1123.         [ Links ]

Soberón, J. 2010. Niche and distributional range: a population ecology perspective. Ecography 33:159–167.         [ Links ]

Soberón, J., J. Golubov and J. Sarukhan. 2001. The importance of Opuntia in Mexico and routes of invasion and impact of Cactoblastis cactorum (Lepidoptera: Pyralidae). Florida Entomologist 84:486–492.         [ Links ]

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.         [ Links ]

Soberón, J. and A. T. Peterson. 2005. Interpretation of models of fundamental ecological niches and species' distributional areas. Biodiversity Informatics 2:1–10.         [ Links ]

Svenning, J. C. and F. Skov. 2004. Limited filling of the potential range in European tree species. Ecology Letters 7:565–573.         [ Links ]

Thomas, C. D., A. Cameron, R. E. Green, M. Bakkenes, L. J. Beaumont, Y. C. Collingham, B. F. N. Erasmus, M. Ferreira de Siqueira, A. Grainger, L. Hannah, L. Hughes, B. Huntley, A. S. Van Jaarsveld, G. E. Midgely, L. Miles, M. A. Ortega–Huerta, A. T. Peterson, O. L. Phillips and S. E. Williams. 2004. Extinction risk from climate change. Nature 427:145–148.         [ Links ]

Thuiller, W., D. M. Richardson, P. Pysek, G. F. Midgley, G. O. Hughes and M. Rouget. 2005. Niche–based modelling as a tool for predicting the risk of alien plant invasions at a global scale. Global Change Biology 11:2234–2250.         [ Links ]

Waltari, E., R. J. Hijmans, A. T. Peterson, A. Nyári, S. L. Perkins and R. P. Guralnick. 2007. Locating Pleistocene refugia: comparing phylogeographic and ecological niche model predictions. PLoS ONE 2:e563.         [ Links ]

Warren, D., R. Glor and M. Turelli. 2008. Environmental niche equivalency versus conservatism: quantitative approaches to niche evolution. Evolution 62:2868–2883.         [ Links ]

Wiens, J. and C. Graham. 2005. Niche conservatism: integrating evolution, ecology and conservation biology. Annual Review of Ecology and Systematics 36:519–539.         [ Links ]