Climatic Niches and the Spatial Distribution of Species

The observation of ecological properties of species and their areas of distribution being related is not new (Grinnell 1917; James et al. 1984), but the increasing availability of information on the variation of environmental parameters in geographic

Department of Herpetology, Zoologisches Forschungsmuseum Alexander Koenig, D-53113 Bonn, Germany

Department of Biogeography, University of Trier, D-54286 Trier, Germany e-mail: [email protected]

J. Dambach

Section of Lower Arthropods, Zoologisches Forschungsmuseum Alexander Koenig, D-53113 Bonn, Germany

J.C. Habel and T. Assmann (eds.), Relict Species: Phylogeography and Conservation Biology, 373 DOI 10.1007/978-3-540-92160-8_22, © Springer-Verlag Berlin Heidelberg 2010

space, species distribution data, and computation capacities during the last decade now allow large scale assessments of relationships between distributions observed and explanatory parameters. Relationships can be assessed by calculating 'environmental' or 'ecological' niches and their subsequent projection into geographic space (Guisan and Zimmermann 2000). Here, GIS-based environmental data offer huge opportunities to assess variations in environmental factors within the species ranges, especially when combined with spatial modeling techniques (Kozak et al. 2008; Waltari et al. 2007). Such techniques were proposed as a useful supplementary tool despite long time/term established techniques for the identification of refuges and potential migration pathways (Waltari et al. 2007).

Model techniques can be classified into two different groups: (1) mechanistic models, which predict the potential distribution of a species based on its physiological tolerances measured in laboratory experiments and (2) spatial model techniques, which derive from the distribution model based on statistical relationships between distribution patterns observed and environmental parameters. In the latter case, Climatic Envelope Models (CEMs) use exclusively climatic variables as predictors whereby a wider range of variables is used in Ecological Niche Models (ENMs) (e.g. soil and vegetation layers or remote sensing data). The development of mechanistic models is just at the beginning (Kearney and Porter 2004; Kearney et al. 2008), but CEMs and ENMs have been applied to predict species' potential distributions (PDs) under current, past, and future climate scenarios for some time now (e.g. Araujo et al. 2004; Araujo and Guisan 2006; Heikkinen et al. 2006; Hijmans and Graham 2006; Pearman et al. 2008; Waltari et al. 2007), invasive species biology (e.g. Peterson 2003; Peterson and Vieglais 2001; Rödder 2009; Rödder et al. 2008), conservation priority setting (e.g. Araujo et al. 2004; Kremen et al. 2008), and ecology and evolutionary biology (e.g. Graham et al. 2004; Kozak et al. 2008; Peterson et al. 1999). CEMs were especially suggested to be useful for identification of potential Pleistocene refugia with high accuracy (Peterson and Nyari 2008; Waltari et al. 2007) and are therefore discussed below in detail.

In CEM approaches, the climate envelope is understood as a part of a species' fundamental or realized niche depending on variables selected and assumptions made (Soberon 2007; Soberon and Peterson 2005). As defined by Hutchinson (1957; 1978) and later extended by Soberon and Peterson (2005), a species' fundamental niche represents the complete set of environmental conditions under which a species can persist, i.e. under which its fitness is greater than or equal to one in the absence of competitors or predators. It's realized niche in environmental space (= realized distribution in geographic space) is a subset of the fundamental niche considering dispersal limitations and biotic interactions, such as food availability, competition, or interactions with pathogens (Fig. 1). Niche variables can be subdivided concerning specific classes regarding the spatial extent in which they operate and if competition may play a role or not. The Grinnellian class is defined by fundamentally non-interactive variables (e.g. climate) (Grinnell 1917) whereby the Eltonian class focuses on biotic interactions and resource-consumer dynamics (Elton 1927). The former operates on a coarse scale and is the main subject in CEM approaches, whereby the latter can principally be measured only at local scales and is commonly not addressed in CEM approaches (Soberon 2007).

Accessibility Biotic interaction

Fig. 1 Relationships between abiotic (= fundamental) niche, biotic interaction and accessibility after Hutchinson (1957) as modified by Soberon and Peterson (2005). The potential distribution is a subset of the abiotic niche considering biotic interactions, whereby the realized distribution is a subset of the potential distribution considering accessibility. Dots represent species records

Accessibility Biotic interaction

Fig. 1 Relationships between abiotic (= fundamental) niche, biotic interaction and accessibility after Hutchinson (1957) as modified by Soberon and Peterson (2005). The potential distribution is a subset of the abiotic niche considering biotic interactions, whereby the realized distribution is a subset of the potential distribution considering accessibility. Dots represent species records

Was this article helpful?

0 0

Post a comment