Oecologia (2009) 160:817–825 DOI 10.1007/s00442-009-1348-3
C O M M U N I T Y E C O L O G Y - O RI G I N A L P A P E R
Spatial and environmental factors contributing to patterns in arboreal and terrestrial oribatid mite diversity across spatial scales Zoë Lindo · Neville N. Winchester
Received: 7 October 2008 / Accepted: 31 March 2009 / Published online: 2 May 2009 © Springer-Verlag 2009
Abstract Understanding the conditions under which species traits, species–environment relationships, and the spatial structure of the landscape interact to shape local communities requires quantifying the relative contributions of space and the environment on community composition. Using analogous sampling of arboreal and terrestrial oribatid mite communities across a large spatial scale in a temperate rainforest, we quantiWed the variation in oribatid mite community structure relating to environmental and spatial factors, and tested whether terrestrial and arboreal communities demonstrated a diVerence in their patterns of community composition based on the assumption of diVerences in dispersal potential. The expectation that terrestrial oribatid mite communities are spatially structured while arboreal communities are environmentally structured was supported by our analyses at the level of variation in beta diversity, but not by assessing beta diversity itself. We found that terrestrial oribatid mite communities with active, cursorial dispersal demonstrate spatial constraint consistent with reduced long-distance dispersal opportunities and high environmental dissimilarity among sites. Arboreal communities, which potentially disperse long distances via passive Communicated by Volkmar Wolters. Electronic supplementary material The online version of this article (doi:10.1007/s00442-009-1348-3) contains supplementary material, which is available to authorized users. Z. Lindo (&) Department of Biology, McGill University, 1205 Docteur PenWeld, Montreal, QC H3A 1B1, Canada e-mail:
[email protected];
[email protected] N. N. Winchester Department of Geography, University of Victoria, PO Box 3020, Victoria, BC V8W 3N5, Canada
aerial vectors, show a spatial signature associated with patterns in beta diversity and a correlation with environmental dissimilarities among sites. In the arboreal community, moisture content of the substrate, total tree height, and average sampled branch height were signiWcant factors explaining beta diversity patterns. For ground-dwelling species, predator abundance and soil type were important local determinants of community variability. Both communities showed clear spatial structuring, suggesting that dispersal limitation continues to inXuence community composition across multiple forest watershed locations. Our results provide evidence of dispersal-maintained diversity patterns in response to local environmental factors in arboreal and terrestrial communities. The relative importance of stochastic dispersal assembly may be dependent on strong deterministic eVects associated with micro-site and macro-site environmental variation, particularly across large spatial scales. Keywords Beta diversity · Dispersal · Canonical correspondence analysis · Mantel test · Oribatid mites
Introduction The relative importance of the environment and dispersal limitation in shaping the compositions of communities at diVerent spatial scales is currently a topic of active consideration (Van de Meutter et al. 2007; Van der Gucht et al. 2007; Vanschoenwinkel et al. 2007; Jones et al. 2008). Community structure at the local scale is viewed along a continuum of dynamics under the metacommunity concept (Leibold et al. 2004). Recent meta-analyses have addressed this issue at two levels of abstraction (sensu Legendre et al. 2005 and Tuomisto and Ruokolainen 2006): the variation in
123
818
community composition (beta diversity) (Cottenie 2005), and the variation in beta diversity across spatial scales (Soininen et al. 2007a). Both Cottenie (2005) and Soininen et al. (2007a) identiWed four potential determining characteristics of metacommunity type: organism size, sampling spatial scale (including extent and grain size), habitat type and properties of the environment, and organism dispersal type or ability. From these characteristics, Cottenie (2005) and Soininen et al. (2007a) found that dispersal type or ability was the most important factor determining metacommunity type, particularly with respect to diVerences between passive dispersers that may have high dispersal potential and active dispersers with constrained dispersal. Species dispersal mode (e.g., active vs. passive) is often related to the distance traveled while dispersing. Most dispersal events occur over short ranges, as dispersal success decreases with distance from the source, but long-distance dispersal is the underlying mechanism of rescue eVects (Brown and Kodric-Brown 1977), and an important parameter for rates of spread of invasive species (Clark 1998). For many taxa, such as seed- or spore-bearing plants and arthropods, the tracking of propagules or individuals is not possible (but see Flinn 2007), and dispersal may be characterized by largely stochastic long-distance dispersal events (Clark 1998; GriYn et al. 2002). Our model system is the oribatid mite (Acari: Oribatida) community, which can be found in arboreal and terrestrial habitats in temperate rainforests (Lindo and Winchester 2008). Oribatid mites are small (
