MEPS 296:13-29 (2005)  -  doi:10.3354/meps296013

ABSTRACT: Increasing evidence indicates that spatial and temporal patterns in ecological systems are not independent of the scale of measurement. In this study we used a hierarchical sampling design to examine spatial patterns in assemblages of algae and invertebrates in midshore and lowshore habitats of rocky coasts in the Mediterranean Sea, increasing the range of scales usually covered by this type of study in marine habitats. To put our results in a broader context, we also conducted a review of the literature, targeting studies that explicitly investigated spatial patterns with hierarchical designs. We addressed 2 main questions: (1) To what extent does small-scale variability contribute to large-scale patterns of variation? (2) Is there a pattern of variability that can be generalised across species and habitats? The review of the literature indicated that hierarchical analyses of spatial pattern have been limited to a narrow range of habitats and taxa and that very few studies have addressed regional scales of variation (1000s of kilometers). The available data, however, did identify a general pattern: variability was larger at small spatial scales (meters) in almost all habitats, whereas variation over larger spatial scales (10s to 100s of kilometers) depended on the specific habitat and taxa examined. In our case study, we obtained measures of spatial variability through the use of 2 alternative methods: hierarchical nested design and independent estimate of spatial variance between pairs of quadrats. Both approaches provided further support to the patterns displayed by the analysis of the literature. Most response variables exhibited large variation over small spatial scales, while the significance of mid- to large-scale variability differed between midshore and lowshore habitats and among taxa. Although a proper understanding of large-scale patterns will require additional comparisons across wide geographical areas, small-scale variability emerges as a general property of benthic assemblages in marine coastal habitats. We suggest that this pattern is common to a wide range of natural systems where assemblages are influenced by complex sets of physical and biological processes like those operating in the marine environment.

KEY WORDS: Spatial scale · Patchiness · Hierarchical ANOVA · Spatial variance · Rocky shores · Algae · Invertebrates