Nash, K.L., C. R. Allen, C. Barichievy, M. Nyström, S. Sundstrom, and N.A.J. Graham. Habitat structure and body size distributions: ecosystem appropriate metrics and methods. Oikos 123:971-983.
1. Body size is linked to a range of ecological and life history traits, and thus provides a useful proxy for understanding individual, population and community level attributes and processes. Habitat structure across multiple scales has been proposed as a key driver of body size distributions for associated communities. Thus, understanding the relationship between habitat and body size is fundamental to developing predictions regarding the influence of future disturbances and habitat degradation on the distribution of body size within communities. 2.To date much of the work assessing the relationship between cross-scale habitat structure and body size distributions has focused on terrestrial taxa with determinate growth, and has primarily analysed discontinuities in species mean size distributions (SMSDs). The suitability of this approach for marine taxa, exhibiting indeterminate growth, has yet to be determined. 3.Here we provide a cross-ecosystem comparison of bird and fish body mass distributions, evaluating three size distribution indices: SMSDs, local size density relationships (LSDRs), and individual size density relationships (ISDs) and two types of analysis (discontinuity vs. modality), to highlight their respective suitability for exploring habitat-body size relationships in different taxa and ecosystems. 4.All three indices of body size distribution are useful for examining the relationship between cross-scale habitat structure and body size for species with determinate growth, such as birds. In contrast, for species with indeterminate growth, such as fish, the relationship between habitat structure and body size may be masked when using summary metrics, and thus ISDs are more useful. Furthermore, evaluating modality and abundance patterns in size distributions provides a more powerful way of assessing habitat impacts, than assessing discontinuity patterns. 5.It is critical to consider the life-history and ecology of taxa when selecting analytical approaches for evaluating body size distributions. Inappropriate use of summary metrics for species with indeterminate growth may hide the influence of drivers and provide misleading research outcomes. Critically, although ISDs are traditionally used to study aquatic systems, they present a common currency for comparing body size distributions across ecosystems.