Water and substrate control of cliff communities: patterns among species and phyla

Aims Cliff communities tend to be compositionally simple but taxonomically diverse at the phylum/division level. As a result, they provide a setting for examining the responses of distinct phyla (e.g. Anthophyta, Bryophyta and lichens), species within these phyla and communities comprised of these phyla, to the same environmental gradients. Our aims were both descriptive and hypothesis‐driven. We sought to identify major community shifts across water–substrate gradients in a poorly studied system (inland North American cliffs), while considering extant hypotheses concerning the relationships of organisms to environments at different taxonomic scales. Location Sub‐alpine cliffs in northeastern Yellowstone National Park. Methods Cliff faces on two mountains were blocked at nine waterfalls. At each block, community composition was recorded from random samples obtained for three levels of water availability (xeric, mesic and hygric) on two substrates (limestone and andesite). Patterns of cover and richness were compared using Friedman's Rank F ‐tests. Community patterns were discerned using permutation MANOVAs, ordinations, and cluster analyses. Results Phyletic composition varied in response to water availability. In general, lichens dominated dry sites, while vascular plants dominated wet sites. Anthophyta was the most narrowly specialized phylum, and Bryophyta was the most generalized. Beta diversity was higher at drier sites, due to the dominance of the diverse lichen group. As a result, communities on lime and andesitic substrates were indistinguishable in hygric environments, but significantly different at drier sites. This may have occurred because, while water is limiting for neither substrate in hygric environments, andesites are more likely to absorb and store atmospheric water in drier environments. Indicator species analyses identified 27 distinct species distinguishing water and substrate conditions, or their interactions. Conclusions Our experimental design uniquely accounted for extraneous variables in its examination of water and substrate effects on cliff communities. Water strongly controlled community composition both among and within phyla through both direct effects and interactions with substrate.