LOCAL AND REGIONAL DIVERSITY IN A PATCHY LANDSCAPE: NATIVE, ALIEN, AND ENDEMIC HERBS ON SERPENTINE

Susan Harrison

Ecology · 1999 · 198 citations · 14 references

Concepts

Abstract

Metacommunity models predict that communities in patchy environments will show different patterns of diversity than communities in continuous habitats. The species richness of herbs was measured in 24 small patches (0.5–3 ha) of serpentine soil and 24 equivalently spaced sets of sampling sites on four large areas (>5 km2) of serpentine soil in the North Coast Ranges of California. Patchy and continuous sites were compared with respect to local diversity (species richness within 5 × 50 m transects), regional diversity (species richness at all 24 sites combined), and among-site differentiation (average proportion of species unshared between any one site and the other 23 sites of its kind). For all herb species together, regional diversity and average local diversity were higher on small patches, mainly because of a significantly higher number of alien species. For the subset (14%) of herb species that were endemic to serpentine, regional diversity was similar on patchy and continuous sites, but it was partitioned differently; local diversity was lower, and among-site differentiation was higher on small patches. Herb diversity also showed significant gradients with soil calcium levels, and these effects were not independent of site type (patchy or continuous). Soil calcium was correlated positively with the number of alien species on small patches, and negatively with the number of serpentine-endemic species on continuous sites. Also, total herb diversity was negatively correlated with elevation on continuous sites. However, patchy and continuous sites did not differ in their average calcium levels or elevation. These results agree with theory predicting that, relative to communities in widespread habitats, local communities in patchy habitats will contain more habitat generalist and fewer habitat specialist species. However, the results also suggest that the effects of habitat geometry interact with those of habitat quality in this system.

References

14