Local and regional zooplankton species richness: A scale-independent test for saturation

Assemblages of coexisting species are formed by immigration from a regional pool of colonists and local interactions among species and with the physic al environment. Theory suggests that the shape of the relationship between regional and local species richness may indicate the relative roles of disp ersal and local interactions in limiting local diversity. Here we examine p atterns of regional and local species richness in freshwater crustacean zoo plankton to test whether linear (suggesting dispersal limitation) or curvil inear (suggesting saturation, via strong local control) functions best fit the data, Local richness appeared saturated when regions of different spati al extents were included on the same graph, However, this pattern was influ enced by differences in scale among surveys. We corrected for the effects o f regional scale by plotting mean local richness against the residuals of t he species-landscape area relations. Controlling for the extent of the regi onal scale produced much more linear patterns, suggesting strong dispersal limitation, We present a simple graphical model to explain how variation am ong surveys in the geographic size of regions can produce apparent saturati on of local diversity even if the underlying pattern of local and regional richness is linear. We also compare the predictive power of residual region al richness on local richness with that of several local features in a mult iple regression model. Local richness exhibits strong relationships with bo th residual regional richness and pH. We argue that the relative strengths of local and regional processes depend on the definition of the regional scale. A variety of evidence suggests th at local processes play a major role in generating differences in zooplankt on diversity among lakes within a biogeographic region. Evidence for the im portance of dispersal limitation comes largely from comparisons of lakes ac ross very large scales. Our analysis suggests that linear patterns of local and regional diversity are not incompatible with strong local interactions .