Effects of repeated burning on species richness in a Florida pine savanna: A test of the intermediate disturbance hypothesis

We studied the effect of burning frequency on the density and species richness of understory flowering stems in a Florida sandhill. Flowering stems were censused weekly for 54 weeks in six sites that had been burned one to six times in the previous 16 years. We concurrently measured overstory characteristics such as species composition, density and basal area. We used maximum likelihood and Akaike's Information Criterion to compare linear, quadratic, saturating, and null models of community response to repeating burning. We did not find a relationship between species richness, diversity or flowering stem density and fire frequency. Tree density was related to fire frequency and may represent an indirect pathway for fire effects on understory characteristics. While we found no support for the Intermediate Disturbance Hypothesis, an analysis of our experimental design indicated that we had low statistical power. We develop the hypothesis that a saturating model of response to fire best describes understory species richness in our system. We test this hypothesis using the most extensive published fire data set we are aware of and find support for a saturating model.