Revisiting the pyrodiversity–biodiversity hypothesis: long‐term fire regimes and the structure of ant communities in a Neotropical savanna hotspot

Abstract

Summary: The idea that variable fire regimes (pyrodiversity) may increase habitat heterogeneity and, consequently, increase biodiversity at the landscape level is a relatively old one. However, this idea has been recently challenged by studies showing that the biota of fire‐prone ecosystems is so resilient to fire that variation in fire regime has little ecological consequences. We evaluated how communities of ants – a dominant faunal group – differ with variation in the frequency and seasonal timing of fire in Brazilian Cerrado (savanna), a global biodiversity hotspot. We compared the number and composition of ant species foraging on the ground and in trees among experimental treatments with three fire frequencies (unburned for 37 years and burned every 2 or 4 years for the past 16 years) and three seasonal timings (early, mid and late dry season) of fire with a biennial frequency. There were no significant differences in ant species richness in trees, and there were no significant differences in the richness of ground‐dwelling ants among plots burned at different times of the dry season. However, unburned plots had fewer ground‐dwelling ant species than those burned at 2 or 4 year intervals. We detected marked differences in species composition across regimes varying in both the frequency and the timing of fire, which were related to fire‐induced changes in vegetation structure. Over 40% of the overall ant species richness recorded in our study site was attributed to differences in species composition among fire treatments. Synthesis and applications. Our study shows that Cerrado ant diversity at the landscape scale is promoted by a diversity of fire regimes, and therefore supports the validity of the pyrodiversity–biodiversity hypothesis for the Cerrado biome. Our findings suggest that the current policy of fire suppression in Cerrado protected areas has a negative effect on overall biodiversity and therefore should be reconsidered.