Hidden influence of fire on locally rare and cryptic reptile species.

Fire responses in animal communities have been well studied, but little is known about the effects of fire on rare species due to limitations in data and modeling frameworks. Using a suite of methods to account for rarity, dominance, and incomplete sampling, we aimed to determine how post-fire succession influenced locally rare and cryptic reptile species in semiarid mallee woodlands in southern Australia. Contemporary fires in this system are typically high-intensity wildfires or prescribed fires, which raze aboveground vegetation. Reptiles were sampled from pitfall traps over two summer seasons (1400 trap nights) at 14 sites spanning three fire history categories based on time since the last fire: recently burnt (2-3 years post fire), medium (7-9 years), and long unburnt (42-48 years). The data comprised 2138 individual reptiles from 41 species. The effects of fire history were evident in diversity metrics where proportional abundances were weighted toward rare species (Shannon's Diversity, Fisher's α), but not in those emphasizing common species (Simpson's Diversity, Berger-Parker Index) or species richness. There were no effects of fire history on the richness or abundance of locally rare species when examined directly. However, fire effects on rare species diversity were detectable in one sub-assemblage including 20 species with fewer than eight total observations. This result indicated higher diversity in unburnt habitat; an effect that was masked when numerically dominant species were included in the analysis. When accounting for incomplete sampling using rarefaction and extrapolation, unburnt habitat was more diverse, accumulated species more quickly, and required a greater sampling effort to obtain sample coverage comparable to recently burnt and medium habitat. Overall, common and abundant reptile species appeared to dominate recently burnt and mid-succession habitats, while reptile communities in long unburnt habitat were more diverse because they had more rare or cryptic species. The data suggest that responses of rare species to fire history are harder to detect than common, dominant species which are often favored by disturbance (and scientific research). Fire management which maintains some early- and mid-successional habitat, while strategically conserving unburnt habitat, should benefit rare and cryptic reptile species in this system.