A comparison of hull methods for estimating species ranges and richness maps

Abstract

Background: The combined analysis of species ranges allows the detection of overall species richness patterns. As such, the reliability of richness maps is directly related to the accuracy of individual range estimates.Aims: We compared how species ranges and richness patterns are influenced according to the use of different methods (convex hull and alpha hull) and evaluated if differences in range estimates are related to the Wallacean shortfall.Methods: We used occurrence records of 386 species from the tribe Bignonieae (Bignoniaceae) to build richness maps by stacking species ranges, using the convex hull and alpha hull methods. We evaluated the behaviour of the richness estimates in relation to other variables (i.e., other richness measures, environmental and spatial variables). We correlated the differences in range estimates to proxies of Wallacean shortfall (i.e., species description date and variation of pseudo-ranges generated by jack-knife).Results: Convex hull estimated larger ranges than alpha hull. The Wallacean shortfall affected range estimates equally regardless of the hull method used. Both hull methods generated similar richness maps. Richness estimates showed similar correlation patterns in relation to other variables regardless of the hull method used.Conclusions: Despite the slight differences in range estimates, both hull methods can be used in equivalence to map species richness.