Estimating the biodiversity of a system with covariates and dependence structures

Entropy is widely used in ecological and environmental studies, with data often presenting complex interactions. Difficulties arise when the interest is to link such measure to available covariates or data dependence structures, as all existing approaches to entropy estimation assume independence. We focus on improving the estimation of the probabilities of the species in biodiversity studies, accounting for any data dependence and correlation. Estimating entropy is then straightforward and may be an informative index of a system’s biodiversity. An application is presented about the biodiversity of rainforest tree data.