A conditional probability index to quantify the amplitude and the direction of spatiotemporal changes in communities

Abstract

AbstractMonitoring and mapping species diversity using indicators can allow the detection of changes in communities. Conclusions regarding these changes greatly depend on the choice of indicator. Here, we propose a new metric, the distance biochange index (DBCI), that enables the characterization and quantification of both the level and direction of change in biological communities relative to a given reference state. The proposed metric uses conditional probabilities to assess the probability of observing a complete change in a given community and can be decomposed into four conditional probabilities of change: no change, complete change in species composition only, complete change in species richness only, and complete change in both species composition and richness. In this study, we compared the properties of DBCI and BCI, a similarity version of DBCI, with those of other widely used indices. We also proposed a new approach, based on the use of partial derivatives, to assess the sensitivity of six similarity indices over a wide range of contrasting scenarios of change. Finally, we extended the application of DBCI to a simulated case study of the predicted evolution of suitable habitats for 20 species under climate change. Results from this simulation demonstrated that DBCI provides an accurate assessment of the level and direction of change. Results also show that DBCI can be used to reflect the effect of ecological gradients on species composition and species richness in biological communities.