Longitudinal species turnover rates are predictable and should guide location of sampling sites for South African river surveys to assess aquatic biodiversity

Abstract

Representative samples of biota collected relative to environmental gradients are important for measuring present distributions and predicting shifts in distribution of taxa in response to climate change or reduced river connectivity. Based on river ecology theory and established measures of species diversity, we present a method to identify suitable river segments for sampling and monitoring changes in taxon diversity. Alpha and beta diversities of selected aquatic macroinvertebrates were assessed in seven South African rivers. Data were drawn from historical and field samples and represented longitudinal species patterns down longitudinal river axes.
Representative sampling sites were identified using a logistic regression model to predict the probability of site pairs that were more than 50% similar as a function of up-/downstream distance. Alpha diversities peaked in the upper third of river lengths; beta diversities showed predictable exponential decay rates down river axes up to and excluding the start of estuarine conditions. Application of the model to a 370-km long river indicated that 14 sites should be selected for sampling to capture overall biodiversity patterns. Additional factors, such as confluences tributaries, which influence alpha diversity at sites, are identified and incorporated into site selection.

Keywords: alpha and beta diversity, aquatic macroinvertebrates, climate change, logistic regression, probability, representivity, sampling design