Using the Rao's Q diversity index as an indicator of protected area effectiveness in conserving biodiversity

Abstract

Developing innovative monitoring systems for biodiversity outcomes in protected areas (PAs) are important to enable effective adaptive management. Here we show how to quantitatively detect and monitor temporal and spatial patterns in environmental heterogeneity, an important indicator of ecological integrity and biodiversity patterns. We used a 28-year time series (1991–2018) generated from freely available Landsat satellite imagery and Rao's quadratic index to calculate relative heterogeneity and trends in heterogeneity for 41 PAs across South Africa. We selected PAs where mega-herbivore assemblages were similar and where management objectives were broadly aligned with South African legislation to protect biodiversity. There was a three-fold difference in heterogeneity among PAs, mainly linked to variation in topography and rainfall. Heterogeneity decreased in 12 (29%), increased in three (7%) and remained stable in 26 (64%) PA's. These trends were not explained by overall heterogeneity, PA size, or rainfall – i.e., PAs that were smaller, drier, and more homogenous were not more likely to show decreasing trends than PAs that were larger, wetter, and generally more heterogenous. Rather trends in heterogeneity are likely the result of interactions between regional process (e.g., rainfall) and local factors such fire regimes, megaherbivore densities, and park management. The framework presented here can be extended to include every PA nationally, or even globally, and the data product fully automated. This presents an opportunity for conservation management to incorporate this important biodiversity indicator in PA monitoring programs as well as other large-scale ecological monitoring initiatives.