Landscape metrics as biodiversity indicators for plants, insects and vertebrates at multiple scales

Abstract

Landscape metrics are widely used to investigate spatial structure and pattern of landscapes. Numerous metrics are available, yet only little empirical research examined their indicator value for species richness pattern. In this study we explored the relations of 52 standard landscape metrics with the species richness patterns of six taxa (woody plants, orchids, Orthoptera, amphibians, reptiles, and birds) and overall biodiversity, taking as a case study a Mediterranean forest landscape - Dadia National Park in Greece. We computed landscape structure variables for circular areas of five different extents (hereafter scales) around the sampling plots. Additionally, we tested the performance of sets of three landscape structure variables as predictors of species richness, comparing sets composed by different methods such as expert knowledge, ordination methods, decision trees and random choice. We also evaluated optimal sets, testing all possible combinations of three variables. Landscape metrics were good indicators of overall biodiversity, and of the species richness of woody plants, Orthoptera and reptiles. Metrics quantifying patch shape, proximity, texture and diversity resulted in numerous significant univariate models, while metrics describing patch area, similarity and edge rarely contributed to significant models. Scale affected the performance of the metrics. Woody plants, Orthoptera and birds were usually better predicted at smaller scales, and reptiles frequently at larger scales. Regarding the different methods of composing sets, optimal sets always performed significantly better than all other methods. Among these, expert knowledge performed even slightly worse than random, while the statistical methods performed slightly better. The revealed pattern of relations and performances will be useful to understand landscape structure as driver and indicator of biodiversity, and to improve management decisions in Mediterranean forests and other mosaic-landscapes.