Indicators of biodiversity in an intensively cultivated and heavily human modified landscape

Nowadays, the loss of biodiversity in agroecosystems due to the intensification of farming practices is happening very fast, and therefore, stopping or slowing it down should be a priority for conservation. To detect changes in these environmental contexts, one approach contemplates focusing on a limited set of indicator species that can alert us to ongoing changes in progress. In this research, we aimed to measure the biodiversity of vertebrates using a multi-taxa approach in an intensively cultivated and highly inhabited area located in northern Italy. We investigated the relationships between biodiversity and environmental characteristics and we identified the taxonomic groups that can be used as indicators of biodiversity. Data collection was carried out in 2016 with different methods depending on the taxonomic group, in 131 sampling units chosen using a Tessellation Stratified Sampling. Then we calculated for each sampling unit a standardized Biodiversity Index, which was related to environmental variables concerning the land use and the landscape configuration using Multiple Linear Regression Analysis and Information-Theoretic approach. We used correlation analyses and the Indicator Species Analysis (IndVal) to identify the taxonomic groups and species that can be used as indicators of biodiversity. Biodiversity was positively related to the number of patches of natural vegetation, whereas it was negatively affected by the number of patches of artificial surfaces and by habitat diversity. Our findings agree with those obtained by many other researchers, which pointed out that agroecosystems provide adequate shelters, suitable foraging habitats and nesting sites. The negative effect of habitat diversity was explained by the area-heterogeneity trade-off. Therefore, sites with high heterogeneity will not contain enough cover of residual natural vegetation, essential to maintain high biodiversity, because increasing compositional heterogeneity within a fixed area simultaneously reduces the surface of each cover type. The analyses showed that birds and reptiles might be used as biodiversity indicators of vertebrates. Eurasian Magpie and Green Whip Snake, both generalist species, were associated with sites of low biodiversity, whereas seven birds, both generalists and farmland specialists, were associated with sites of medium biodiversity. In high biodiversity sites there were not indicator species. To conclude, in less natural environments, such as urban and agricultural landscapes, a combination of specialist and generalist indicator species seems adequate to monitor biodiversity changes. Our findings increase the knowledge of these very dynamic ecosystems, being important both to plan strategies for biodiversity conservation and to guarantee ecosystems services useful for humans.