Modelling biodiversity distribution in agricultural landscapes to support ecological network planning

Mossman, HL, Panter, CJ and Dolman, PM (2015) Modelling biodiversity distribution in agricultural landscapes to support ecological network planning. Landscape and Urban Planning, 141. ISSN 1872-6062

Preview

Available under License In Copyright. Download (730kB) | Preview

Abstract

Strategic approaches to biodiversity conservation increasingly emphasise the restoration of ecological connectivity at landscape scales. However, understanding where these connecting elements should be placed in the landscape is critical if they are to provide both value for money and for biodiversity. For such planning to be effective, it is necessary to have information of the distributions of multiple taxa, however, this is of poor quality for many taxa. We show that sparse, non-systematically collected biological records can be modelled using readily available environmental variables to meaningfully predict potential biodiversity richness, including rare and threatened species, across a landscape. Using a large database of ad-hoc biological records (50 501 records of 502 species) we modelled the richness of wetland biodiversity across the Fens, a formerly extensive wetland, now agricultural landscape in eastern England. We used these models to predict those parts of the agricultural ditch network of greatest potential conservation value and compared this to current strategic network planning. Odonata distribution differed to that of other groups, indicating that single taxon groups may not be effective proxies for other priority biodiversity. Our results challenged previous assumptions that river channels should comprise the main connecting elements in the Fens region. Rather, areas of high ditch density close to a main river are likely to be of greater value and should be targeted for enhancement. This approach can be adopted elsewhere in order to improve the evidence-base for strategic networks plans, increasing their value for money.