Making the connection: combining habitat suitability and landscape connectivity to understand species distribution in an agricultural landscape

• Published in: Landscape ecology Vol. 36; no. 10; pp. 2795 - 2809
• Main Authors: Youngquist, Melissa B., Boone, Michelle D.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.10.2021; Springer Nature B.V
• Subjects: Agricultural land; Amphibians; Biodiversity; Biomedical and Life Sciences; Conservation; Dispersal; Ecology; Environmental Management; Frogs; Geographical distribution; Habitats; Immigrants; Land cover; Land use; Landscape Ecology; Landscape/Regional and Urban Planning; Life Sciences; Metapopulations; Mosaics; Nature Conservation; Ponds; Precipitation; Regression analysis; Regression models; Research Article; Species; Sustainable Development; Graphab; Species distribution model; MaxEnt; Landscape network
• ISSN: 0921-2973; 1572-9761
• DOI: 10.1007/s10980-021-01295-7

Context The current biodiversity crisis has intensified the need to predict species responses to landscape modification and has renewed attention on the fundamental question of what influences the dynamics of species distributions. Landscape composition can affect two main components that dictate distributions: habitat suitability and habitat connectivity. Elucidating the relative importance of these factors and associated landscape features can help prioritize management action for species conservation. Objectives Our objective was to use species distribution models and network-based landscape connectivity models to understand which landscape factors were most predictive of the distribution of an anuran, Blanchard’s cricket frog ( Acris blanchardi ), in an agriculturally-dominated landscape. Methods We conducted our study in Ohio, USA, near the edge of the cricket frog’s contracting range. To obtain a current assessment of cricket frog distribution, we surveyed 367 pond and stream locations across three North–South transects. We then tested seven regression models, combining habitat suitability and landscape connectivity metrics, to determine which factors best predicted cricket frog presence. Results We detected cricket frogs in 24% of surveyed locations and they were more likely to occupy pond sites than stream sites. Cricket frog presence was best predicted by models with habitat suitability and the number of interconnected habitat patches. We found that, while there was high variation in habitat suitability across the study area, landscape connectivity was relatively uniform where we surveyed. Conclusions Agricultural landscapes around the world are often mosaics of land cover types, which may functionally provide connectivity for some species. In such areas, conservation management should focus on preserving and restoring regions of highly suitable habitat. This focus may be particularly relevant for species that do not appear to be dispersal limited and, therefore, able to maintain metapopulation dynamics.