Swimming in Big Data: Using Network Analysis to Predict the Spread of Invasive Fishes

River-lake networks create a series of analytical challenges for studying invasive species spread with climate change. Although it is relatively easy to conceptualize freshwater networks as connected and flow directed systems represented by a series of lines (rivers), interspersed by nodes (e.g., lakes, confluences), there are only a handful of studies that have successfully analysed how hydrologic connectivity affects fish species movement and diversity; or how hydrologic connectivity is affected by barriers such as dams. Given the importance of freshwater networks for the maintenance of services like waterpower generation, water extraction, and fisheries, it is remarkable that most of these studies have only been performed at single watershed scales. We examine whether some network configurations in the Great Lakes Basin (GLB) of Ontario are more resistant to invasion by warm and cool-water fishes. Our approach uses circuit theory to model the relative accessibility of inland river-lake networks given projected climate change models. We then examine how development of proposed dams might modify access to suitable thermal habitat. We showcase the challenges, strengths and limitations of our approach using a large dataset and intensive analytical techniques, suggesting that ‘Big Data Ecology’ can inform strategic watershed management across large spatial scales.