Semi-automated analysis of high-resolution aerial images to quantify docks in glacial lakes

Lake resources can be negatively affected by environmental stressors originating from multiple sources and different spatial scales. Shoreline development, in particular, can negatively affect lake resources through decline in habitat quality, physical disturbance, and impacts on fisheries. The development of remote sensing techniques that efficiently characterize shoreline development in a regional context could greatly improve management approaches for protecting and restoring lake resources. The goal of this study was to develop an approach using high-resolution aerial photographs to quantify and assess docks as indicators of shoreline development. First, we describe a dock analysis workflow that can be used to quantify the spatial extent of docks using aerial images. Our approach incorporates pixel-based classifiers with object-based techniques to effectively analyze high-resolution digital imagery. Second, we apply the analysis workflow to quantify docks for 4261 lakes managed by the Minnesota Department of Natural Resources. Overall accuracy of the analysis results was 98.4% (87.7% based on K^) after manual post-processing. The analysis workflow was also 74% more efficient than the time required for manual digitization of docks. These analyses have immediate relevance for resource planning in Minnesota, whereas the dock analysis workflow could be used to quantify shoreline development in other regions with comparable imagery. These data can also be used to better understand the effects of shoreline development on aquatic resources and to evaluate the effects of shoreline development relative to other stressors.