Conservation organizations must locate populations of endangered species in order to monitor their welfare. Accurate endangered species surveys are an important part of site assessments used when acquiring development permits. However, traditional methods of surveying endangered species locations are time and resource intensive and are subject to an individual’s interpretation. Additionally, habitat studies often lack a geospatial analysis component. Greater efficiency in the endangered species mapping can be achieved by analyzing its’ habitat characteristics and creating a probabilistic map of the species locations. We have developed a workflow using GRASS GIS machine learning algorithms to analyze environmental and high resolution lidar data. Variables such as climate, slope, aspect, hydrology, soil and landcover conditions favorable for Virginia Spiraea plants were examined based on its’ known locations in North Carolina and used to derive probabilistic maps of plant locations. We then demonstrate how the resulting map will be used to prepare targeted UAS surveys to efficiently locate individual Virginia Spirea plants at ultra-high resolutions. The developed methodology, based on the open source geospatial software stack, can be further expanded to include other species.
- About Us