Unmanned ground vehicles (UGVs) are used increasingly often in military operations and can carry out tasks with varying degrees of autonomy. However, currently there exists no established method for tactical terrain analysis to support military decision-making for autonomous UGVs. Military planning requires adequate tactical understanding of the mission environment, as well as analysis and optimization techniques to generate a course of action. This contribution presents a novel approach to extracting tactical geospatial information from terrain data and shows how this information can be used for planning of autonomous UGV missions. First, we generate an abstract representation of the mission environment from high-fidelity terrain data. Then terrain features such as elevation, vegetation density, and soil composition are combined to calculate tactical terrain information, such as line of sight, terrain traversability, and concealment. Using this tactical information, optimal observation points for given enemy areas and routes towards these points are derived. In addition, we show how different evaluation criteria influence the best course of action as proposed by the automatic tactical terrain analysis. The research contributes to the development of autonomous UGV systems by bridging the gap between terrain analysis and tactical analysis in the context of planning and decision-making for autonomous systems, enabling UGVs to become more self-reliant.