From particle flows in physics over the rhythm of commuter flows in cities to the graceful motion of schools of fish, investigating large-scale movement data allows us to understand complex coherences in almost all areas of society. Yet, analyzing spatio-temporal datasets can be challenging, as it is hard to model behaviors that one has no knowledge about in the first place. Given the unique, tightly interlocked combination of the dimensions of space and time, manual exploration of movements through observation is often without alternative to ensure the most complete gain of knowledge possible. This thesis aims to improve the visual analysis process by providing a novel visualization approach, making use of dimensionality reduction techniques to create a static, space-efficient visualization of multiple movers. Initially, the practical application in the field of Movement Ecology is considered, from which this thesis draws its real-world reference. Through a user study and analysis of existing work processes, research gaps and user needs are identified. Based on these insights, Visual Summaries for movement overviews are introduced at hand of the MotionRugs approach, which uses space partitioning and transformation techniques to provide a space-efficient representation of multiple moving entities. Specifically, raw movement data is transformed into a frame-by-frame representation. Afterward, a linearization strategy reduces the spatial dimensions to a one-dimensional sequence. By coloring these sequences according to feature values of the underlying movers, a static representation develops, allowing to identify feature developments and spatial events. As the linearization process impairs the accuracy of the visualization, a comprehensive exploration of the advantages and limitations of the proposed approach is provided through expert feedback, use cases, and both a qualitative and quantitative evaluation. Also, a solution to the reduction in spatial accuracy using two-dimensional color spaces is proposed. The main contribution of this thesis is a novel movement visualization technique,which is demonstrated using real-world data and expert-backed use cases. A comprehensive evaluation of all aspects of the technique is provided, and solutions for shortcomings are proposed. The thesis concludes with a discussion of advantages and limitations of Visual Summaries for movement data and provides starting points for future research opportunities in the field.