Search

According to Moore's Law, the cost of devices drops by half every two years. This ongoing price reduction has led to a surge in data collected via GPS sensors, accelerometers, and magnetometers. The increased availability of ways to collect movement data has significantly boosted our ability to understand and work with movement-related information. Movement data takes various forms, such as people navigating cities, cars traveling on highways, or animals roaming through forests. However, on its own, movement data lacks intrinsic meaning. It must be integrated with environmental, physical, and temporal contexts to gain a comprehensive understanding. Compared to their human counterparts, who move in urban environments with abundant semantic information, animals' movements often lack such semantics and context. Researchers or domain experts who deeply understand the environment grounded in extensive research data often provide contextual information. Domain expertise is an indispensable asset in initiating practical data analysis. Visualization is pivotal in fostering collaboration between domain scientists and data analysts. This collaboration empowers both parties to collectively contribute context and significance to the data. Moreover, visualization plays a crucial role in understanding movement, as seeing the data visually prompts deeper comprehension. It aids in understanding how movement interacts with the environment.However, visualization comes with its challenges; with the exponential increase in data, clutter has emerged as a significant issue in movement visualization. Dealing with large datasets can be overwhelming; the sheer number of data points poses a considerable challenge for interpretation. Moreover, errors in the data, often originating from sensor inaccuracies, further complicate the analysis. Designing tools with the end-user in mind is crucial to address these challenges effectively. User-centric design not only helps simplify the interpretation of vast data but also aids in managing and mitigating errors, ensuring that the insights drawn from the data are meaningful and accurate. Also, animal research demands a lot of specialized knowledge, and that's why tools have unique requirements, causing a gap between researchers and the people making these tools.During this thesis, we collaborated with researchers studying arboreal animals on BCI Island in Panama. Our aim was to tackle challenges like handling large datasets, loading contextual information, and finding the right spatial and temporal scales for analysis, all while making the visual tool easy to use. To make this happen, we created a collection of user-friendly exploratory and visual analytics features. Users can explore the island, adjust data spatially and temporally which can be used as inputs to analytical models, identify common movement patterns animal highways allowing for hypothesis verification of behavior patterns, and even download the dataset for in-depth analysis.