"HerdGPS-Preprocessor"-A Tool to Preprocess Herd Animal GPS Data; Applied to Evaluate Contact Structures in Loose-Housing Horses.
Authors: Salau Jennifer, Hildebrandt Frederik, Czycholl Irena, Krieter Joachim
Journal: Animals : an open access journal from MDPI
Summary
# Editorial Summary Precision livestock farming increasingly relies on GPS positioning to understand animal behaviour and social dynamics, but the practical challenges of large-scale, long-duration data collection—prohibitive hardware costs, sensor battery limitations, and repeated human handling during recharging—have hindered widespread adoption in equine research. Salau and colleagues addressed these obstacles by developing HerdGPS-Preprocessor, a software tool designed to manage the substantial raw data streams generated by continuous GPS tracking of multiple animals, automatically sorting and formatting positional information whilst simultaneously generating contact matrices that enable straightforward social network analysis. Applied to 40 horses housed in a loose-box facility over four weeks, the tool revealed dynamic changes in contact structure throughout the day and across different days, with notably denser social networks and larger subgroup formations (cliques) emerging in the hours immediately before pasture access became available. The ability to quantify social proximity patterns using standardised network parameters—density, diameter and clique configuration—offers equine professionals a practical means to assess welfare indicators and social stability, particularly in boarding and rehabilitation settings where understanding herd dynamics directly influences management decisions and individual horse wellbeing. This toolkit substantially reduces the technical and logistical barriers to implementing GPS-based behavioural monitoring, making population-level social analysis increasingly feasible for practitioners seeking objective data on grazing group behaviour and social preference hierarchies.
Read the full abstract on PubMed
Practical Takeaways
- •This tool provides a practical solution for large-scale GPS monitoring of multiple horses without constant sensor maintenance disruption, enabling better understanding of natural herd movement patterns and social dynamics
- •Network analysis of horse contacts reveals time-dependent social structures that may reflect anticipation of feeding or pasture access opportunities
- •Facility managers can use this approach to optimize timing of pasture access and feeding schedules based on observed natural grouping patterns
Key Findings
- •HerdGPS-Preprocessor software successfully processed continuous GPS data from 40 horses over one month, converting raw positional data into formatted contact lists for network analysis
- •Contact networks showed significant hourly variation, with denser networks and larger cliques observed in hours preceding access to additional pasture
- •Network density, diameter and clique structure differed substantially between different times of day and different days of the week