Event detection from novel data sources: Leveraging satellite imagery alongside GPS traces

TL;DR

The paper tackles rapid detection of breaking events by fusing privacy-preserving mobile GPS traces with satellite imagery to enable real-time or historical event inference. It presents a practice-oriented data fusion framework that couples mobility-derived indicators (e.g., radius of gyration, stays, visits) with visual and spectral changes captured in satellite data through an image-aware, rule-based workflow. Key contributions include the dual-module methodology, a tornado case study in Muskogee, OK, and an open-source implementation for reproducibility. This approach has practical implications for disaster response, search-and-rescue, and conflict monitoring, especially in rural or connectivity-challenged environments, leveraging commercially available data streams and scalable geospatial processing. The methodology is underpinned by a satellite-imagery utility metric $u$ and an anomaly-detection mechanism using a $Z$-score threshold, and it uses NDVI and grayscale changes to corroborate observed mobility signals with environmental impact.

Abstract

Rapid identification and response to breaking events, particularly those that pose a threat to human life such as natural disasters or conflicts, is of paramount importance. The prevalence of mobile devices and the ubiquity of network connectivity has generated a massive amount of temporally- and spatially-stamped data. Numerous studies have used mobile data to derive individual human mobility patterns for various applications. Similarly, the increasing number of orbital satellites has made it easier to gather high-resolution images capturing a snapshot of a geographical area in sub-daily temporal frequency. We propose a novel data fusion methodology integrating satellite imagery with privacy-enhanced mobile data to augment the event inference task, whether in real-time or historical. In the absence of boots on the ground, mobile data is able to give an approximation of human mobility, proximity to one another, and the built environment. On the other hand, satellite imagery can provide visual information on physical changes to the built and natural environment. The expected use cases for our methodology include small-scale disaster detection (i.e., tornadoes, wildfires, and floods) in rural regions, search and rescue operation augmentation for lost hikers in remote wilderness areas, and identification of active conflict areas and population displacement in war-torn states. Our implementation is open-source on GitHub: https://github.com/ekinugurel/SatMobFusion.

Figures (6)

• Figure 1: Images before and after the Aru Glacier collapse on 17 July 2016. Images are from Sentinel-2 on 18 June 2016 (a) and 21 July 2016 (b) with 10 m resolution. From tian2017twoglaciers
• Figure 2: Methodological flowchart; the top bins describe the order of processes employed in this paper, while the bottom outlines key geospatial analysis metrics to carry out the inference task
• Figure 3: Usage decision tree for the proposed method
• Figure 4: EDA of mobility metrics in the before, during, and after periods
• Figure 5: Visits per hour to the ROI in the before, during, and after periods sorted by hour of day