Ground Cameras with ADS-B Data for High-Resolution Contrail Detection & Attribution

Aircraft contrails are a significant contributor to aviation-induced climate effects, but attributing individual contrails to specific aircraft remains a challenge, especially when relying solely on satellite data. Satellite imagery offers extensive coverage but is limited by temporal and effective spatial resolution due to the large imaging distance. We propose a methodology that integrates freely available ground-based webcams, with ADS-B flight data to detect and attribute contrails to specific flights. 

We project flight trajectories from ADS-B data with latitude-longitude-altitude coordinates into the ground camera frame of reference using camera extrinsics pre-calculated from geographic points of reference. We then use a contrail segmentation method based on Canny edge detection and curve fitting to identify contrails in the frame of reference of the ground camera, capitalizing on the ground camera’s proximity and high framerate to attribute detected contrails to flights. Detected contrails can also be projected from ground camera space to satellite perspective, enabling the direct comparison between ground camera and satellite-based detection and attribution methods. With higher-accuracy ground-camera based attribution, we also enable the validation of more challenging satellite-based contrail attributions.

By leveraging the proximity of ground cameras for higher effective spatial resolution and their continuous capture for higher temporal resolution, our system addresses key limitations of satellite data. Our framework lays the groundwork for improved understanding of contrail creation, lifecycle, persistence, and more reliable monitoring of aviation-induced climate impacts. This method has the potential to enhance and validate operational contrail monitoring and avoidance and improve the integration of ground-based observations with satellite-based methods.