Mapping Macroplastics at Scale: Calibration and Multi-Year Global Analysis from the Automated Debris Imaging System (ADIS)

To understand the full scope of plastic pollution in aquatic environments, continuous and large-scale monitoring techniques are essential, especially for tracking mass balance and long-term trends. Over recent years, advancements in automated monitoring of floating marine debris have gained momentum. Satellite-based mapping of coastal regions has progressed significantly with the development of multispectral, hyperspectral, thermal infrared, polarimetry, and synthetic aperture radar (SAR) methods, among others. 

However, tracking debris in the open ocean presents unique challenges. In this environment, floating debris is widely dispersed, making satellite detection difficult. Large floating macroplastics (>50 cm) represent a significant portion of the ocean plastic mass budget, yet remain under-quantified by traditional methods, such as surface trawls and visual surveys that are costly and labor intensive. 

This presentation highlights recent advancements in the Automated Debris Imaging System (ADIS), a technology developed to tackle these limitations. Designed to be deployed onboard existing vessels of opportunity, ADIS uses optical imaging combined with deep learning algorithms to detect and capture geo-tagged snippets of floating macroplastics (>50 cm) along sailing routes. Recently endorsed as an official UN Decade of Ocean Science action, ADIS offers promising potential for global-scale debris tracking. 

Since the initial proof of concept, we have built a comprehensive global dataset comprising over 10 million images, identifying thousands of floating macroplastics. By comparing ADIS detections with concurrent surface net samples and visual observations from selected transects, we can now rigorously evaluate the system's accuracy and precision, underscoring its value for global marine debris monitoring.