Hidden Industrial Activity in the World’s Oceans Revealed by Satellite Data
New research published in Nature shows that a large portion of industrial activity at sea remains hidden from public view. The findings imply that threats to the marine environment are more severe than previously understood. For example, about 75 percent of industrial fishing, especially in Asia, occurs without any documentary or visual record. In addition, roughly 25 percent of other activities such as energy development are undeclared.
The study combined artificial intelligence with satellite imagery to map maritime activity. Global Fishing Watch led the analysis and produced the first global map that traces large ship traffic and coastal infrastructure. This approach uncovered substantial activity that had largely escaped notice by traditional monitoring systems.
David Kroodsma, director of research and innovation at Global Fishing Watch and a co-author of the study, described this shift as an emerging Industrial Revolution in the oceans. He noted that on land we have detailed maps of roads and buildings, while the oceans have remained largely invisible. The new study helps close blind spots and illuminates the scale and intensity of human activity at sea.
Nearshore marine infrastructure is highlighted in the study as a focal point. The researchers from Global Fishing Watch, along with colleagues from the University of Wisconsin–Madison, Duke University, UC Santa Barbara, and SkyTruth, analyzed two million gigabytes of satellite images from 2017 to 2021. Their goal was to detect ships and nearshore assets across six continents where industrial activity concentrates in surprising ways.
One notable finding is that ships that did not transmit their positions could still be identified, revealing evidence of their existence and activities. By cross-referencing GPS data with five years of optical and radar images, the team could determine which vessels were engaged in fishing even without publicly available position data. Advanced computer systems then matched vessels to fishing behavior, effectively uncovering hidden operations.
Fernando Paolo, a co-author and senior machine learning engineer at Global Fishing Watch, noted that historically ship movements were not adequately documented. This gap limited understanding of how the world’s ocean resources are utilized. By combining space-based data with modern machine learning, the project maps unidentified industrial activity on a scale never before seen.
The study shows that three out of every four industrial fishing vessels are effectively hidden from public records. Even within marine protected areas, many vessels are concealed and, in some regions, there are high concentrations of activity where public monitoring had shown little prior presence.
Jennifer Raynor of the University of Wisconsin described a surprising discrepancy in publicly available data. Public sources often suggest similar fishing levels in Asia and Europe, but the detailed maps reveal Asia dominates, with seven out of ten observed fishing boats located there and only one in Europe. By bringing dark vessels into view, the team has created the most complete public picture of global industrial fishing to date.
The research also tracks changes in oceanic human activity over time. During the COVID-19 period, global fishing activity declined by nearly 12 percent, with an 8 percent drop in China and a 14 percent decrease elsewhere. In contrast, transport and energy-related maritime activity remained steadier through the period.
Beyond fishing, the study documents shifts in energy infrastructure. Oil platforms increased by about 16 percent, while offshore wind turbines grew substantially, with wind energy installations doubling during the study years. By 2021 the number of turbines was set to surpass oil platforms. China led the surge in offshore wind growth, showing dramatic expansion from 2017 to 2021.
Patrick Halpin, a Duke University professor of marine geospatial ecology and a co-author, emphasized that the Anthropocene footprint now extends beyond land. He argued that to understand the full industrial footprint of the oceans, growth in offshore wind, aquaculture, and mining must be considered alongside fishing, shipping, and oil and gas. The study portrays a dense, crowded, and complex blue economy increasingly driven by industrial activity in the sea.
The research also points to the potential of these mapping technologies to address climate change. Detailed ship traffic maps can refine estimates of marine greenhouse gas emissions, while infrastructure maps can guide wind energy development and help monitor environmental impacts from oil exploration.
Overall, the work reinforces a growing understanding that modern oceans are a significant arena of industrial development. It shows how technology can illuminate hidden flows and contribute to more informed governance of marine resources. A marked citation for this study appears in Nature, and it is attributed to the teams at Global Fishing Watch and partner institutions. The study is described in Nature under the article number s41586-023-06825-8 and is discussed in related analyses and press materials as cited by the authors and collaborators.
This evolving picture underscores the need for transparent, data-driven oversight of ocean use. Ongoing satellite monitoring and advanced analytics are likely to play a central role in understanding and managing the evolving blue economy, including fishing, shipping, energy extraction, wind energy, aquaculture, and related marine infrastructure.