By Preligens, 04/13/22
Read the article on SpaceNews website here.
The war in Ukraine has put the importance of information dominance on full display. Western media have made extensive use of commercial satellite imagery to document Russia’s invasion of neighboring Ukraine. And while this imagery has played an important role in galvanizing world opinion against Russia, the United States and its allies nonetheless stand at a pivotal moment for information dominance.
Earth observation represents an intense international contest with billions of dollars at stake in the government and commercial sectors. By most metrics, the U.S. maintains a commanding lead in space technologies. The U.S. operates approximately 2,800 satellites; China operates fewer than 500, while Russia has fewer than 200. U.S. organizations also have lowered barriers to entry, slashing launch costs to under $2,000 per kilogram to low Earth orbit. Despite the high levels of investment and achievement, too few resources go to analytics and generating insights from Earth observation information, opening opportunities for adversaries to overtake the U.S. and allies.
The path to preserving U.S. and allied information dominance starts with acknowledging the competitive landscape. During the 2021 GEOINT Symposium, Dave Gauthier, director of the Commercial and Business Operations Group at the U.S. National Geospatial-Intelligence Agency, shared NGA’s Olympic Games-themed assessment of the world’s commercial Earth observation capabilities. The evaluation highlighted the growing global competition between commercial operators in Earth observation, with Washington and allies confronted by rising rivals. Of the nine categories, companies from the U.S. collectively claimed three gold medals, as did companies from China.
The number of Earth observation satellites has expanded fivefold since 2012, with additional increases anticipated for the foreseeable future. As more satellites reach orbit, the era of persistent surveillance approaches. Soon anywhere on Earth can be imaged at any time. If Earth observation data were flowing water, a once dripping faucet has morphed into a garden hose and will soon be a firehose. Insufficient investments in information processing and exploitation mean many government agencies worldwide will drown in data.
The value of Earth observation data also has evolved, with more sources becoming available. As with most commodities, prices go down when supplies increase. Legacy business models have changed, too, with satellite data accessible in new formats and platforms. Consumers can still task and acquire individual images, but innovative licensing agreements and APIs now permit access to vast libraries for product development or support to fleeting operations.
To ensure the West’s continued information dominance, the U.S. and its allies must be able to process and analyze terabytes of Earth observation data before adversaries. The maturing Earth observation market means accessing data presents fewer challenges than leveraging what any global competitor can acquire. In many countries currently engaged in conflict, information dominance permeates military doctrine and drives operations. As the conflict in Ukraine unfolds, the U.S. and its allies have shown the value of dispelling Russian misinformation campaigns with Earth observation data.
Meanwhile, government agencies continue to confront issues such as cybersecurity and non-attributable tasking, as well as the urgent need to scale analytics capabilities to make sense of the ever-increasing volume of Earth observation data.
Scaling analysis of Earth observation data requires balancing traditional tradecraft and automation, with human-machine teaming at the center. Established practices for deriving insights from satellite imagery and other remote sensing sources benefit from decades of success. In addition, applying artificial intelligence and machine learning solutions to redundant procedures frees analysts to focus on high-priority issues, enabling more rapid contextualization and interpretation of complex events. Once implemented, well-executed human-machine teaming workflows allow for scaled analysis of Earth observation data while maintaining confidence in results and assessments.
Last year, I contributed to a white paper published by the U.S. Geospatial-Intelligence Foundation titled: “The State of Artificial Intelligence and Machine Learning (AI/ML) in the GEOINT Community and Beyond.” The white paper highlights a series of recommendations for AI/ML adoption and countering adversaries. As a call to action, the moment has arrived for the U.S. and its allies to cement information dominance in the GEOINT field. Simply put, creating conditions for government, commercial, and academic collaboration in AI/ML will secure the U.S. and allies’ competitive advantage over adversaries.
The U.S. and its allies should expand space policies to include robust support for organizations building AI/ML solutions for Earth observation data. A list of engineering feats in the space sector would fill a book — and reveal disparities between systems and analytics. Investments in launch, spacecraft, sensors and communications dwarf what has been dedicated to analyzing Earth observation data. To date, the inequity of resources represents a missed opportunity for the U.S. and its allies. It also contrasts with adversaries such as Russia and China that prioritize the development of AI/ML capabilities. Maintaining the status quo risks not only missed opportunity but also casting the U.S. and its allies as runners up in the race for information dominance.