A decade ago, a United States carrier strike group departing San Diego could cross the Pacific with only intermittent detection. China might catch a glimpse — a satellite pass here, a signal intercept there — but the gaps were wide enough to matter. Wide enough to maneuver. Wide enough to survive.
Today, China’s space-based surveillance architecture has matured to the point where U.S. naval movements can be monitored continuously, from homeport to operating area, with targeting-relevant precision.
The Pacific is not the only pressure point. In the Arctic, the threat is structural and accelerating. Russia has spent years militarizing its northern coastline, while China, despite holding no Arctic territory, has deepened its presence through shipping routes, resource partnerships and scientific outposts. A Hudson Institute analysis published in January 2026 warned that the U.S. alliance system may be unprepared to credibly counter the growing Russia-China alignment in the High North. At a congressional hearing in March, defense experts told lawmakers that the Arctic now sits squarely within homeland defense, and that sustaining a persistent operational picture of the region is no longer optional.
The pressure this creates on the defense and intelligence goes beyond access to imagery. Satellites are proliferating and data volumes are surging. At the GEOINT Symposium in May, National Geospatial-Intelligence Agency Director Lt. Gen. Michele Bredenkamp stated that the agency’s elevated operational tempo is now the permanent baseline, emphasizing a strategic shift toward a unified “data agency” model that leverages commercial capabilities to digest massive satellite data volumes.
Her predecessor, Vice Adm. Frank Whitworth, had put the risk more bluntly, warning that without the right infrastructure to process the coming wave of satellite data, those dense files risk falling on a cutting room floor.
The paradox of data abundance
For decades, the progress of Earth observation was measured by access. Governments and defense agencies built the ability to see more of the planet more often and in greater detail. That progress transformed terrain analysis, infrastructure monitoring, mission planning and battlefield situational awareness. For years, the focus was on the three Rs: resolution, revisit and reach.
The question has now shifted — from visibility to measurement, from whether we can see something to whether we can track its change consistently enough to act on it.
The instinct, when facing an intelligence gap, is to collect more. More satellites, more passes, more data. That instinct is understandable and increasingly insufficient. The intelligence community has more collection capacity today than at any point in history. Yet much of that capacity still delivers moments rather than measurements.
A snapshot of a port facility, a single pass over a contested airstrip or an Arctic coastline captured under different atmospheric conditions than the last one may exist in isolation, unable to answer the question that matters: What has changed, and how fast? When comparison breaks down, prediction breaks down with it. Analysts absorb the gap, reconciling inconsistencies in collection conditions before they can even begin to answer the mission question. In a low-stakes environment, that is an efficiency problem. In a high-stakes one, it is a decision advantage problem.
More imagery does not automatically mean more clarity. For the warfighter and the intelligence officer, that distinction carries real operational weight.
Why GeoAI still misses the pattern
Artificial intelligence has become central to how national security agencies process geospatial data at scale. The promise is great — faster detection, broader coverage, reduced analyst burden. But the promise is only as good as the data underneath it.
Geospatial artificial intelligence (GeoAI)depends on patterns that only emerge when observations are consistent enough to compare. When atmospheric conditions, viewing angles or calibration standards shift between passes, the model loses its footing. It cannot tell whether something on the ground has actually changed or whether the data itself has. Analyst trust erodes, signals get buried and the system produces volume, but not confidence.
When the underlying data is scientifically consistent and calibrated across time, patterns become visible that would otherwise stay hidden — the slow accumulation of change that precedes a crisis rather than announces it. Military mobilization, infrastructure development and shifts in maritime activity do not happen overnight. A model trained on trusted time-series data can catch these early. One trained on inconsistent data is, at best, guessing.
The shift from collection to measurement
The earlier era of Earth observation asked how much we could capture. The next asks how accurately we can detect change — and how confidently we can project it into the future.
This requires a different standard of data: consistent time of day, precise geometry, high signal-to-noise ratio and a trusted archive that grows more valuable with every passing day.
A contested port observed daily tells a very different story than one observed sporadically. Patterns emerge — vessel movements, construction activity, changes in force posture — that a single pass would never reveal. By the time an event is confirmed, the window to respond has often already closed. Measurement buys back that time.
That is what it means to move from collection to measurement — not seeing more; knowing sooner.
Building smarter bridges for geospatial intelligence
The future of geospatial intelligence will be shaped by the ability to operationalize change detection at daily, global scale. Meeting that standard does not require one platform, one vendor or one sensor. It requires a common measurement standard: data that is calibrated, consistent, comparable over time and ready for AI systems to use directly.
That shift has to start with the requirements we set. Policymakers and acquisition leaders need to stop treating resolution and revisit as the only tests that matter. Mission owners need to be clear about what they are asking the data to support: earlier warning, reliable GeoAI and decisions made under pressure. The real test is whether the data holds up from one pass to the next, whether analysts can trust the pattern and whether it gives them more confidence, not another feed to sort through.
Commercial providers have to meet that bar as well. The answer cannot simply be more imagery, more feeds and more files. The value is in delivering data that is consistent enough to support automated analysis and reliable enough to help identify change before the window to respond has closed.
The history of Earth observation was about learning to see. The next chapter is about learning to measure, with enough consistency and continuity to turn observation into foresight.
The world does not need more maps. It needs the ability to predict outcomes. For those who bear responsibility for national security, that is, ultimately, a measurement problem.
Nicole Toigo is president of EarthDaily Federal, where she leads the company’s work delivering near real-time geospatial intelligence solutions for the U.S. government across defense, search and rescue, and national security missions.
SpaceNews is committed to publishing our community’s diverse perspectives. Whether you’re an academic, executive, engineer or even just a concerned citizen of the cosmos, send your arguments and viewpoints to opinion (at) spacenews.com to be considered for publication online or in our next magazine. If you have something to submit, read some of our recent opinion articles and our submission guidelines to get a sense of what we’re looking for. The perspectives shared in these opinion articles are solely those of the authors and do not necessarily represent their employers or professional affiliations.