Roughly 99 percent of the gas in Earth’s upper atmosphere carries no electrical charge — and that neutral majority happens to be the portion our instruments have the hardest time detecting. The remaining fraction, the ionosphere’s ionized gas, is comparatively simple to track: charged particles disturb the passage of radio waves and leave behind a signature that can genuinely be measured.
Faced with that problem, a Kyoto University team abandoned direct observation of the neutral atmosphere and instead began monitoring its effect on objects passing through it. Thousands of objects, to be precise: Starlink satellites.
What came out of it is a two-dimensional latitude-longitude portrait of thermospheric density at an altitude of roughly 500 kilometers — the first tomographic analysis of its kind, according to the team.
Medical imaging math, pointed upward
Tomography is the method behind a CT scanner: image something opaque by measuring what travels through it from many different angles. In this case, the thing travelling through is a satellite, and the measurement is drag.
Working from publicly available orbital information on Starlink satellites, the researchers derived atmospheric density from the slow decay of those orbits. The approach yielded density estimates surrounding approximately 1,200 satellites operating at an altitude of 482 kilometers.
The term thermospheric density describes the neutral atmosphere spanning roughly 100 to 1000 kilometers in altitude. Even that high up, enough gas remains to decelerate a satellite — which is exactly why nailing the figure matters for anyone forecasting a spacecraft’s position a week from now.
The check against SWARM
Any new technique is judged by what it lines up with. The density patterns generated by the Kyoto group turned out to be strongly consistent with data from the European Space Agency’s SWARM satellites, which record shifts in atmospheric density along the paths of their own orbits.
That agreement is the noteworthy part. SWARM consists of dedicated hardware performing in-situ measurement. Starlink orbital data, by contrast, is a free byproduct of a commercial broadband constellation.
This is the second pass, not the first
The study extends earlier work by the same researchers, who used Two-Line Element data — TLE — from Starlink satellites to estimate how thermospheric density shifted across time and altitude. As orbital information goes, TLE is about as generic as it gets.
What the latest analysis contributes is the horizontal axis: a map of how density changes with latitude and longitude, revealing more of the thermosphere’s geographic structure.
“This is a multidisciplinary study between space science and space engineering,” said corresponding author Mamoru Yamamoto. “Reading papers from both research fields, we realized that deeper dialogue between researchers from both fields is necessary.”
Why anyone outside atmospheric science should care
Satellites and debris continue to pile up in low Earth orbit. Sharper density figures translate directly into sharper forecasts of where all that material will drift, lowering the chances of collisions — satellite against satellite, and satellite against debris.
Down the line, the method could also enable near-real-time readings of atmospheric density in a satellite’s vicinity, a boon for space weather forecasting and satellite operations. The operative word is “eventually.” For now there is a snapshot and a validated method, not a live feed.
The economics, though, are difficult to dispute. Each Starlink satellite already transmits its orbit, and each one is steadily being dragged by an atmospheric layer that is otherwise invisible. What the Kyoto team supplied was the realization that the constellation had been taking the measurement all along.