The Price of a Close Approach

1. From 475 kilometers to a few meters

On 12 and 13 October the two satellites were about 475 km apart, with Cosmos 2612 trailing and closing slowly. Then it sped up. LeoLabs told Axios that the gap was more than 200 miles on Thursday and roughly three miles on Friday afternoon. The company called it a deliberate approach, said behavior like this is relatively uncommon in low Earth orbit, and said the intent remains uncertain.

Marco Langbroek, a Dutch analyst who has tracked military satellites from Leiden for years, was skeptical over the summer that Cosmos 2612 was after X36 at all. The orbits were similar, but the geometry did not produce frequent close passes, and he thought the match could be coincidence. The 30 September maneuvers changed his mind. In a post on Saturday, he put the separation at 1 to 2 km as of 18:15 UTC, based on the latest public orbital elements, and calculated that Cosmos 2612 might pass within a few meters of X36 in the early hours of 18 October UTC.

That pass, if it happened, would have taken place in the last few hours. As of this writing there is no public report of it, and no report of damage or lost service. Langbroek's figure is a prediction from two-line element sets, a public orbit format that NASA's conjunction guidance notes lacks the uncertainty information you need for meter-level estimates. It tells you the approach was continuing. It is not a measured distance.

Clayton Swope at CSIS described the Friday distance to Axios as bumper-to-bumper. Aaron Brynildson, who teaches space law at the University of Mississippi, told the same reporters that no rule sets a minimum distance between satellites. Countries usually keep around six miles apart as a matter of practice.

ICEYE declined to comment. The Pentagon said it had nothing to release.

2. Why this satellite

ICEYE signed a contract in August 2022 giving Ukraine's government access to its radar constellation, and this January it announced expanded work with a customer inside Ukraine's Ministry of Defence. Synthetic aperture radar sees through cloud and at night, which makes it useful for watching military equipment move. A Ukrainian official told Axios that the military uses ICEYE's capabilities but could not say whether X36 specifically is part of that work.

This is also not the first approach. Russia launched six satellites, Cosmos 2609 through 2614, from Plesetsk in April. According to CSIS, five of them made plane changes in May to match the ICEYE satellite's orbit, and Cosmos 2609 came within about 2.5 km of it on 26 May. Then the planes drifted apart until Cosmos 2612 went back on 30 September.

Possible reasons include inspection, collecting signals, testing a capability, or sending a message to the countries buying the imagery. None of these has been established, and each would mean something different for ICEYE's customers.

3. The attack that already happened came through a VPN

If you want to know what conflict does to a commercial space business, the best evidence is from 24 February 2022, the morning Russia invaded Ukraine, and it involved no spacecraft maneuvers at all.

At about 03:02 UTC, Viasat's KA-SAT network started seeing large volumes of malicious traffic from customer modems in Ukraine. Around 04:15 UTC, modems across much of Europe began dropping offline, and over roughly 45 minutes tens of thousands of them disappeared from the network. According to Viasat's own incident report, the attacker had come in through a misconfigured VPN appliance, moved into the network segment used to manage the system, and used legitimate management commands to overwrite the flash memory on residential modems in bulk.

The satellite was fine. Customers in Ukraine and across Europe lost their connections anyway. Viasat stabilized the network within hours, fully within days, and shipped nearly 30,000 replacement modems to distributors. In May 2022 the European Union attributed the attack to Russia.

For an operator or investor, KA-SAT is the best-documented case of what conflict costs a commercial space company. The damage was done on the ground. Most of the people who lost service had nothing to do with the war, and the company recovered by buying and shipping hardware.

4. Where the money actually sits

Three numbers get passed around in space-economy coverage, and they measure different things.

Do not add these. Annual revenue, capital raised, and a ten-year forecast are different kinds of quantity, and the forecast is not $1.8 trillion "at risk" in any accounting sense.

The SIA breakdown is where the useful detail is. Of the $303 billion, $165.2 billion is ground equipment, and $125.8 billion of that is GNSS equipment: the navigation chips in phones, cars, ships and farm machinery. Satellite services, which includes broadband and TV, are $105 billion. Remote sensing, the business ICEYE is in, is $3.6 billion. Launch is $12.4 billion. Everything filed under space sustainability, which covers servicing, life extension, situational awareness and debris removal, comes to about $500 million.

So the largest single slice of the satellite industry's revenue is receivers on Earth that depend on a signal from orbit. That signal can be jammed from the ground without anyone touching a satellite.

A 2019 study presented to the U.S. GPS advisory board modeled a 30-day GPS outage at more than $1 billion a day in U.S. losses, about $30.3 billion in total. Maritime operations accounted for $10.4 billion of that, mostly ports, and telecommunications for $9.8 billion. It is an old, conditional scenario. It still gives a sense of how much ordinary commerce runs on timing and position data.

The imaging business works the other way. Remote sensing revenue is small, but governments now buy it in wartime, which raises demand and exposure together.

5. What it costs before anything breaks

Nobody has published what the last three weeks have cost ICEYE, and the company has said nothing. We can list the kinds of cost an operator takes on when another country's satellite parks next to theirs, without assigning numbers to them.

Someone has to watch it around the clock. If the operator decides to move, it spends propellant that was budgeted for the satellite's working life, and a 90 kg satellite does not carry much. Imaging tasks may be dropped while the spacecraft maneuvers. Customers, especially government customers, will ask what is happening and what the operator is doing about it. If the operator carries insurance, the policy wording on hostile acts, war and cyber events decides what is recoverable, and those exclusions vary.

Then there is financing. A company raising money for a constellation now has to answer questions about a risk that was abstract a year ago. We do not know yet whether any insurer or lender has changed terms because of this encounter. We would like to.

If you operate satellites, or work in space insurance or finance and have seen this come up, we want to hear what a close approach actually costs you in staff hours, fuel, lost collections or changed terms. Send it to info@populationmars.com. Real numbers will go in a follow-up piece, with your name or without it.

6. The United States does this too

A fair reading of Cosmos 2612 needs the same standard applied to everyone.

The U.S. Space Force operates the Geosynchronous Space Situational Awareness Program, satellites built to approach and inspect other objects in geostationary orbit, and the Secure World Foundation has documented U.S. approaches to foreign satellites. In September 1985 the U.S. Air Force destroyed its own Solwind satellite with a missile fired from an F-15. China destroyed one of its weather satellites in 2007. Russia's November 2021 missile test against its own Cosmos 1408 produced more than 1,500 trackable fragments, according to U.S. Space Command.

On 14 September, Air Force Secretary Troy Meink told the Air & Space Forces Association's conference in National Harbor, Maryland, that the Space Force now has "on-orbit space control weapons." He would not say whether they are kinetic or non-kinetic, or whether they have been tested, and he said the wording was deliberate. That is a public acknowledgment of weapons already in orbit. It does not tell us when they were deployed, how many there are, or what they do.

The law gives little guidance on how close is too close. The Outer Space Treaty bans nuclear weapons in orbit and national claims of territory, and requires "due regard" for other countries' activities. It sets no keep-out distance. The Liability Convention makes a state liable for damage in space only where fault can be shown, and showing fault depends on evidence that is often classified or does not exist.

7. The navy comparison, with dates

A common argument holds that American naval power kept sea lanes open after 1945, trade grew, and space needs the same thing. Naval power was part of it. So were the Bretton Woods monetary agreements of 1944 and the trade rules that became GATT in 1947, along with reconstruction lending and new shipping technology.

The Strait of Hormuz this month shows the limit of the military half on its own. The EIA's October 2026 Short-Term Energy Outlook describes constrained flows, convoys and rerouting, in a region where the U.S. Navy has operated for decades.

Orbit is also harder to patrol than a strait. Every satellite is moving, changing orbit costs fuel and time, debris from a fight stays in the way of everyone afterward, and the most damaging attack so far went through a ground network.

8. The missing product is evidence

Langbroek added a note to his Saturday post that is the most important sentence in this story for the industry. If ICEYE-X36 failed tonight, he wrote, it would look suspicious, but it would be very hard to attribute the failure to Cosmos 2612 without ambiguity.

Satellites fail on their own every year from radiation damage and software faults. An operator that loses a spacecraft next to a foreign military satellite has to show its insurer and customers what happened, and possibly a court. Today that evidence is scattered across classified sensors, commercial trackers like LeoLabs and Slingshot, public orbital elements, and the operator's own telemetry, and none of it is collected in a form designed to settle a dispute.

That is a real business gap. Tracking companies already sell conjunction alerts. The next step is evidence an insurer or a government will accept: maneuver records with stated uncertainty, timestamps that can be checked independently, and a clear record of who consented to what. Servicing and debris-removal companies face the same question from the other side. A spacecraft that can approach and grab a dead satellite can also approach a live one, so Astroscale or ClearSpace will need to prove what their vehicles did and did not do.

The same problem will follow industry out of Earth orbit. NASA's lunar relay contract with Intuitive Machines, announced in September 2024, has a ceiling of $4.82 billion over up to ten years. That is a maximum, not money spent. It is also a shared communications and navigation service that every lunar mission will lean on, and the rules for who can approach it, and how a failure gets investigated, have not been written.

9. What to watch

Integrity ISR expects two of Cosmos 2612's sister satellites, 2611 and 2613, to cross the orbital plane of another ICEYE satellite, X37, around 6 November and 15 November. If either one repeats the climb-and-descend sequence that 2612 used, it will show up in public tracking data a week or two before the closest approach.

Langbroek posts his analysis at SatTrackCam Leiden. LeoLabs publishes its tracking commentary on its own site and through reporters. ICEYE has not said whether X36 is healthy. That is the next fact worth having, and we will update this piece when someone states it on the record.