The earlier post about the status of the space-based segment of the Russian early-warning system got some media attention, which led colleagues to question my conclusion. And they may have a point - it may be a bit early to write these satellites off.
I was looking at the mean motion (the number of revolutions a day), which was a very good indicator of a working status of satellites of the earlier generation. The old Oko satellites had to be rather meticulous about station keeping as they were detecting missiles against the background of space (see this article for details). Tundra satellites apparently have true look-down capability, so they are more flexible with their orbits. Bart Hendrickx has a great overview of the Tundra/EKS satellites.
Since the notion of a station is different for Tundra satellites, the fact that we don't see standard orbit corrections does not necessarily tell us that the satellite stopped functioning. Jonathan McDowell noted that apogee longitudes of the four most recently launched satellites - Cosmos-2541, Cosmos-2546, Cosmos-2552, and Cosmos-2563 - haven't changed much--unlike those of Cosmos-2510 and Cosmos-2518, which are clearly off station.
Scott Tilley made a similar observation, also looking at apogee longitudes. He noticed, though, that starting around September 2025 all four satellites are shifting their apogees west. Since all four started the movement at the same time, Scott suggested that it's a planned maneuver to take satellites to a station that is easier to maintain, rather than a sign of dysfunction. Furthermore, he registered radio activity consistent with historical on-orbit operations and noted that there are no signs of uncontrolled behavior.
These are fairly convincing arguments, so it's quite possible that I jumped the gun in declaring the satellites non-operational. But maybe not--my understanding is that station-keeping for these kinds of orbits is a rather delicate process and a failure to perform distinct regular maneuvers is a sign that a satellite is not quite well. The groundtracks (see the image) also do not quite suggest a healthy constellation. But maybe that's because I've seen too many graphs of Oko satellites' mean motion. The current constellation is different. I guess the situation will be clearer in a few months.
Orbital data suggest that as of the end of 2025 Russia may have only one functioning early-warning satellite of the Tundra type. This is a significant decline from the situation in March 2025, when three satellites of the constellation - Cosmos-2541 (launched in September 2019), Cosmos-2552 (November 2021), and Cosmos-2563 (November 2022) - appeared to be operational.
Now it appears that for Cosmos-2541, the orbit correction maneuver successfully conducted in March 2025 was the last one. Another satellite of those three, Cosmos-2563, appears to have failed at some point after the last successful maneuver in July 2025. Images below show the changes in mean motion that testify to the failures.
The only satellite that doesn't show clear signs of failure is Cosmos-2552, launched in November 2021. However, based on recent patterns, it should have performed an orbit correction sometime in November 2025 (see the main image in the post). But it is too early to say that Cosmos-2552 has ended its operations.
I should note again that the apparent loss of early-warning satellites is not necessarily a cause for alarm. Russia does not rely on the space-based segment of its early-warning system to the extent the United States does. For a discussion, see this 2015 post or my Science & Global Security article.
Oreshnik is an elusive missile. It shows up in various statements and there are some pretty tangible signs of its existence, but it is still not quite clear what the actual status of the missile is.
The most notable recent appearance of Oreshnik was in the Belarusian leader's 18 December 2025 address, in which he announced that Oreshnik was delivered to Belarus on 17 December 2025 and that it is "being placed on combat duty." Is it really?
My take is that there are reasons to be skeptical. Yes, there is strong evidence of preparations for the deployment, such as the potential deployment site found by Jeffrey Lewis' team (and confirmed by US intelligence). But there are also reasons not to read this evidence too literally, at least at this point in time.
This post is an attempt to collect what we know about Oreshnik to see how various pieces of the story fit together.
The name Oreshnik first appeared on 21 November 2024, when Russia used this missile to strike the Yuzhmash plant in Ukraine. (Here is a video of warheads hitting their targets.) In a special televised address the president of Russia described the strike as a test of a medium-range "non-nuclear hypersonic ballistic missile."
The missile was launched from Kapustin Yar, which is about 800 km away from Yuzhmash. This means that it is indeed a medium-range missile. Note that the reason it was described as a test is that technically the Russian INF moratorium proposal was still on the table. Apparently, the (not quite convincing) logic was that a test is different from deployment, so it would not violate the moratorium. Speaking at a meeting with designers on 22 November 2024, the Russian president confirmed the very narrow reading of the moratorium when he said that Oreshnik is only one of "a whole line of medium- and shorter-range systems." Nevertheless, the framing of the launch as a test was an interesting detail.
According to the report of the commander of the Strategic Rocket Forces at the 22 November 2024 meeting, the missile was developed in accordance with the presidential decision of July 2023. It appears that the missile was tested at least twice before November 2024 - in October 2023 and June 2024 (MilitaryRussia.ru citing a report by the Ukrainian Main Intelligence Directorate).
The Pentagon assessed that the missile was based on RS-26 Rubezh. The Pentagon spokesperson also said that "the United States was pre-notified briefly before the launch through nuclear risk reduction channels." Technically, since the missile was not a true ICBM, Russia was not under an obligation to send a notification, but it apparently decided to do so to avoid potential misunderstanding. If the missile is indeed based on RS-26, its signature during the boost phase would be very similar to Yars (since RS-26 was said to be based on RS-24 Yars). One can never be too careful. The Kremlin spokesman later said that the notification was "an automated warning [that] was sent 30 minutes before the launch." This is most certainly an error - these notifications are not automatic, and by all indications the warning was sent about 24 hours in advance, as required by the US-Russian agreement. We know that rumors of an upcoming ICBM launch were circulating in Ukraine the day before the strike.
The Belarusian involvement in the Oreshnik story started in December 2024. After a bilateral meeting on 6 December 2024, the president of Belarus asked for the missile to be deployed in Belarus (with a condition, though - that "the targets for these weapons" be determined by Belarus). The Russian president responded that such a deployment "is feasible." He also noted that it would be possible after the serial production of these missiles "is ratcheted up" and they are deployed with Russia's Strategic Rocket Forces (RVSN). Moreover, the Russian president suggested that even though the missiles will be operated by RVSN, it would be up to Belarus "to identify the targets."
This kind of targeting arrangement does not look particularly realistic and suggests that something different is going on here. The exchange clearly followed the pattern that had been established earlier regarding the deployment of nuclear weapons in Belarus. The Belarusian leader makes bold statements and Russia plays along, especially during joint appearances.
A few days after this exchange, on 16 December 2024, the president of Russia said in an address to the Ministry of Defense that the serial production of Oreshnik "should begin in the near future." It's notable, though, that the Russian president did not mention the plan to deploy these missiles in Belarus (although he did say that these systems will be used "to protect Russia and our allies' security").
This pattern will persist through the entire year. While the president of Belarus has been constantly mentioning the deployment plan, his Russian counterpart has been more reserved. With the notable exception of joint appearances. Note, though, that he never seems to volunteer any details, opting to confirm the words of the Belarusian president. For example, at a meeting in August 2025, the Russian president confirmed that the site for the deployment in Belarus had been selected and that the missiles would be delivered by the end of the year.
At the same August meeting, the Russian president said that the industry had produced the first serial production system, which "has already been delivered to the troops." A few days later, the Ministry of Foreign Affairs issued a statement that formally withdrew the INF moratorium proposal made in October 2020. The statement, however, did not mention Oreshnik or any other specific system, noting instead that "decisions on the specific parameters of response measures will be made by the leadership of the Russian Federation" based on the analysis of the situation.
News about Oreshnik continued to come mostly from Belarus. However, the deployment announcement on 18 December 2025 came after the president of Russia said at a meeting at the Ministry of Defense on 17 December 2025 that "[b]y the end of the year, the medium-range missile system armed with the Oreshnik hypersonic missile will be placed on combat duty." Notably, he said nothing that would suggest that the missiles would be deployed in Belarus. Neither did he mention Oreshnik in Belarus when asked directly during the end-of-the-year press conference. The Minister of Defense, who spoke after the president, also said nothing about Belarus when mentioning Oreshnik. The Chief of the General Staff, who spoke the next day at a meeting with military attachés, told them that "a brigade has been formed equipped with a new medium-range missile system, Oreshnik."
I would say that "a brigade has been formed" is pretty far from "missiles have been delivered to Belarus where they are entering combat duty." I am far from suggesting that Oreshnik is a phantom missile - the evidence suggests that it is not. But at the same time, I do believe that we should be very skeptical about reports of its deployment in Belarus. I find it very hard to believe that the Strategic Rocket Forces, which are expected to operate the missile, will be happy about being stationed outside Russia. Besides, the case for the military utility of the deployment is virtually non-existent. Note that the Krichev-6 site is literally seven kilometers from the Russian border (see the image above).
The "division of labor" between Russia and Belarus regarding virtually all news about Oreshnik also makes me suspicious. We have seen a similar pattern with nuclear weapons - Russia lets Belarus make all kinds of statements about them and even builds a storage facility that should be capable of accepting these weapons if necessary. But there has been no confirmation of the deployment from the Russian side and there are no signs of weapons being actually delivered to Belarus (I believe they will never be deployed there, but that's a topic for another post). The same seems to be the case with the missile base. The infrastructure, of course, could become useful someday, but we are not there yet. For the moment, it appears that both sides are involved in a rather strange political spectacle. I still hope that it will not involve actual movements of missiles (not to mention nuclear weapons), but we cannot exclude that it will.
Russia appears to have made another attempt to launch a Yars ICBM from Plesetsk. It issued a NOTAM notification with a window opening at 6:00 UTC on 25 December 2025. It probably notified the United States about the upcoming launch: the United States dispatched an RC-135 Cobra Ball aircraft to monitor the launch. The aircraft returned to the base after an 18-hour flight, but the launch did not take place on the 25th as the NOTAM was not cancelled. Cobra Ball flew another mission on December 26th (taking off around 2:00 UTC). The NOTAMs were cancelled around 08:20 UTC on December 26th and the aircraft returned to the base.
It is not clear whether a launch took place on the 26th or was cancelled as was apparently the case with the attempt in early December. Given that no reports about the launch appeared in the Russian media, a cancellation is more likely.
In the traditional 17 December interview marking the anniversary of the Strategic Rocket Forces (established on that day in 1959), commander Sergey Karakayev reported that his force completed the deployment of Yars missiles in the silos of the missile division in Kozelsk. (Video of the deployment by Zvezda TV.)
This brings the number of silo-based Yars missiles in the 28th Missile Division to 30. The missiles are deployed in three regiments - 74th, 168th, and 373rd. The older three regiments of the division - 119th, 214th, and 372nd - were disbanded around 2007-2009. Here is the kmz file with silo coordinates.
The decision to keep three regiments and use them to deploy "new-generation missiles" was made in 2008. In 2011, the Rocket Forces announced that these silos would be used to deploy Yars missiles. The first two missiles were deployed in 2014.
In an interview to Krasnaya Zvezda the commander of the Strategic Rocket Forces said that the first regiment of silo-based Years missiles at the Tatishchevo division began combat duty in 2025. Apparently, the plan is to replace all single-warhead Topol-M missiles with MIRVed Yars. The first two Topol-M missiles were deployed at Tatishchevo in December 1997.
Although the Golden Dome will soon mark its first anniversary, there is still not much clarity on what the system will consist of, what it will be asked to do, or, more importantly, what it will be able to do. One idea, however, surfaces with some regularity - space-based missile defense. This does not mean that this idea has some promise - the physics hasn't changed since the last time it was discussed in the 1980s. Building a system that would be even marginally capable is still prohibitively difficult.
The current enthusiasm about space-based defense largely stems from the fact that the launch cost has been reduced quite dramatically and constellations of thousands satellites do not seem particularly impossible. Another reason to believe that the idea of space-based defense will get some traction is that it would be relatively easy to arrange a test of the capability (especially if it does not involve an actual intercept). Finally, space-based missile defense could provide anti-satellite capability, which some people in the US seem to find useful.
As for the actual missile defense capability that the Golden Dome would provide, it is always useful to remember that missile defense works perfectly ... against missiles that don't exist. This was the lesson from the 1960s, when it turned out that missile defense systems of the time cannot do anything with actual ICBMs that both the United States and the Soviet Union started producing "like sausages" (well, almost). This was also the case in the 1980s, when the US plans to build Strategic Defense Initiative led Soviet designers to see how they could make the life of SDI complicated.
The image on top shows what they came up with - it is a translation of one of the slides in the Katayev archive (it's from my article about Soviet response to SDI). The idea was rather simple - to create more targets for the space-based missile defense layer, they suggested "modular missiles" that would have several second stages to deploy their warheads.
The estimate, which was completed ca. 1987 suggested that the modular scheme would not dramatically reduce the number of warheads that the missiles can carry. At the same time, it would significantly improve the chances of penetrating the space-based layer of missile defense. The two tables below are taken from the same document in the Katayev archive.
Estimated missile defense penetration effectiveness of R-36M2 and RT-23UTTH missiles with MIRVs and modular configurations
| RVSN missile forces (after reductions), with R-36M2 and RT-23UTTH missiles | Warheads penetrating space-based missile defense (% of launched warheads) | ||
|---|---|---|---|
| Interceptors | Interceptors and lasers | ||
| Missiles with MIRVs, enhanced survivability, and shortened boost and post-boost phases | R-36M2 | 13 ... 16 | 5 ... 6 |
| RT-23UTTH | 13 ... 28 | 6 ... 15 | |
| Modular missiles | R-36M3 | 70 ... 75 | 65 ... 70 |
| RT-23M | 40 ... 45 | 35 ... 40 | |
| Note: Modular missile configurations improve penetration effectiveness against U.S. space-based missile defenses by a factor of 4.5-14 for R-36M3 missiles and 1.5-6.5 for RT-23M missiles, compared to other measures. | |||
R-36M2 and RT-23UTTH missiles with MIRVs and modular configurations
| Specification | R-36M2 | RT-23UTTH | ||
|---|---|---|---|---|
| MIRV | Modular | MIRV | Modular | |
| Launch weight, tonnes | 211.2 | 211 | 105 | 105 |
| Throw weight, kg | 8,740 | 6,400 | 3,650 | 3,200 |
| Number of warheads | 10 | 8 | 10 | 5 ... 10 |
The second table shows that this improvement in counter-missile defense performance would not have led to a significant degradation of the missiles' capabilities. R-36M3 would carry eight warheads instead of its regular ten, but the launch weight would not have changed, probably meaning that the missile could have been deployed in the same silos. Things were a bit different for R-23, but not too much.
A colleague noted that all these numbers should be taken with a (big) pinch of salt. I agree, of course. The point is that these kinds of estimates supported the case that the Soviet Union should not worry too much about the SDI program. Since these estimates were produced internally, by the same institutions that were in charge of building missiles, they allowed the political leadership to "untie" missile defense from disarmament discussions (INF and START). They showed that any missile defense eventually meets a missile against which it doesn't quite work.
The Knyaz Pozharsky submarine of the Project 955A/Borey-A class was accepted for service in July 2025. The formal acceptance would normally take place in December, but with Knyaz Pozharsky it was postponed for about six months, presumably to complete the trials which include a test launch of a Bulava missile.
It is not clear if the expected test launch actually took place. In June 2025 Russia submitted NOTAM notifications that correspond to a Bulava test from the Barents Sea to Kura during the 20-21 June 2025 window (kmz file, h/t EM for identifying the dates). There were no reports about the launch, however. This does not necessarily mean the launch failed, although it is unusual to see a successful test not being officially reported. It's not unprecedented - see a launch in 2016 and in 2023 - but rather rare. Also, these were not state acceptance launches.
One sign that may indicate a scrubbed launch is the NOTAM cancellation record. Or, rather, lack thereof. NOTAMs were not cancelled and just expired when the announced launch window ended at 17:00 on 21 June 2025 (it opened at 05:00 on 20 June 2025). Normally, a NOTAM is cancelled right after a successful launch. See, for example, the Sineva launch on 22 October 2025 - the NOTAM covered the window from 21 October to 28 October, but was cancelled at 12:48 on 22 October 2025, right after the launch. This may not be very conclusive, however - the June window was rather narrow and the launch may have taken place close to its end anyway.
At this point I would wait for more evidence one way or another. It does appear that the launch was scrubbed, but I don't think it can be said with certainty.
On 1 December 2025 Russia issued a series of NOTAM notifications that indicated an upcoming ICBM launch from Plesetsk to Kura during the 3-7 December 2025 window. The missile to be tested was certain to be Yars (or some of its modifications, like Sirena-M command missile). However, as Thord Ave Iversen noted, the NOTAMs were cancelled on 2 December 2025. This suggests that the test launch was cancelled or postponed.
The most recent launch of the Yars missile took place during the strategic exercise in October 2025. This may indicate that the planned launch was a development test rather than a test to confirm the reliability of the missile.
On 28 November 2025, local Russian media and then Telegram channels reported an explosion near Yasnyy in Orenburg oblast. The town is close to the old basing area of the 13th missile division.
A full video of the incident shows the missile making an effort to right its trajectory. At about 7 seconds into the flight, the first stage engine visibly fails, the missile begins to tumble, and catches fire shortly afterwards. Interestingly, at about 15 seconds, something is ejected from the head section of the missile. This could be a payload salvage sequence.
The silo that was used for the launch has been identified as one of the silos of the 621st missile regiment, located at 50.97280, 59.55149. This silo was used for Dnepr space launches in the past (Dnepr was a converted R-36MUTTH ICBM). The last Dnepr launch was in 2015, so the silo had not been used since then. However, starting in the spring of 2025, some construction activity was detected at the site.
By all indications, it was an attempt to conduct a regular ICBM test launch with the missile delivering its warhead(s) to the Kura test site in Kamchatka. Russia issued NOTAMs for Yasnyy as well as for Kura.
The type of the missile is rather difficult to determine, but most likely it was an attempted test launch of the Sarmat ICBM. The other two possibilities are a UR-100NUTTH/SS-19 missile with Avangard and an R-36M2/SS-18.
SS-19 can be ruled out as the video shows a cold launch from a silo, while SS-19 uses a hot launch. Here is an Avangard test in 2018 that shows it very clearly. Interestingly, Avangard tests also used one of the 621st missile regiment's silos, the one at 51.030849,59.690144.
As for R-36M2, the missile has not been tested for quite a while. The last launch of the missile apparently took place in October 2013. While some of these missiles may still be deployed, it is extremely unlikely that the Rocket Forces would want to test launch them.
This leaves Sarmat. Normally, Sarmat would be launched from Plesetsk, but the Yubileynaya test silo there was destroyed in a test in September 2024. No activity has been detected in Plesetsk since then, so it appears that the tests were moved to Dombarovskiy.
So, it appears to be another setback for the Sarmat program, and it appears highly unlikely that the missile will be deployed in 2025, as was announced by the Russian president earlier this month.
UPDATE: Post-incident satellite images of the site (as of November 29, 07:31 UTC).
UPDATE: La Fondation pour la recherche stratégique has published a detailed analysis of the event.
The book
