Astronauts aboard the International Space Station were directed into their docked Soyuz escape craft after the air leak rate in the Russian Zvezda service module’s transfer tunnel doubled to roughly two pounds per day. The cracks responsible for the leak have been tracked since 2019, but the sharp increase came during Progress 95 cargo operations the week of June 1, prompting NASA and Roscosmos to act. The shelter order and the decision to pursue more extensive repairs signal growing concern about the structural condition of one of the station’s oldest pressurized segments.
Why the Zvezda tunnel leak forced a shelter order
The leak is centered in the PrK, the transfer tunnel connecting the Zvezda service module to the rest of the station’s Russian segment. Cracks in that tunnel wall were first detected in 2019, and engineers on both sides of the partnership have monitored them continuously since then. For years the air loss remained manageable, but during the week of June 1, while the Progress 95 cargo vehicle was being loaded and reconfigured, Roscosmos teams recorded a rapid increase in the rate of escaping atmosphere, according to a recent update from NASA.
The timing raises a pointed question: did mechanical forces from the Progress 95 docking or cargo transfer operations stress the already-weakened tunnel wall? Cargo vehicles impose vibration loads and pressure transients when they dock, undock, and shift mass inside the station. If the leak-rate spike tracks with those events, future docking schedules may need to account for the fragility of the PrK structure. Comparing pressure telemetry before, during, and after each Progress arrival would test that correlation directly, though neither NASA nor Roscosmos has publicly released the raw sensor logs needed for independent verification.
For the crew, the practical consequence was immediate. NASA directed the astronauts to move into their Soyuz vehicle, the capsule that doubles as their emergency lifeboat, while Roscosmos teams worked on repairs from the Russian side of the station. The shelter period was brief, but the fact that it happened at all reflects how seriously both agencies treated the rate change. Roscosmos then opted for a more extensive repair after discovering additional problems during the initial fix, a decision that underscores how dynamic and uncertain the structural situation inside the PrK tunnel has become.
Seven years of cracks and a pattern of Russian-segment leaks
The Zvezda PrK leak is not an isolated incident. The ISS has dealt with pressure anomalies on its Russian-built hardware before. In 2018, a separate leak was traced to the Soyuz MS-09 orbital compartment, a discovery that triggered its own investigation and forced crew members to locate and patch a small hole while controllers monitored cabin pressure from the ground. That earlier event established the procedural template NASA and Roscosmos now follow: isolate the affected area, move the crew to a safe location if needed, and coordinate repairs across the two agencies’ engineering teams. Contemporary logs on the station status blog captured how quickly teams pivoted from detection to containment during that incident.
In the aftermath of the Soyuz MS-09 leak, the partners also formalized how they would communicate about such anomalies. A subsequent joint statement laid out the framework for shared investigations, emphasizing transparency between NASA and Roscosmos even when the affected hardware is owned and operated by one side. That interagency architecture now shapes the response to the Zvezda tunnel cracks, from how inspection data is exchanged to how risk assessments are coordinated across mission control centers.
What separates the current situation is duration. The Soyuz MS-09 hole was a discrete event that was sealed relatively quickly. The Zvezda tunnel cracks have persisted for seven years and are now worsening. That trajectory puts pressure on long-range planning for the station, which NASA and its partners intend to operate through at least 2030. Every year the cracks remain open is another year of cumulative metal fatigue, thermal cycling, and micrometeorite exposure acting on a structure that was launched in 2000.
Neither agency has released detailed metallurgical data or imaging showing how the cracks have progressed since 2019. The public record consists of summarized blog updates and official statements rather than raw inspection results. Without that data, outside analysts cannot independently assess whether the repair work is restoring the tunnel to a safe baseline or simply buying time before the next rate spike. The lack of high-resolution imagery and stress analysis in the public domain also makes it difficult to compare the PrK damage to known failure modes in other long-lived space structures.
Unanswered questions about the PrK tunnel’s future
Several gaps in the public record remain open. First, the two-pounds-per-day figure has not been cross-checked through an independent sensor audit, at least not in any document available outside the agencies. NASA’s recent blog post confirms the increase but does not publish the underlying telemetry tables or describe how many pressure sensors feed into the estimate. Without those details, it is unclear how much margin of error is built into the reported leak rate and how quickly controllers would see further accelerations.
Second, the internal decision memos that authorized the crew’s movement to the Soyuz have not been made public, so the exact risk thresholds that triggered the shelter order are unknown. Mission rules typically specify combinations of leak rate, trend over time, and available repair options that justify moving a crew into a lifeboat. In this case, observers can only infer that the doubling of the leak, combined with ongoing cargo operations and the discovery of additional structural issues, crossed whatever line NASA and Roscosmos had agreed upon.
Third, Roscosmos announced it chose a more extensive repair after finding fresh problems, but neither agency has specified what those additional problems were or how they relate to the original crack pattern. Possibilities range from new fracture lines branching off existing cracks to deformation around welds or fasteners in the tunnel wall. The decision to expand the scope of work suggests the damage is more complex than a single, well-bounded flaw, yet the public descriptions remain vague.
The correlation between Progress 95 operations and the leak-rate jump is the most actionable thread to watch. If future cargo dockings produce similar spikes, mission planners will face a difficult choice: accept the risk, redesign docking procedures to reduce vibration loads on the Zvezda tunnel, or seal the PrK permanently and reroute traffic through other modules. Each option would carry operational costs. Slower or gentler dockings could constrain cargo schedules, while isolating the tunnel might reduce redundancy in the Russian segment’s internal pathways.
Longer term, the PrK situation feeds into broader questions about how to manage aging hardware on a station that is now well beyond its original design lifetime. The Zvezda module has been in orbit for more than two decades, experiencing thousands of day-night thermal cycles and countless minor impacts from orbital debris. The current leak may be a localized issue, but it also serves as a proxy for the overall health of the Russian segment and, by extension, the viability of extending ISS operations toward the end of the decade.
For now, NASA and Roscosmos continue to present a united front, emphasizing that the crew is safe and that the station maintains adequate pressure margins even with the elevated leak rate. Yet the combination of an aging structure, incomplete public data, and the need for increasingly invasive repairs ensures that the PrK tunnel will remain under scrutiny. How the partners choose to balance transparency, operational risk, and the desire to keep the ISS flying will shape not only the future of this particular module, but also public confidence in the station’s final years.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
