In recent years, the heavens have become ever more crowded. Not just with stars and galaxies, but with artificial hardware — communication satellites, imaging platforms, Internet-of-things nodes, and more. Among these, the mega-constellation built by SpaceX under the Starlink brand stood out for its ambition: to blanket the Earth in broadband connectivity from low-earth orbit. But now, a startling new discovery raises a grave concern for astronomy and our ability to study the universe. A team of scientists has found that second-generation Starlink satellites are emitting unexpectedly large amounts of unwanted radio-wave leakage, flooding frequency bands reserved for radio-astronomy and threatening to blind our radio telescopes.
This investigative article takes you through the full story: how the emissions were detected, why the findings matter deeply, what the implications are for science and society, and what may — or may not — be done about it.
1. The Discovery: Unexpected Signals from Space
In September 2024, researchers published a paper in the journal Astronomy & Astrophysics titled “Bright unintended electromagnetic radiation from second-generation Starlink satellites.” The first author is C. G. Bassa of the Netherlands Institute for Radio Astronomy (ASTRON). arXiv+2Aanda+2
Using data from the LOFAR (Low-Frequency Array) radio telescope network in Europe, the scientists observed that nearly all of the second-generation Starlink satellites they monitored were emitting unintended electromagnetic radiation (UEMR) at frequencies critical to radio astronomy. Aanda+2ResearchGate+2
Specifically:
-
The observed frequency bands included 10–88 MHz (low band) and 110–188 MHz (high band). arXiv+1
-
Within that range, signals were detected in the 40-70 MHz and 110-188 MHz windows, including some in the 150–153 MHz band that is reserved for radio astronomy. arXiv+1
-
The measured “spectral power flux density” (essentially the strength of the emission) for the second-generation satellites was up to 32 times higher than that of the first-generation satellites (after correcting for distance) in some bands. Astron+1
-
The study asserts the electric field strengths exceed both commercial electromagnetic-compatibility standards and those recommended by the International Telecommunication Union Radiocommunication Sector (ITU-R) for protecting the 150.05–153 MHz band for radio astronomy. arXiv+1
One striking figure from the ASTRON summary: compared to the faintest cosmic sources detected with LOFAR, the unintended emissions from these satellites are 10 million times stronger. Astron
The data came not from occasional passes but from systematic tracking: 97 second-generation Starlink satellites were observed during two one-hour sessions on July 19, 2024, for example, using the central six LOFAR stations. Aanda+1
In short: a previously under-appreciated source of radio-frequency “noise” in orbit has been quantified, and the magnitude is alarming.
2. Why This Matters: Radio Astronomy Under Threat
2.1 What radio telescopes measure
Radio astronomy is one of the most powerful ways we study the universe. Unlike visible-light telescopes, which capture photons in the optical band, radio telescopes observe extremely faint electromagnetic signals at long wavelengths — from cosmic phenomena such as pulsars, black holes, hydrogen lines tracing the early universe, and transient events like fast radio bursts. Because these signals are so faint and require long integrations, any interference — terrestrial or orbital — can destroy the measurement.
2.2 Protected bands & the silence requirement
To preserve this science, certain frequency bands have been reserved internationally for radio astronomy (via ITU-R) and other radio-sensitive science. For example, the 150.05–153 MHz band is internationally defined as a “primary protected” radio-astronomy band. The idea is that virtually no intentional transmissions should appear in that band so that telescopes can make ultra-sensitive measurements. The UEMR detected from Starlink satellites intrudes directly into those protected windows. Aanda+1
2.3 Why unintended satellite leakage is worst
There have been many discussions about optical light pollution from satellites (visible streaks across the sky) and reflections that can ruin long-exposure optical images. But the problem highlighted here is far more insidious: radio-frequency pollution from satellites whose electronics leak signals at frequencies that ground-based telescopes rely on. Unlike visible streaks that can sometimes be mitigated (by masking, post-processing, or scheduling), radio contamination can overwhelm the signal — the faint cosmic target simply disappears under the noise floor.
As the paper states: “Unlike the visible streaks satellites leave in telescope images, radio-wave pollution cannot be filtered out — it simply overwhelms sensitive instruments.” While that exact sentence is a summary rather than direct quote from the paper, it faithfully conveys the reported risk.
2.4 A growing problem
The scale of the problem grows rapidly because of the following facts:
-
SpaceX is launching second-generation Starlink satellites (v2-Mini and v2-Mini Direct-to-Cell) at high cadence. According to the ASTRON summary, about 40 new satellites per week are being launched in that generation. Astron
-
Other companies and mega-constellations — for example OneWeb, Amazon Kuiper — are planning or deploying large numbers of satellites, all of which could potentially contribute to the radio-noise floor.
-
Lower altitude orbits increase the strength of leakage signals as received on Earth simply because the distance is shorter; the paper notes that the newer satellites operate at lower altitudes (e.g., ~448 km, ~482 km, ~360 km) compared to the ~550 km for first generation. That amplifies the impact. arXiv
In effect: the rapidly expanding fleet of satellites increases both frequency coverage (more bands) and intensity (closer orbits) of radio noise. For radio-astronomy, this is equivalent to adding thousands of loudspeakers behind a microphone tuned to a faint whisper.
2.5 The threat to cosmic discovery
What is at risk? A lot. Here are some key science domains that depend on ultra-quiet radio frequency environments:
-
Epoch-of-Reionization studies: Tracing the first stars and galaxies by detecting faint hydrogen signatures at low-frequency radio bands. Contamination can wipe out these signals entirely.
-
Transient astronomy: Fast radio bursts (FRBs), pulsars, solar and planetary radio bursts — many of these occur at frequencies where leakage is now being detected. If satellites dominate the noise, detecting and localizing these events becomes much harder. The ASTRON and other studies warn that time-domain science is especially vulnerable. AstroBytes+1
-
Deep-sky surveys: Radio maps of the sky rely on extremely low background noise. Artificial emissions shift the baseline and force higher integration times (or reduce sensitivity).
-
Innovation spill-over: Historically, radio astronomy has led to breakthroughs (e.g., the discovery of cosmic microwave background fluctuations, pulsars, gravitational-wave electromagnetic counterparts). If our radio window is compromised, the innovation cascade could slow significantly.
In short: if we lose the “quiet sky” for radio astronomy, we risk losing our ability to go deeper, to detect the faintest cosmic signals, and to push the boundaries of discovery.
3. The Evidence: What the Study Shows
3.1 Satellite generations and comparisons
The study compares two main sets of satellites: first-generation Starlink (v1.0 and v1.5) and second-generation (v2-Mini and v2-Mini DTC, where DTC stands for Direct-to-Cell). The key findings:
-
In earlier work (2023), unintended emissions were detected from first-generation satellites in the 110–188 MHz band, with spectral power flux densities (S_ν) in the range of ~0.1 to 10 Jy (broadband) and up to ~500 Jy for narrowband features at 125, 135, 150 MHz. arXiv
-
In the new study (2024), the second-generation satellites show S_ν values ranging from 15 to 1,300 Jy in the 56–66 MHz band, and 2 to 100 Jy in the 120 and 161 MHz bands. arXiv+1
-
After normalising for distance (scaled to 1,000 km), the second-generation satellites emit up to 32 times stronger UEMR compared to the first generation, at least in certain bands. arXiv+1
-
The newer satellites are in lower orbits; this alone could make them appear brighter, but the distance correction shows they are intrinsically louder. Aanda+1
3.2 Frequencies and bands of concern
-
The UEMR extends below the 88 MHz low-band into frequencies previously unused, meaning the “noise footprint” is broader than before.
-
The 150.05–153 MHz band (reserved for radio astronomy) shows significant aggregate strength increases: roughly 15 dB and 7 dB higher for the v2-Mini and v2-Mini DTC types (compared to first gen). On a linear scale those correspond to factors of ~32 and ~5 respectively. Aanda+1
-
The electric-field strengths (when scaled to 10 m at 1,000 km) exceed ITU-R RA.769-2 thresholds for that band. Meaning: they are in a regime that regulatory frameworks intended to keep free of interference.
3.3 Scale of observations
-
In the two one-hour LOFAR sessions on July 19, 2024, 97 Starlink satellites (of second generation) were detected passing through the beam pattern. DNA: the beam-forming comprised 91 tied-array beams and tracked equatorial positions culminating near zenith (maximum elevation ~87.5°) to minimise distance. Astron+1
-
Each pass generated dynamic spectra: time vs frequency intensity maps, where the satellite signature is a bright track. These data lay bare both narrowband “combs” and broadband emission.
-
The study characterises features such as periodic “comb” spacing at 48.8 kHz, 65 kHz, 97.5 kHz for some satellites, and 50 or 150 kHz for others. These are spectral structures likely tied to internal electronics or switching-mode power supplies in the satellites. ResearchGate+1
3.4 Conclusions drawn by the authors
The authors categorise the emission as intrinsic (i.e., emitted by the satellite itself) rather than reflections of terrestrial signals (though earlier narrowband detections at 143.05 MHz may be radar reflections). They emphasise this is not simply an optical light-pollution issue — this interference is electromagnetic, broad in frequency, and increasing in magnitude. They contend the satellite operator (SpaceX) should be engaged to identify the hardware components that produce the UEMR and to design mitigation strategies for both current and future hardware. arXiv
4. Who’s Responsible and What Are Their Roles
4.1 Satellite operator — SpaceX/Starlink
As the owner and operator of the Starlink constellation, SpaceX bears responsibility for the engineering and deployment of the hardware, including mitigation of electromagnetic interference (EMI). Although the satellites are designed to transmit in Ku-band and Ka-band (~10 GHz and above) for user links, the unintended emissions in the ~40-200 MHz regime appear to come from other components (power electronics, telemetry, etc). The study’s authors encourage SpaceX to work with the radio-astronomy community to identify root causes and implement fixes.
4.2 Radio astronomy community
Institutions such as ASTRON, Observatoire de Paris (LESIA), Max Planck Institute for Radio Astronomy, and organisations linked to SKA (Square Kilometre Array) are actively tracking the interference. They monitor satellite emissions, quantify impact, and advocate for mitigation frameworks. For example, Observatoire de Paris published a news article summarising the study and emphasising that radio telescopes may be “blinded” if nothing is done. Paris Astronomy Research Center
4.3 Regulatory bodies
Entities like ITU-R define the frequency allocations and interference thresholds for radio astronomy and other services. The paper notes that while intentional transmissions are regulated, unintended electromagnetic radiation (UEMR) from satellites falls into a regulatory grey area — the existing frameworks may not fully cover space-based sources of leakage. This regulatory gap is a central concern. arXiv
4.4 Other satellite operators
Mega-constellations are not limited to Starlink. Amazon’s Kuiper, OneWeb, and others will bring thousands more satellites into orbit, possibly elevating the noise floor further. The study emphasises that unless mitigation becomes standard, the cumulative effect will worsen.
5. Implications & Deep Consequences
5.1 Immediate scientific impact
-
Reduced sensitivity: Radio telescopes rely on detecting extremely faint signals; elevated background noise forces longer integration times, reduces sensitivity, and may render some observations impossible.
-
Time-domain astronomy at risk: The detection of fast radio bursts, variable sources, and transient phenomena depends on clean data. If large constellations appear regularly in the beam, much of that science may be compromised.
-
Lost parts of the spectrum: The fact that emissions span and exceed the 150–153 MHz “protected” band means that entire swathes of the low-frequency spectrum may become unusable for unbiased science.
-
Increased cost & complexity: As contamination rises, astronomers may need to build more elaborate shielding, schedule around satellites (impractical at high numbers), or relocate to more remote sites — all increasing cost.
5.2 Long-term strategic consequences
-
Slowed discovery: If radio astronomy’s capabilities are reduced, breakthroughs in cosmology, astrophysics, planet detection, and SETI (search for extraterrestrial intelligence) may slow.
-
Science-policy conflict: The rise of commercial satellite constellations vs public-good science creates a policy tension. Space is increasingly a shared domain where commercial interests may compete with scientific ones.
-
Sky-as-public-resource: The night sky has long been treated as a public resource for humanity’s exploration; radio-frequency space may need similar recognition, but regulation lags.
-
Call for global framework: The study underlines the urgent need for international coordination and possibly more stringent regulation of UEMR from satellites before that noise floor becomes permanent.
5.3 Risks for innovation
Consider that radio astronomy has driven technologies we now use ubiquitously: radio receivers, antennas, signal processing, cosmological models, even GPS and wireless communications have roots in radio-science. If the radio frontiers become compromised, the downstream innovation pipeline may suffer.
6. What Could Be Done: Mitigation and Strategies
6.1 Engineering fixes
-
Better shielding and design: Satellite electronics (power supplies, telemetry systems) may be modified to reduce broadband leakage. The study suggests collaboration between manufacturers and astronomers to identify the offending components.
-
Lower emission hardware: Future generations of satellites might adopt stricter electromagnetic compatibility (EMC) standards, analogous to terrestrial electronics.
-
Orbit & operation changes: Satellites could operate in orbits or orientations that minimise emission toward Earth-based telescopes, though this is difficult at scale.
-
Software filtering and avoidance: While radio noise cannot simply be subtracted like optical streaks, telescopes could schedule observations during satellite absence or flag contaminated data — but this becomes less viable with tens of thousands of orbiters.
6.2 Regulatory and policy responses
-
Expand UEMR regulation: International bodies (e.g., ITU) may need to establish explicit limits for unintended emissions from space-based platforms, not just ground transmitters.
-
Constellation licensing conditions: Licensing agencies (e.g., FCC in the U.S., national regulators abroad) might require constellations to pass interference-tests and commit to mitigation before deployment.
-
Astronomy-industry partnerships: Establish formal frameworks where satellite developers collaborate with astronomical institutes to identify and mitigate impacts proactively.
-
Public-interest standards: Recognise low-frequency radio astronomy as a public good and give it protective status in spectrum planning, orbit-allocation, and launch-licensing regimes.
6.3 Practical steps for astronomers
-
Monitoring & mapping: Continued systematic measurement of satellite emissions (LOFAR, SKA prototype arrays) to build an empirical picture of contamination.
-
Data flagging & cleanup: Develop pipelines that identify satellite-noise tracks and flag or exclude contaminated data, though this is limited when the noise floor rises.
-
Site selection & instrumentation: Consider remote or radio-quiet zones, better shielding, and improved antenna design to cope with higher backgrounds.
-
Raise awareness: Communicate the risks to policymakers, funders, and wider public — the “silent noise” is less visible than optical streaks but no less destructive.
7. What’s Next: Open Questions and Challenges
-
Cause identification: Exactly which component(s) of the second-generation satellites produce the broadband leakage? The study asks for operator cooperation.
-
Cumulative effect: With tens of thousands of satellites planned, how will the aggregate effect scale? Some modelling is underway (e.g., future SKA-Low site impact). arXiv
-
International regulation: Will governments and international bodies move swiftly enough to set enforceable limits?
-
Commercial vs scientific priorities: How will operators balance connectivity goals against scientific externalities? Are there economic incentives to reduce UEMR?
-
Public awareness: Since the damage is invisible to naked eye (unlike streaks) how will the issue gain broad attention and pressure?
-
Mitigation vs remediation: For satellites already in orbit, how practical is retroactive mitigation? Can design changes on future launches compensate for existing noise?
8. A Window That May Close
The metaphor is stark: imagine trying to listen for a whisper in a stadium where someone has started broadcasting full-volume music. At present, radio telescopes try to hear whispers from the earliest stars, from black holes, from the empty spaces between galaxies. The invaders are steadily raising the noise floor of that stadium.
The second-generation Starlink satellites, by leaking emissions up to 32 times higher than their predecessors, are accelerating that loss of silence. The risk isn’t just a nuisance. It’s a systemic threat to our ability to study the universe.
The key takeaway: unless mitigation happens promptly and globally, we may reach a point where radio-astronomy loses its edge, not because telescopes fail, but because the sky itself becomes too loud.
9. Final Word: Balancing Progress and Preservation
Innovation in space communications is laudable. The ability to provide broadband Internet globally via low Earth-orbit constellations unlocks remote connectivity, disaster response, and global infrastructure. But such progress must come with responsibility — especially when that infrastructure occupies the night sky we share with all of humanity.
The leakage of radio waves from second-generation Starlink satellites is not just a technical hiccup. It is a wake-up call. If we value our cosmic heritage — our ability to listen to the faintest signals of the cosmos — then we must demand that humanity’s technologies in orbit respect that heritage.
It’s time for satellite operators, scientists, regulators, funders, and the public to act together. The silent sky is a resource — essential, fragile, and finite. Let us not lose access to it, not because of brightness in our skies, but because of noise we cannot hear.