≡ Menu

BLC1: The ‘Proxima Signal’ and What We Learned

If we were to find a civilization at Proxima Centauri, the nearest star, it would either be a coincidence of staggering proportions — two technological cultures just happening to thrive around neighboring stars — or an indication that intelligent life is all but ubiquitous in the galaxy. ‘Ubiquitous’ could itself mean different things: Many civilizations, scattered in their myriads amongst the stars, or a single, ancient civilization that had spread widely through the galaxy.

If a coincidence, add in the time factor and things get stranger still. For only the tiniest fraction of our planet’s existence has been impinged upon by a tool-making species, and who knows what the lifetime of a civilization is? Unless civilizations can live for eons, how could two of them be found around stars so close? Thus the possibility that BLC1 — Breakthrough Listen Candidate 1 — was a valid technosignature at Proxima Centauri was greeted with a huge degree of skepticism within the SETI community and elsewhere.

Image: This picture combines a view of the southern skies over the ESO 3.6-metre telescope at the La Silla Observatory in Chile with images of the stars Proxima Centauri (lower-right) and the double star Alpha Centauri AB (lower-left) from the NASA/ESA Hubble Space Telescope. Proxima Centauri is the closest star to the Solar System and is orbited by the planet Proxima b, which was discovered using the HARPS instrument on the ESO 3.6-metre telescope. Credit: Y. Beletsky (LCO)/ESO/ESA/NASA/M. Zamani.

But it was a good idea to look, because you don’t know what you’re going to find until you take the data, which is why SETI happens. While a great deal of attention has focused on the Alpha Centauri stars as targets for a future space probe, little attention has been paid to them in SETI terms. The southern hemisphere sky was examined by Project Phoenix in the 1990s (202 main sequence stars) and a second search, conducted by David Blair and team, likewise used the Parkes radio telescope in Australia to cover another 176 bright southern stars in that decade. Neither of these searches took in Proxima Centauri, which is, after all, a very faint M dwarf.

The discovery of Proxima Centauri b, a habitable zone world, brought new attention to the dim star. When the Parkes Observatory was turned toward Proxima Centauri in 2019 as part of the activities of Breakthrough Listen, the system received a thorough examination. We learn in two new papers in Nature Astronomy just how thorough:

We searched our observations towards Prox Cen for signs of technologically advanced life, across the full frequency range of the receiver (0.7–4.0 GHz). To search for narrowband technosignatures we exploit the fact that signals from any body with a non-zero radial acceleration relative to Earth, such as an exoplanet, solar system object or spacecraft, will exhibit a characteristic time-dependent drift in frequency (referred to as a drift rate) when detected by a receiver on Earth. We applied a search algorithm that detects narrowband signals with Doppler drift rates consistent with that expected from a transmitter located on the surface of Prox Cen b…Our search detected a total of 4,172,702 hits—that is, narrowband signals detected above a signal-to-noise (S/N) threshold—in all on-source observations of Prox Cen and reference off-source observations. Of these, 5,160 hits were present in multiple on-source pointings towards Prox Cen, but were not detected in reference (off-source) pointings towards calibrator sources; we refer to these as ‘events.’

Image: This is Figure 4 from et al. (citations below). Caption: The signal of interest, BLC1, from our search of Prox Cen. Here, we plot the dynamic spectrum around the signal of interest over an eight-pointing cadence of on-source and off-source observations. BLC1 passes our coincidence filters and persists for over 2 h. The red dashed line, purposefully offset from the signal, shows the expected frequency based on the detected drift rate (0.038 Hz s−1) and start frequency in the first panel. BLC1 is analysed in detail in a companion paper. Credit: Smith et al./Breakthrough Listen.

In other words, a technosignature at Proxima Centauri, if actually there, should disappear depending on whether or not the telescope is pointing directly at the target system, which is how all but 5,160 hits were eliminated. The BLC1 signal made it through subsequent examination partly because it did not lie within the frequency range of local radio-frequency interference. The authors like to refer to it as a ‘signal of interest’ rather than a ‘candidate’ signal, but accept the BLC1 nomenclature because it is so widely adopted in coverage of the event.

BLC1 was a narrowband signal, which screened out natural astrophysical sources, and intriguingly, it persisted for several hours, much longer than would be accounted for by a passing aircraft or satellite. Moreover, it showed a drift rate that one would expect from a transmitter not on Earth’s surface, one that changed smoothly over time, “as expected for a transmitter in a rotational/orbital environment.” So you can see why it merited a deeper look.

As presented in the two Nature Astronomy papers, that second look has proceeded in the hands of Sofia Sheikh and colleagues at the University of California Berkeley. The scientists have examined archival observations of the Proxima Centauri system using an analysis thoroughly explained in the second of the two papers, one that included, in addition to drift rate study, a search for reappearances of the signal on other days and at other frequencies.

That involved searching for other signals near 982 MHz, where BLC1 appeared, looking first for signals with the same frequency and drift, and moving on to other frequencies (I’m simplifying mercilessly here — see the paper for the intricacies of the analysis). We wind up with a population of BLC1 look-alike signatures that, unlike BLC1, also appear when the telescope is not directly pointed at Proxima Centauri. The authors find every one of these to be caused by radio frequency interference, and determine that BLC1 is consistent with this population of look-alikes in terms of absolute drift rate, frequency and signal to noise ratio. Thus:

Using this procedure, we find that blc1 is not an extraterrestrial technosignature, but rather an electronically drifting intermodulation product of local, time-varying interferers aligned with the observing cadence.

That’s a useful and not unexpected finding, but the value of BLC1 is apparent. It has allowed scientists to develop a set of procedures for the analysis of technosignatures which were fully deployed here and explored in the papers. Sheikh and team have developed a technosignature verification framework built around this, the first signal of interest from Breakthrough Listen that required exhaustive investigation to rule out an alien technology. The value of that for future SETI work should be obvious:

…this signal of interest also reveals some novel challenges with radio SETI validation. It is well understood within the community that single-dish, on–off cadence observing could lead to spurious signals of interest in the case in which the cadence matches the duty cycle of some local RFI. The blc1 signal provided the first observational example of that behaviour, albeit in a slightly different manner than expected (variation of signal strength over position and time, which changed for each lookalike within the set). This case study prompts further application of observing arrays, multi-site observing and multi-beam receivers for radio technosignature searches. For future single-dish observing, we have demonstrated the utility of a deep understanding of the local RFI environment. To gain this understanding, future projects could perform omnidirectional RFI scans at the observing site, record and process the data with instrumentation with high frequency resolution such as the various BL backends, and then use narrowband search software such as turboSETI to obtain a population with which to characterize the statistics (in frequency, drift, power, duty cycle and so on) of local RFI.

The 10-part technosignature verification framework appears at the end of the Sheikh paper and summarizes both what BLC1 taught us through this analysis, but also how we can proceed more efficiently with persistent, narrowband technosignature searches in the future. I would say that’s a good outcome, one that moves the field forward thanks to this unusual detection.

The papers are Smith et al., “A radio technosignature search towards Proxima Centauri resulting in a signal of interest,” Nature Astronomy 25 October 2021 (full text); and Sheikh et al., “Analysis of the Breakthrough Listen signal of interest blc1 with a technosignature verification framework,” Nature Astronomy 25 October 2021 (full text).


Comments on this entry are closed.

  • Alex Tolley October 27, 2021, 13:58

    Yes, the more rigorous analysis should benefit SETI. However, it will also most likely eliminate all “signals of interest” quickly. Done quickly it could reduce the endless speculation about ETI, especially if very nearby. It might even shift the focus to looking for artifacts/probes in our system that have been around long enough to compensate for the time factor between civilizations, however long or short-lived.

    Wasn’t Sheikh also the person who came up with a set of key points for ETI artifacts as noted by Jason Wright? Any more rigorous approaches to plug the loopholes allowing “ETI in the gaps”

    • Paul Gilster October 27, 2021, 15:36


  • Geoffrey Hillend October 27, 2021, 17:08

    Wikipedia says that ten days before BLC1 that an intense optical flare accompanied by radio signals from Proxima Centauri. Our Sun’s radio frequency of solar flares is from 10 MHZ to 400 GHZ . How do we know the 900 MHZ signal did not come from Proxima Centauri or a more distance source passing through the Alpha Centauri system? Ibid. The main problem is that a technosignature radio waves do not have the power to reach us just as our techno radio signals are not strong enough to reach the nearest star system. A deliberate attempt to communicate with us or accidental still has to have some data or code to provide that an intelligence exists in order to for us to differentiate it from radio signals from stars.

    I eagerly await the JWST’s observation of the spectra of Proxima b. If there is no oxygen there, I doubt there will be anyone there to send a us a signal. Planets around Alpha Centauri A or B have yet to be proven.

    • Alex Tolley October 27, 2021, 21:37

      You are assuming any ETI is biological. They may be machine entities, or something else using machines or other means to communicate with em radiation.
      (But please don’t let this comment imply that I think there are robots on Proxima b or anywhere else in the system. I am quite sure that the team is correct that this is rfi)

      • Geoffrey Hillend October 29, 2021, 1:06

        That’s a good idea since robots don’t need oxygen and could use it as a base or humanoid ET’s could use it as a robot base. I am biased against machine EI becoming considered a living organism or alive.

  • Floyd mathews October 27, 2021, 19:06

    I guess most of us were probably expecting this finding that BLC-1 was not ETI, because Sophia Sheikh so indicated in her presentation on April 12, 2021 at the Breakthrough Discuss virtual conference, which I summarized in my post below.

    At least our astronomers hopefully learned a lot from BLC-1 about how to filter out misleading radio data?

    Also raises our hopes/expectations for future progress.

    Floyd Mathews April 12, 2021, 18:50
    I just watched today Sofia Sheikh’s presentation on Breakthrough Discuss in which she concluded that BLC-1 was RF Interference, not ETI.

    She made the following main points:

    1. The peculiar narrow frequency characteristics of BLC-1 exactly match a certain category of known terrestrial electronic clocks. However the precise source location of the presumed clock RFI at present remains unknown, but is definitely not the Proxima Centauri system.

    2. Papers elaborating on her presentation are now being written and peer reviewed for publication in the near future.

    3. Re-observations hopefully searching for a repeat of this signal by multiple radio telescopes are now scheduled for April 29, 2021.

    If you want to watch her presentation, you can look for the rebroadcast on the YouTube link which Paul posted in today’s article. It commences about 4 hours and 38 minutes into the rebroadcast and runs about 20 minutes.

  • kzb October 27, 2021, 19:27

    The inhabitants of Proxima b learned long ago, to modulate their radio output so that they wouldn’t be detected by this method?

  • Edwin October 27, 2021, 19:29

    Yes this has been a very interesting story to follow and a great update to read here.

    Here are a few related papers I have seen recently.

    The Search for Deliberate Interstellar SETI Signals May Be Futile

    Missions to and Sample Returns from Nearby Interstellar Objects

    No Transits of Proxima Centauri Planets inHigh-Cadence TESS Data

    Laser Communication with Proxima and Alpha Centauri using the Solar Gravitational Lens

    The Impact of the increasing number of satellites is going to be a big issue in the future too.

    Characterizing the All-Sky Brightness of Satellite Mega-Constellations

    Cheers Edwin

  • Dale October 27, 2021, 20:07

    So pleased the papers have been released! I’m not a scientist and so don’t fully understand the final analysis but does this definitely rule out the signal coming from Proxima Centauri? Is it possible when the telescope was pointing away that it was picking up peripheral parts of the signal bouncing off local features? As a guitarist myself,I did appreciate the Spinal Tap “turned up to 11” quote featured in the original paper!

    • Paul Gilster October 28, 2021, 5:20

      I don’t know about signal ‘bounce’, Dale, but as to the nature of the signal, Sofia Sheikh puts it this way: “Given a haystack of millions of signals, the most likely explanation is still that it is a transmission from human technology that happens to be ‘weird’ in just the right way to fool our filters. We still can’t say with 100 per cent certainty that BLC1 isn’t a signal from alien technology – but the probability that it is alien is now extraordinarily low.”

      I think that’s the best we can do. Loved your guitar reference, as a one-time jazz guitar guy myself.

      Read more: https://www.newscientist.com/article/2294772-promising-radio-signal-isnt-aliens-just-human-generated-interference/#ixzz7AZxSR4EP

  • ljk October 28, 2021, 13:03

    A Call To Action on Ensuring That Extraordinary Claims About ET Life Come With Extraordinary Evidence

    OCTOBER 27, 2021


    The global scientific search for signs of life beyond Earth has produced cutting-edge and paradigm-shifting science for several decades now, and it has clearly found eager audiences around the world. This search is a high-priority goal of NASA and other space agencies, as well as institutions, universities and companies.

    While the successes in this broadly defined field of astrobiology are legion, the field has also struggled with a problem that flows precisely from its high-impact subject.

    That problem is how to best keep its scientific claims evidence based and how to take into account all the myriad factors that can undermine the strength of a “finding.” And then comes the question of how to best communicate with the public the nature of the findings and all the caveats involved.

    There appears to be a widely-held view that some scientific claims and media reports about potential life beyond Earth have become not only a distraction in the field, but have served to undermine some public confidence in the endeavor.

    Full article here:


    This is how the authors introduce the paper:

    “Ours could realistically be the generation to discover evidence of life beyond Earth. With this privileged potential comes responsibility.”

    “The magnitude of the question, “are we alone?”, and the public interest therein, opens the possibility that results may be taken to imply more than the observations support, or than the observers intend. As life detection objectives become increasingly prominent in space sciences, it is essential to open a community dialogue about how to convey information in a subject matter that is diverse, complicated, and has high potential to be sensationalized.”

    “Establishing best practices for communicating about life detection can serve to set reasonable expectations on the early stages of a hugely challenging endeavor, attach value to incremental steps along the path, and build public trust by making clear that ‘false starts’ and ‘dead ends’ are an expected and potentially productive part of the scientific process.”

    Their “Confidence of Life Detection” (CoLD) scale image link:


    • Alex Tolley October 28, 2021, 20:31

      The author keeps conflating life with ETI. The probability of life ( from bacteria to advanced life) existing contemporaneously with us is incomparably greater than the emergence of ETI and even smaller probability of ETI being contemporaneous with us.

      While evidence of both needs to be evaluated carefully, this is much more important for ETI. Some fossil evidence for life on earth is controversial, but that is fine. But claiming the existence of a godlike civilization millions of years ago on earth would require extraordinary evidence. The same applies to ETI. We need to be very careful about such claims as the importance and social impact could be very great.

    • Mike Serfas October 29, 2021, 16:10

      There are some easy memes that take on an unwholesome life, and the one about extraordinary claims and extraordinary proof is certainly one of them. It is not so reliably and intransigently wrong as “Hanlon’s Razor”, but it is surely bad enough: it invites the reader (the more uncreative the better) to formulate a definition of “extraordinary” wholly out of his unexamined prejudices, and to use that as a criterion to dismiss research findings out of hand.

      The fact that most of those claims actually *are* bad and fail in terms of merely ordinary proof, needless to say, is not helping.

  • FrankH October 28, 2021, 14:29

    While expensive and more complicated, why can’t these searches be done with two independent radio telescopes on different continents (or four, to fully cover both hemispheres).

    This would reduce or eliminate interference like BLC1

    • Ron S. November 1, 2021, 20:15

      This is discussed in the paper since it is an obvious technique. They diplomatically describe it as “resource intensive”. SETI is not well funded and few observatories give the matter high priority. Neither is likely to change in the near future.

      The procedure to eliminate candidates due to RFI in its various manifestations is also resource intensive. I believe that the most notable outcome of their work on blc1 is that they have developed software tools to do a lot of the hard work to evaluate whether candidate signals are RFI or are of human technology origin.

  • Jeff Wright October 29, 2021, 1:06

    The Universe must have its little jokes….there is probably a type of degenerate stellar matter that perfectly replicates the first five minutes of an I LOVE LUCY episode-that will doubtless fake us out again…

    • Alex Tolley October 29, 2021, 11:21

      Not a crazy idea at all. 5 minutes of video with 360×240 pixels at 30 frames/second and 256 shades of gray = 2e11 random values. If binary, perhaps 1.6e12 random bits.

      With over 10^10 stars per galaxy and 10^10 galaxies there is the possibility that there is an object that by chance sends out the exact sequence needed. Like the Jorge Luis Borges infinite library there will be many other versions that are close enough to be considered a fragment, even a single frame from any of the 180 I Love Lucy episodes. It doesn’t even have to be that show, but rather any identifiable show or movie or image that we have accumulated. The odds of a random “binary” emission that happens to contain an identifiable image or video sequence must be quite good if we search hard enough. Seems like a task for a server farm in Google’s stables.

      • Ron S. October 29, 2021, 14:50

        Decades ago I remember working with early digital communications systems over analog channels and no error correction protocol. Noise interprets as random characters. Every so often the random sequence would be perfectly understandable short English phrases and sentences, despite the spelling errors, extra spaces, etc. Very spooky when you see it but completely expected.

        Intelligent messages and noise have a lot in common.

        • Jeff Wright November 4, 2021, 1:58

          That’s most AM radio…