If you’ve been following exoplanet research over the last couple of years, you’ve definitely heard of K2-18b. Located 124 light years away in the constellation Leo, it’s attracted a lot of attention as it sits squarely in its red dwarf host star’s habitable zone, and measurements of the James Webb Space Telescope show its atmosphere is rich in carbon dioxide and methane. It’s one of the prime candidates for a “Hycean” world - one where a thick hydrogen-rich atmosphere covers a global liquid water ocean. It is such an intriguing target for Search for Extraterrestrial Intelligence (SETI) researchers that they turned two of the most powerful radio telescopes in the world to watch K2-18b’s system. A recent paper, available in pre-print on arXiv, shows that there is likely no artificial narrow-band radio signals that are equivalent to our technology level coming from the planet, despite millions of potential hits.
Capturing the data necessary to process this not only involved the Karl G. Jansky Very Large Array (VLA) in New Mexico, but also the MeerKAT radio telescope in South Africa. These are two of the most powerful radio telescopes on the planet, and having them coordinate like this on an observational campaign is exceedingly rare.
But it wasn’t just the physical hardware that was important for this work - the “data pipeline” as astronomers call the software filters and logic that come after the data is collected is equally as critical, especially in radio astronomy. Earth-based signals are the source of the vast majority of radio signals these telescopes pick up, and advanced filtering algorithms, such as the Commensal Open-Source Multi-Mode Interferometer Cluster system that the VLA used and the Breakthrough Listen User Supplied Equipment (BLUSE) system that MeerKAT used, are critical pieces of any modern-day radio astronomy program.
Fraser updates the debate about “Hycean” worlds.The logic of that filtering is still the responsibility of the humans in the loop, though, and the paper describes five different constraints they imposed on the data to screen for potential alien technosignatures. First was RFI masking - essentially they removed all data from signals that fell within frequency bands that were known to be heavily contaminated by terrestrial interference. If the aliens were talking on those channels, we’ll have to use some other method - like a radio telescope on the far side of the Moon - to hear them.
Doppler effects, like those that change how an ambulance sounds when it approaches or passes you, are even more prominent when the signal is passing between planets. Any signal with essentially no Doppler change was eliminated outright, as it could only have come from Earth. Perhaps the most debatable logical filtering choice was to eliminate all signals with a signal-to-noise ratio of less than 10 or more than 100. While this eliminated extremely weak false positives, as well as strong instrumental data artifacts typically only seen in one antenna, it could also have eliminated relatively weak actual signals.
Another filtering technique is to use multibeam analysis. In this instance, the telescopes formed coherent “beams” across the sky, with one pointing directly at K2-18b and the other pointing elsewhere. In these cases, a signal coming from the exoplanet would have appeared only in the beam pointed directly at it, while Earth-bound interference bleeds into multiple beams simultaneously. A final check, which wasn’t necessary due to the timing of the survey, is transit filtering. Any signal that appears from K2-18b should go away when the planet passes behind its parent star, but since it didn’t have such a “secondary transit” during the observational window, no such filtering was necessary.
Fraser discusses one potential solution to all the Earth-based radio noise - build a telescope on the Moon.In short, despite millions of potential signals throughout the observational window, none passed these filters. There were no definitive technosignatures in the narrowband radio spectrum from K2-18b. While that might sound disappointing, it’s exactly the kind of thing that science needs to advance. By thoroughly scanning the planet and finding nothing, they are able to place “upper bounds” on the power of a transmitter from that system - in terms of power it would be something equivalent to the collapsed Arecibo radar in Puerto Rico. If there is a civilization there, they certainly aren’t shouting at us with anything larger than that level of radio telescope.
Perhaps the most important result, though, is the proof of concept for their automated filtering system. Processing the millions of signals the two telescopes discovered by hand would have been next to impossible. So when even larger radio telescopes, like the Square Kilometer Array, come online, these techniques will be ready to help another survey make sense of the mass of data it collected. While K2-18b might be quiet today, we will continue to get much better at listening, if it ever does start to speak to us.
Learn More:
C. D. Tremblay et al. - A Narrowband Technosignature Search Toward the Hycean Candidate K2-18b Using the VLA and MeerKAT
UT - How the Evidence for Alien Life on K2-18 b Evaporated
