Centauri Dreams

As a quick follow-up to yesterday’s article on quantifying technosignature data, I want to mention the SETI Institute’s invitation for applicants to the Davie Postdoctoral Fellowship in Artificial Intelligence for Astronomy. The Institute’s Vishal Gajjar and his collaborators both in the US and at IIT Tirupati in India will be working with the chosen candidate to focus on neural networks optimized for processing image data, so-called ‘CNN architectures’ that can uncover unusual signals in massive datasets.

“Machine learning is transforming the way we search for exoplanets, allowing us to uncover hidden patterns in vast datasets,” says Gajjar. “This fellowship will accelerate the development of advanced AI tools to detect not just conventional planets, but also exotic and unconventional transit signatures including potential technosignatures.”

As AI matures, the exploration of datasets is a critical matter as these results from missions like TESS and Kepler are packed with both exoplanet data as well as stellar activity and systematics that can mislead investigators. Frameworks for sifting out anomalies should help us distinguish unusual candidates including disintegrating objects, planets with rings, exocomets and perhaps even megastructures and other technosignatures, all flagged by their deviation from our widely used transit models.

The data continue to accumulate even as our AI tools sharpen to look for anomalies. I can think of several Centauri Dreams readers who should find this work right up their alley. If you’re interested, you can find everything you need to apply for the fellowship here. The deadline for applications is March 15, 2025.

It’s one thing to talk about technology as we humans know it, but applying it to hypothetical extraterrestrials is another matter. We have to paint with a broad brush here. Thus Jason Wright’s explanation of technosignatures as conceived by SETI scientists. The Penn State astronomer and astrophysicist defines technology in that context as “the physical manifestations of deliberate engineering.” That’s saying that a technology produces something that is in principle observable. Whether or not our current detection methods are adequate to the task is another matter.

Image: Artist’s concept of an interesting radio signal from galactic center. But the spectrum of possible technosignature detections is broad indeed, extending far beyond radio. Credit: UCLA SETI Group/Yuri Beletsky, Carnegie Las Campanas Observatory.

A technosignature need not be the sign of industrial or scientific activity. Consider: In a new paper in The Astronomical Journal, Sofia Sheikh (SETI Institute) and colleagues including such SETI notables as Wright himself, Ravi Kopparapu and Adam Frank point out that the extinction of ancient megafauna some 12,800 years ago may have contributed to changes in atmospheric methane that fed into a period of cooling known as the Younger Dryas, to be followed by growing human agricultural activity whose effects on carbon dioxide and methane in the atmosphere would be detectable.

As a technosignature, that one has a certain fascination, but it’s not likely to be definitive in ferreting out extraterrestrials, at least not at our stage of detection technology. But Sheikh and team are really after a much less ambiguous question. We know what our own transmission and detection methods are. How far away can our own technosignature be detected? By studying the range of technosignatures we are producing on Earth, the authors produce a scale covering thirteen orders of magnitude in detectability, with radio still at the top of the heap. The work establishes quantitative standards for detectability based on Earth’s current capabilities.

We might, for example, use the James Webb Space Telescope and the upcoming Habitable Worlds Observatory to provide data on atmospheric technosignatures as far out as 5.7 light years away. That takes us interstellar, with that interesting system at Proxima Centauri in range. Let’s tarry a bit longer on this one. While carbon dioxide is implicated in manmade changes to Earth’s atmosphere, the paper points to other sources, zeroing in on one in particular:

…there are other atmospheric technosignatures in Earth’s atmosphere that have very few or even no known nontechnological sources. For example, chlorofluorocarbons (CFCs), a subcategory of halocarbons, are directly produced by human technology (with only very small natural sources), e.g., refrigerants and cleaning agents, and their presence in Earth’s atmosphere constitutes a nearly unambiguous technosignature (J. Haqq-Misra et al. 2022). Nitrogen dioxide (NO2), like CO2, has abiotic, biogenic, and technological sources, but combustion in vehicles and fossil-fueled power plants is a significant contributor to the NO2 in Earth’s atmosphere (R. Kopparapu et al. 2021).

And indeed nitrogen dioxide (NO2) is what the authors plug into this study, drawing on earlier work by some of the paper’s authors. Note the fact that biosignatures and technosignatures overlap here given how much work has proceeded on characterizing exoplanet atmospheres. It turns out that the wavelength bands that the Habitable Worlds Observatory will see best in its search for biosignatures are also those that include the NO2 technosignature, a useful example of piggybacking our searches.

But of course the realm of technosignatures is wide, including everything from the lights of cities to ‘heat islands’ (inferring cities), orbiting satellites, radio transmissions and lasers. I’m aware of no other study that combines the various forms of technosignature in a single analysis. If you start looking at the full range of technosignatures according to distance, you find objects on a planetary surface to be the toughest catch, with heat islands swimming into focus only from a distance as far as outer planetary orbits in the Solar System. The current technology with the most punch is planetary radar, whose pulses should be detectable as much as 12,000 light years away, although such a signal would be a fleeting and non-repeating curiosity.

SETI does find signals like that now and then. But precisely because they are non-repeating, we simply don’t know what to make of them.

Image: The maximum distances that each of Earth’s modern-day technosignatures could be detected at using modern-day receiving technology, in visual form. Also marked are various astronomical objects of interest. Credit: Sheikh et al.

Think back to the early days of SETI and ponder how far we’ve come in trying to understand what other civilizations might do that could get us to notice them. SETI grew directly out of the famous work by Giuseppe Cocconi and Philip Morrison that laid out the case for artificial radio signals in 1959, followed shortly thereafter by Frank Drake’s pioneering work at Green Bank with Project Ozma. Less known is the 1961 paper by Charles Townes and R. N. Schwartz that got us into optical wavelengths.

And while ‘Dysonian SETI’, which explicitly searches for technosignatures, is usually associated with vast engineering projects like Dyson spheres, the point here is that a civilization will produce evidence for an outside observer that will continue to deepen as that observer’s tools increase in sophistication. The search for technosignatures, then, actually grows into a multi-wavelength effort, but one that spans a vast range. Making all this quantitative involves a ‘detectability distance scale.’ The authors choose one known as an ichnoscale. Here’s how the paper describes it:

Using Earth as a mirror in this way, we can employ the concept of the ichnoscale (ι) from H. Socas-Navarro et al. (2021): “the relative size scale of a given technosignature in units of the same technosignature produced by current Earth technology.” An ι value of 1 is defined by Earth’s current technology. This necessarily evolves over time—for this work, we set the ichnoscale to Earth-2024-level technology, including near-future technologies that are already in development.

Considering how fast our detection methods are improving as we build extremely large telescopes (ELTs) and push into ever more sophisticated space observatories, learning the nature of this scale will become increasingly relevant. While we realized in the mid-20th Century that radio was detectable at interstellar distances, we’re now able to detect not just an intentional signal but radio leakage, at least from nearby stellar systems. That’s an extension of the parameter space that involves levels of power we have already demonstrated on Earth. The ichnoscale framework quantifies these signatures that will gradually become possible to detect as our methods evolve.

We see more clearly which methods are most likely to succeed. This is an important scaling because the universe we actually live in may not resemble the one we construct in our imaginations. Let me quote the paper on this important point:

…the focus on planetary-scale technosignatures provides very specific suggestions for which searches to pursue in a Universe where large-scale energy expenditures and/or travel between systems is logistically infeasible. While science fiction is, for example, replete with mechanics for rapid interstellar travel, all current physics implies it would be slow and expensive. We should take that constraint seriously.

And with this in mind we can state key results:

1. Radio remains the way that Earth is most detectable at ι = 1.
2. Investment in atmospheric biosignature searches has opened up the door for atmospheric technosignature searches.
3. Humanity’s remotely detectable impacts on Earth and the solar system span 12 orders of magnitude.
4. Our modern-day planetary-scale impacts are modest compared to what is assumed in many technosignature papers.
5. We have a multiwavelength constellation of technosignatures, with more of the constellation becoming visible the closer the observer becomes.

Let’s pause on item 4. The point here is that most notions of technosignatures assume technologies visible on astronomical scales, and indeed it is usually assumed that our first SETI detections, when and if they come, will involve civilizations vastly older and superior in technology than ourselves. Planets bearing technologies like those we have today are a supremely difficult catch, because the technosignatures we are throwing are tiny and all but trivial compared to the Dyson spheres and starship beaming networks we typically consider. And this point seems overwhelmingly obvious:

We should be careful about extrapolating current technosignatures to scales of ιTS = 10 (or even ιTS = 2) without considering the changing context in which these technologies are being developed, used, and (sometimes) mitigated or phased out (e.g., the recovery of the ozone hole; J. Kuttippurath & P. J. Nair 2017). As another example, we are becoming aware of the negative health effects of the UHI [urban heat index] (as detailed in, e.g., A. Piracha & M. T. Chaudhary 2022); thus, work may be done to mitigate the concentrated regions of high infrared flux discussed in Section 4.3.

Indeed. How many of the technosignatures we are producing are stable? Chlorofluorocarbons in the atmosphere are subject to adjustment on astronomically trivial timeframes. The chances of running into a culture that is about to realize it is polluting itself just before it takes action to mitigate the problem seem remote. So all these factors have to be taken into account as we rank technosignature detection strategies, and it’s clear that in this “multiwavelength constellation of technosignatures” the closer we are, the better we see. All the more reason to continue to pursue not just better telescopes but better ways to get ever improving platforms into the outer Solar System and beyond. Interstellar probes, anyone?

The paper is Sheikh et al., “Earth Detecting Earth: At What Distance Could Earth’s Constellation of Technosignatures be Detected with Present-day Technology?” Astronomical Journal Vol. 169, No. 2 (3 February 2025), 118 (full text). The Cocconi and Morrison paper is “Searching for Interstellar Communications,” Nature 184 (19 September 1959), 844-846 (abstract). The 1961 paper on laser communications is Schwartz and Townes, “Interstellar and Interplanetary Communication by Optical Masers,” Nature 190 (15 April 1961), 205-208 (abstract).