On the matter of city lights as technosignatures, which we looked at on Friday, I want to follow up with Thomas Beatty’s work on the issue in the context of an assortment of nearby stars. Beatty (University of Arizona, Tucson) assumes Earth-like planets examined via direct-imaging by LUVOIR, a future space telescope in planning, or HabEx, a different architecture for a likewise powerful instrument. What he’s done is to take data from the Soumi National Polar-orbiting Partnership satellite to find the flux from city lights and the spectra of currently available lighting. He goes on to model the spectral energy distribution from such emissions as applied to exoplanet settings at various distances.
Why look at city lights in the first place? Because they’re another form of technosignature that may be within the realm of detection, and we’d like to find out what’s possible and what any results would imply. In particular, Beatty reminds us, the National Academies’ Exoplanet Science Strategy and Astrobiology Strategy reports are on record as recommending space-based, direct imaging that is capable of directly detecting emissions from habitable zone planets. This would obviously support biosignature searches but would also open up a hunt for technosignatures.
Technosignature searches can take place within the context of ongoing biosignature investigations on the same planets. Both LUVOIR and HabEx should be capable of this, generating data sets that can be scanned for both biological and technological returns. One area of investigation has been satellites — could we detect satellite constellations in orbit around an Earth-like planet? Large-scale photovoltaic arrays would show a different signature than vegetation on the surface. Various forms of pollution in atmospheres are within LUVOIR’s range, so the field is wide.
A lack of a specific technosignature is itself interesting, as it helps us begin to constrain the field. Just as we started searching for planets around Proxima Centauri by first ruling out gas giants of a certain mass, then Jupiter-class objects, then ice giants of Neptune size, we first learn what is not there and then can specify what remains to be searched for. The lack of SETI detections at radio or optical frequencies, for example, makes it less likely that technological civilizations are broadcasting powerful beacons aimed at us from stars near the Sun, thus paring down earlier possibilities.
But back to city lights, which Jean Schneider and colleagues first studied in 2010 (citation below). We’ve learned through the work of Avi Loeb and Elisa Tabor that artificial illumination from the nightside of Proxima b could be detected, though with great difficulty, by the James Webb Space Telescope. LUVOIR will up the ante and widen the range. Beatty points out that city lights are compelling because they would presumably be long-lived artifacts of a technological culture and would offer a unique spectroscopic signature that is unlike anything produced by natural processes.
Image: This is Figure 1 from the paper. Caption: Figure 1. The nightside of Earth shows significant emission from city lights in the optical. This is a composite, cloud-free, image of Earth’s city lights compiled using Day/Night Band observations taken using the Visible Infrared Imaging Radiometer Suite instrument on the Soumi National Polar-orbiting Partnership satellite (Roman et al. 2018). Searching for the emission from city lights is a compelling technosignature because it requires very little extrapolation from current conditions on Earth, should be relatively long-lived presuming an urbanized civilization, and offers a very distinct spectroscopic signature that is difficult to cause via natural processes. This places the emission from nightside city lights high on the list of potential technosignatures to consider (Sheikh 2020).
Beatty considers the detectability of city lights first as a function of stellar distance and the amount of a planet’s surface covered by urbanization on Earth-like planets around G-, K- and M-dwarf stars. He then calculates their detectability on planets orbiting stars within 10 parsecs of the Sun, and finally estimates detectability on two dozen known, potentially habitable planets around stars close to the Sun. The tables within this paper are worth scanning, but here are some of the highlights:
We learn that LUVOIR should be able to detect city lights on Proxima b at an urbanization level of 0.5% (10 times Earth’s). Lalande 21185 b, Luyten’s Star b and Tau Ceti e and f would show detectable emission from city lights at urbanization levels of 3% to 10% in LUVOIR imaging.
Detection of city lights should be easiest on M-dwarf planets, and Beatty notes in particular planets around Proxima Centauri, Barnard’s Star, and Lalande 21185, but he points out how quickly the habitable zone around this kind of star moves within the inner working angle (IWA) of LUVOIR with distance, making it beyond the capacity of the instrument.
Earth-analog planets around Sun-like stars can be imaged at greater distances, but the distance drives the minimum detectable urbanization fraction higher. Here Beatty suggests Alpha Centauri A and B, Epsilon Eridani, Tau Ceti and Epsilon Indi A as potential targets.
And this brings up memories of Isaac Asimov’s global city Trantor: What Beatty calls an ‘ecumenopolis’ — a planet-wide city — would be detectable at much larger distances. The author surveys 80 nearby stars on which such a city would be at least marginally detectable.
Thus the work moves from the study of Earth’s urbanization fraction (0.05%) up to an ecumenopolis, showing how detectability scales with the amount of planetary surface covered. The paper assumes 100 hours of observing time for generic Earth-class planets around stars within 10 parsecs. Earth itself would not be detectable by LUVOIR in this range, but planets around M-dwarfs near the Sun would show detection for urbanization levels of 0.4% to 3%. City lights on planets orbiting nearby Sun-like stars would be detectable at urbanization levels in the range of 10 percent.
From the paper:
The possibility of directly detecting technosignatures on the surfaces of potentially habitable exoplanets is thus starting to be in the realm of practicality. Perhaps unsurprisingly, the 15m LUVOIR A architecture would be the most capable observatory for detecting city lights on the nightsides of nearby exoplanets, though LUVOIR B [smaller than LUVOIR A) or HabEx with a starshade would also have significantly sized detection spaces. Much of this proposed capability has been spurred by the goals of characterizing the atmospheres of and detecting biosignatures on potentially habitable exoplanets, but it also would afford us the opportunity to search for other, technological, signs of life.
In short, we’re going to be looking hard at many of these planets within a few decades as we search for biosignatures. The same data may show technosignatures, the strength of which we need to examine to see what’s possible. We are simply defining the limits of the search.
The paper is Beatty, “The Detectability of Nightside City Lights on Exoplanets,” in process at Monthly Notices of the Royal Astronomical Society (abstract). The Schneider et al. paper is “The Far Future of Exoplanet Direct Characterization,” Astrobiology Vol. 10, No. 1 (22 March 2010). Abstract. Thanks to my friend Antonio Tavani for the pointer to this paper.
Just a reflection on becoming acquainted with the concept of techno-signature. Versus a bio-signature, it seems to reflect what a species on another planet or star system would do ( perhaps inadvertently) to make itself evident via examination across space. The Dyson spheres and Kardeshev scales appeared to be appropriate for societies diligent with technology over vast periods of time and elaboration, employing vast energy levels and constructions. How they spread over parsecs was a TBD. Visible across the depths of space, maybe even extra galactic, the odds of detection were better than locals who might have just discovered fire. What they would be doing with all that energy, however, is the best argument I can think of for sustainability in contrast. For if and when it crashes, it will crash big.
And in contrast to that the above discussion addresses concepts familiar to our own civilization such as street lamps, satellite constellations and LEDs.
In this instance, we are looking for markers indicative of beings that would be quite like ourselves. If they arrived or had stayed on Proxima c for quite some time then they would respond to its environment and challenges much like we would… Whether the native denizens would do business that way, well let’s get into that…
I remember speaking with people who had participated in the design of the Viking Lander; they had life detection discussions with Carl Sagan – and they would both become aggravated with each others’ approaches.
Carl Sagan was worrying about covering all the bases of his imagination and the engineers were exasperated by the contingency plans and payload penalties of devices that they doubted would ever accomplish a thing. One issue was photo data rates. A martian could walk right by and take a selfie by the lander and the existing instrumentation would never have stirred. Or they could have had a picnic event nearby after it deactivated…
There is not really a “side” to be on in this issue – unless at some later date we realize we missed on account of “if only we had…”. But with regard to civilizations employing street lights, let’s consider for a moment the social species on this planet with their hives, burrows
and ant hills. Do they employ interior illumination? Do they cope without it somehow? On a super earth with a thicker atmosphere and cloudier weather than ours, will telephone poles and lamp posts be practical? Life is possible without fire in our oceans, but will combustion in the atmosphere be a given? If 100 bars or 1000 bar pressure prevails at the surface, would another form of life come to ascendancy there or at a level higher and float around?
And then for satellite constellations surrounding planets in habitable zones about red dwarf stars, we should also ask our neighbors how they keep those systems stabilized. It’s no small issue placing a satellite around a Galilean satellite. Zones of stability will narrow down unless they employ active controls.
The implications of such possibilities make it harder to extrapolate what social conveniences, appliances and infrastructure might be developed vs. what we might have imagined for LGMs.
We should be aware that the illumination levels of cities are highly variable – from the type of illumination (incandescent, gas, LED, etc), amount of illumination (high vs low intensity, shaded vs unshaded, display vs functional), source density (full street illumination or partial), and lastly city density (European vs US east Coast, vs US West Coast) – this last of which is what we see in the posted image. North Korea is the classic example of low illumination (very little power available) and thus stands out as a dark area on any nightside image. Europe was probably rather dark in WWII as night bombing raids required blackout blinds at night. Megacities have high density and are equivalent to many smaller cities, and perhaps be the future of Earth cities based on historical trends). Conversely, energy efficiency, technological and cultural changes, and ecological concerns may result in a reduction of externally observable illumination.
Although a rather long way off, has anyone calculated the capabilities of a solar gravity lens telescope performance in this regard? What about the use of quantum interferometers that are being proposed that will potentially offer extremely large aperture telescopes on Earth?
.
How is the advent of LED lighting affecting Earth’s light emissions?
Actually the term Ecumenopolis originates with architect and city planner Constantinos Doxiadis, a.k.a. Doxiades father, because his son is also an important architect. There is an article in Wikipedia about the term. For me my first experience with an Ecumenopolis was Coruscant in The Phantom Menace
It seems to me that if an advanced civilization wanted a city so large it would cover an entire surface of a planet, it would be much easier to build a 3D structure in space to take advantage of the volume to surface area ratio. In that case one would look for a heat signature not light I think.
If we’re looking for Earth-like planets (those are the only ones for which we have evidence that life may occur in them) we’re looking for planets, we’re looking for roughly a 70/30% ocean-to-landmass ratio.
Probabilities alone indicate that we might find be able to find more instances of life in oceanworlds than we can find in Mars-like or Venus-like planets.
I’m not convinced that city lights can be a significant signal happening in sufficient frequency to make a big difference. Most cities could be under clouds, or under the ocean, and we would never see them.
City lights only make sense if we assume that apart from the features of the planet, we are looking for human-like civilizations, made by lifeforms who can’t see by night, need a lot of illumination, and build their largest cities often on land or above water, near the shore of their lakes and seas.
Slightly off topic but I have wanted to share this for some time.
I want to propose that a very rough back-of-the-envelope calculation
shows that even using quite generous estimates of
radio transmitting civilization lifetimes, because of deep time, the
distance to the nearest “radio-active” civilization is likely to be so
large that it may explain why we cannot detect them.
An interesting “upper bound” on prevalence of technical civilizations
can be estimated if we assume *every* star (out of 400 billion) in our
galaxy hosted a technical civilization that was “active” for 1000
years. But the catch is that we spread them out evenly over 12 billion
years: 400/12 leads to 33 stars birthed per year. Assuming the radio
technology in each is active for 1000 years, we get that 33,000
civilizations would overlap in activity over our own 1000 year
lifetime which seems quite promising! But, now lets assume our galaxy
has a cuboid shape of 100K by 100K by 3K light years so that it
contains 30,000 cubes of volume, each of which is 1000 LY on a
side. Assuming we spread the 33K civilizations evenly throughout the
galaxy leads to the conclusion that the nearest co-temporaneous
civilization is approximately 1000 light years away. We can send them
a signal, but our civilization will no longer be radio-transmitting by
the time they send a reply. In order to reduce the mean distance to
500 light years, we need 8 times (say 10 times) as many civilizations
so that they would need to be radio transmitting, on average, for 10K
years. In order to have a “vibrant” alive galaxy, we need clustering:
the civilizations need to be born at about the same time, or be
clustered together in space (galactic HZ) or be extremely long lived
(at least millions of years in age). This all makes a lot of assumptions, but not purposefully pessimistically: yet we get a situation where even assuming 33K civilizations because of the unfathomably large dimensions of space and time, we end up being quite far away from our alien neighbors.
To put your comment in context:
Why is the study even thinking about the capabilities of the LUVOIR telescope that is perhaps 15-20 years away in operation and might detect city illumination of an ecumenopolis (many times larger than the totality of Earth’s cities, and probably biosphere destroying for biological ETI) within ~30 ly rather than the more likely 1000+ ly that is often used as the nearest civilization (on its homeworld without stellar colonization)?
Exactly.
Well, if we assume street lighting and radio broadcasts are linked, and they seem to be here, then this idea is not that wide afield. Same general problem. Two constraints that could be examined for more favorable outcomes:
1. Twelve billion years might be an unnecessarily wide distribution over time within the galaxy. In that context, even the Earth has not been around that long – or the sun. The population of stars has in effect gone through generations too. The fainter ones still hanging around, but our G star of a later generation.
2. Why limit a civilization to 1000 years of radio activity?
The issue might be akin to saying that our civilization is long overdue
for abandoning the sail. It’s been around for millenia, after all. In our own instance, despite what communication advances we might achieve, if we were to send spacecraft into interstellar space within the next century and they would remain operable for over a millenium, then we would not be likely to abandon our long wavelength communication investments entirely. Low data rates seem to be an answer to low signal to noise ratios too.
Anyway, your case and model ought to be explored.
Being around for millenia is not the same as being radio active for millenia. That’s why I was careful to talk not about the longevity of the civilization, but just constrain myself to talking about their radio active subset.
If there were hypothetically a civilization on a tidally-locked planet around an M-class star, you would think they would have evolved on the night side and would thus not see the need for artificial lighting?
Perhaps we should be looking for artificial *shades* instead on these planets?
I’m not sure that city lighting is a long-lived phenomenon. It’s inefficient. Earth’s city lights may have less than a century left. Augmented reality glasses, once they’re good enough, are much more convenient than cell phones. If everybody has them, then the public parts of a city can be continually mapped by lidar at a tiny fraction of the energy cost of lighting (and a tiny fraction of the detectability from interstellar distances). A century is more than long enough for that much progress.
A summary of Berkeley’s SETI efforts:
https://alumni.berkeley.edu/california-magazine/summer-2021/seti-at-berkeley-turns-ears-to-stars