SETI: New Tools for Screening Out Radio Interference

Two new techniques for examining interesting SETI signals come into view this morning, one out of Breakthrough Listen work at UC-Berkeley, the other from researchers working with the Five-hundred-meter Aperture Spherical radio Telescope (FAST), the so-called ‘Heaven’s Eye’ instrument located in southwest China. In both cases, the focus is on ways to screen SETI observations from disruptive radio frequency interference (RFI), which can appear at first glance to flag a signal from another star.

The Chinese work relies upon FAST’s array of receiving instruments, each acting as a separate ‘beam’ to cover slightly different portions of the sky. FAST’s currently operational L-band receiver array consists of 19 beams, to which researchers led by Bo-lun Huang (Beijing Normal University) apply a technique called MultiBeam Point-source Scanning (MBPS). Here the instrument scans the target star sequentially with different beams of the instrument, setting up the possibility of cross-verification and allowing researchers to identify local interference quickly and accurately.

The paper on this work points to the SETI ON-OFF strategy as a more conventional way to analyze a target star. In this case, the star is observed for a short time, followed by a different target six or more beamwidths away from the primary. These become the ‘ON’ and ‘OFF’ of the method, the assumption being that an authentic signal from another civilization would appear only in the ON set of observations. MBPS, on the other hand, can be used by any radio telescope with a multibeam receiver and requires the telescope to slew during the observation periods to provide ongoing comparisons between each beam.

Let me quote the paper on this:

…we are effectively adding new parameters and the observation data can thus be interpreted from different perspectives. The additional parameters introduced by the MBPS strategy include the duration of signals in a single beam, intensity variation of signals, and the difference in central frequencies of different beams which are the results of the observation method of the MBPS. With the three newly introduced parameters, we are then able to put in the most rigorous restrictions on the RFI/ETI identifications by confining the characteristics of an ETI/RFI signal in a new multi-parameter space.

Having run a re-observation campaign on TRAPPIST-1 using this strategy (this followed a set of observations taken in 2021), the team was able to retrieve all 16,645 received signals (!) as RFI. The authors’ confidence level in the technique is high:

We speculate that it would be exceedingly rare for the MBPS strategy to return any suspicious signals, even over the course of several years, because the types of false positives found by other strategies are easily identifiable with the MBPS strategy. However, when a genuine narrowband ETI signal does arrive on Earth, the MBPS strategy is capable of identifying it even amidst a substantial influx of RFI.

Image: An illustration shows how FAST receives radio waves emitted by distant pulsars, the rapidly rotating cores of dead stars. At left, a photo shows the huge telescope in Guizhou province. Can the new methods in the Bo-lun Huang paper help us weed radio interference out of signals from another civilization? Credit: China Daily.

At UC-Berkeley, Bryan Brzycki and team have been analyzing interstellar ‘scintillation,’ the refraction or bending of electromagnetic waves that pass through cold plasma in interstellar space. Rising and falling in amplitude, the waves interfere when they reach Earth by different paths. The phenomenon has been well studied through analysis of pulsars and other distant radio sources, and an obvious analog occurs in the twinkling of starlight created by Earth’s atmosphere. In the case of interstellar scintillation, Brzycki has come up with algorithms that can analyze narrowband signals for this effect, quickly selecting for those that show the phenomenon and thus are not local.

On first glance, this appears extraordinarily useful, as co-author (and Brzycki thesis adviser) Imke de Pater (UC-Berkeley) points out:

“This implies that we could use a suitably tuned pipeline to unambiguously identify artificial emission from distant sources vis-a-vis terrestrial interference. Further, even if we didn’t use this technique to find a signal, this technique could, in certain cases, confirm a signal originating from a distant source, rather than locally. This work represents the first new method of signal confirmation beyond the spatial reobservation filter in the history of radio SETI.”

Image: The Green Bank Telescope, nestled in a radio-quiet valley in West Virginia, is a major listening post for Breakthrough Listen. Credit: Steve Croft, Breakthrough Listen.

A useful tool indeed, though bear in mind that it proves useful only for signals originating more than 10,000 light years from Earth, for to produce the needed scintillation, the signal must do a lot of traveling. If we do make a SETI detection with the aid of scintillation, in other words, it will not be of a civilization we’ll be likely to converse with (unless, of course, we find a way someday to actually visit it).

The Brzycki paper dovetails nicely with the FAST work, as witness its discussion of the ON-OFF strategy discussed above. The italics below are mine:

…RFI can also appear in only ON observations. For example, RFI signals could exhibit intensity modulations that follow the observational cadence of 5 minutes per pointing, a false positive that would pass the directional filter. While we observe false positives like this in practice, having directional requirements still serves as an interpretable basis for determining candidates, which would induce follow-up observations for potential re-detection.

This begs the question: can we differentiate narrowband signals as RFI based on morphology alone? Since ETI signals must travel to us through interstellar space, are there effects that would be observable and sufficiently unique compared to RFI modulations?

Thus the important result: The effect of scintillation does indeed provide a way to single out RFI simply because no local interference will produce the effect. Indeed, as the authors note, ETI might well consider the presence of scintillation in an artificial, narrowband signal as an announcement: ‘we are here.’ Where this work points is to further analysis of radio emission – the authors single out polarization, which they say is only beginning to be studied in the SETI context. Who can doubt their conclusion?

Whether it is because certain effects are stochastic or because human radio emission exploits every facet of radio light possible for communication, extracting non trivial information from a radio signal’s detailed morphology has been and will remain difficult. We may need to push the limits of detectability along hitherto unexplored axes to discover the first technosignature.

The paper from FAST is Bo-lun Huang et al., “A solution to persistent RFI in narrowband radio SETI: The MultiBeam Point-source Scanning strategy,” currently available as a preprint. The paper on scintillation is Brzycki et al., “On Detecting Interstellar Scintillation in Narrowband Radio SETI,” Astrophysical Journal 17 July 2023 (full text).

The “Habitability” of Worlds (Part II)

If we ever thought it would be easy to tell whether a planet was ‘habitable’ or not, Stephen Dole quickly put the idea to rest when he considered all the factors involved in his study Habitable Planets for Man (1964). In this second part of his essay on habitability, Dave Moore returns to Dole’s work and weighs these factors in light of our present knowledge. What I particularly appreciate about this essay in addition to Dave’s numerous insights is the fact that he has brought Dole’s work back into focus. The original Habitable Planets for Man was a key factor in firing my interest in writing about interstellar issues. And Centauri Dreams reader Mark Olson has just let me know that Dole appears as a major character in a novel by Harry Turtledove called Three Miles Down. It’s now in my reading stack.

by Dave Moore

In Part I of this essay, I listed the requirements for human habitability in Stephen Dole’s report, Habitable Planets for Man. Now I’ll go over what we’ve subsequently learned and see how this has changed our perspective.

Dole, in calculating the likelihood of a star having a habitable planet, produced his own ‘Drake equation.’

Image: Dole’s ‘Drake Equation.’

Dole assigns the following probabilities to his equation: PHP=Nsub>S Pp Pi PD PM Pe PB PR PA PL:

Pp = 1.0, Pi = 0.81, PM = 0.19, Pe = 0.94, PR = 0.9, PL = 1.0, PB = 0.95 for a star taken at random, 1.0 if there is no interference with the other star in a binary system. He calculates that for stars around solar mass there is a 5.4% chance of having a habitable planet.

I’ll only summarize his calculations as this is not the primary thrust of this essay. Some of his estimates such as Pp = 1.0, the number of stars with planets, have held up well. Others need adjusting, but by far the biggest factors that determine the likelihood of a planet being habitable for humans are those he didn’t consider in depth.

Since Dole’s report, we’ve learned a lot more about the carbonate-silicate cycle and atmospheric circulation. The carbonate-silicate cycle provides a stronger negative feedback loop over a wider range of insolation than thought at the time of his report. Atmospheric and oceanic heat transport have been shown to work more efficiently also. This leads to a more positive assessment to the range of habitability. Planets with high axial tilts and eccentricities, which Dole had excluded, are now considered potentially habitable; and more importantly, there’s the possibility that tidally-locked planets around M-dwarf stars may be habitable. M-dwarf stars being the most common in the galaxy, this makes a big difference to the number of potentially habitable planets. Nsub>S, the mass range of stars, is now opened up. Pi, the range of inclination, is probably 1.0, and PD, the probability that there is a planet in the habitable zone, which he gave as 0.63 and is still a good estimate, is now extended to M dwarfs. And given that tidally locked planets are no longer excluded, PR, the rate of rotation is not a limiting factor.

On PM, Dole’s assumptions for the size of a habitable Earth-like world have held up well. His calculations on atmospheric retention and escape conclude that planets between 0.4 Earth mass and 2.35 Earth mass could be Earth-like. Planets below 0.4 Earth mass would lose their atmospheres. Planets above 2.35 Earth mass would retain their primordial Hydrogen and Helium atmospheres and become what we now call Hycean planets or Super-Earths.

This gives a range of surface gravities, assuming a composition similar to Earth’s, of between 0.68 and 1.5 G, which would mean from a gravitational perspective most of the range is within what humanity could handle. Dole puts the upper limit at 1.25 G based on mobility measurements made in centrifuges from that time. I would agree with him even though there are a lot of people walking around today with one and a half times their ideal weight. The limiting factor for high G is heart failure at an early age, a condition extremely tall people here on Earth suffer from. If you are a six-foot person on a 1.5 G world, your heart is pumping blood equivalent to that of a nine-foot person. In this case, people of short stature have a distinct advantage. A five-foot person would have the blood pressure equivalent of being seven foot six on a 1.5 G world and six foot three on a 1.25 G world.

However, when it comes to the frequency of Earth-sized worlds in the habitable zone, Dole’s guess at PM = 0.19 is probably too high even when we now include tidally-locked planets around red-dwarf stars. He, like the rest of us until recently, had no clue that sub-Neptunes and super-Earths would be the most frequently-sized planets in the habitable zone of a roughly Sol-mass star.

From our observations, Dole’s guess on orbital eccentricity, Pe, looks like it’s in the ballpark, again due to the inclusion of red-dwarf stars with their tidally circularized orbits. With a lot of these factors, though, slight changes in probability do not make a big difference in the frequency of habitable planets. The big differences come from those he didn’t consider.

Dole noted that water coverage on a planet could determine its habitability. He did not go over this in any detail, however, mainly I suspect because he had no information to go on. He didn’t include a term for it in his calculations. But, we do know from density determinations of transiting Earth-sized planets that there’s a significant possibility that a large percentage of them may be excluded due to being covered by deep oceans. This would mean, even if they had breathable atmospheres, they would not meet Dole’s criteria for habitability.

While Dole went carefully over the range of breathable atmospheres humans could tolerate, he essentially assigned a probability of 1.0 to the formation of this atmosphere once life appears on the planet, PL, and sufficient time has passed, PA, to which he arbitrarily assigns a period of 3 billion years. He made no consideration of how likely it would be for this process to go off the rails.

Yet, if you consider the range of possible atmospheric compositions and pressures on Earth-like planets, those that meet the requirements of human habitability are narrow. This is the one factor that is most likely to winnow the field with the possible exception of average water composition.

When considering what percentage of Earth-like planets could have a breathable atmosphere: Oxygen between 100 and 400 millibars, Nitrogen less than 2.3 bar, CO2 less than 10 millibars, and no poisonous gasses, we are helped by a natural connection of these parameters. Oxygen destroys most poisonous gasses. The Carbonate-Silicate cycle will draw down CO2 to low levels. With Nitrogen we note that Venus has 3 bars of Nitrogen. Earth has a similar stock, but most of it is either dissolved in the oceans or mineralized as nitrates. Mars still has a 2.6% by volume trace of its primordial Nitrogen atmosphere. This points to a certain consistency for terrestrial planets with regard to their Nitrogen stock; however, Oxygen to Nitrogen ratios do vary from star to star. Getting the level of Oxygen within breathable parameters is more problematic, though. It’s a reactive gas that disappears with time. I can see two possible pathways that can lead to a breathable atmosphere, one abiotic and one biotic.

On the abiotic front, there’s a robust mechanism available for generating Oxygen. If the planet is warm enough to have significant quantities of water vapor in the upper atmosphere or has a steam atmosphere, then photolysis and subsequent Hydrogen escape will result in the build-up of Oxygen.

Planets less massive than the Earth-like range lose their atmospheres. Planets more massive retain their primordial Hydrogen, which means any Oxygen resulting from photolysis will recombine to form water. Intermediate-sized planets, however, can build up Oxygen via Hydrogen escape.

How much it builds up depends on the balance of production and removal. The amount produced depends on stratospheric water vapor and UV levels. The rate of removal is determined by three main processes: Oxygen escape, which is dependent on planetary mass, magnetic field strength and the strength of plasma wind from its primary; chemical reaction with reducing gasses, which is proportional to the level of volcanic emissions; and the oxidation of exposed regolith due to volcanism and weathering, the first being proportional to the level of volcanism and the second being proportional to the planet’s temperature.

Abiotic Oxygen atmospheres are probably transitory in nature over geological time periods, but I do see sufficient Oxygen being generated at various stages in an Earth-like planet’s history. The first is from the time when a planet’s red-dwarf primary is sliding down its Hayashi track towards its position on the main sequence. Due to the star’s greater luminosity at this time, an Earth-like planet destined for the habitable zone will spend 100 million to a billion years with a steam atmosphere. Models of this process indicate it could lose up to several Earth oceans of water through photolysis and Hydrogen loss. The loss of an Earth ocean translates into roughly 300 bar of Oxygen, most of which, as with Venus, will finish up oxidizing the crust. If, however, the various factors balance out, so that when the planet’s steam atmosphere condenses as the star arrives at its main sequence position, the water fraction is sufficient to provide both oceans and continents, and the Oxygen production and removal hove balanced out to produce a breathable but non-toxic level of Oxygen, then we should get a habitable planet, albeit one with a highly oxidizing surface chemistry like Mars.

If this all sounds highly unlikely, you are probably right, but there are a lot of red dwarf stars in our galaxy.

Image: Artist’s impression of the ultracool dwarf star TRAPPIST-1 from the surface of one of its planets. We’re beginning to learn whether the inner worlds here have atmospheres, but will we find that any of the seven are habitable? Credit: ESO.

Oxygen generation through photolysis occurs anytime an Earth-like planet has a high level of water loss. Mars is thought to have lost an ocean of water corresponding to 1.4% of Earth’s ocean early in its history, which translates into a total partial pressure of 4.2 bar of Oxygen (under 1 G.) This Oxygen generation would have occurred over a long period, so the partial pressure at any given time was probably low; but you’ll notice that the mineralogy of Mars from around 4 billion years ago is highly oxidizing whereas Earth’s surface didn’t become oxidizing until 2.2 billion years ago.

Also an Earth-like planet suffering from runaway greenhouse such as Venus did two billion odd years ago would also experience a build-up in Oxygen.

If the presence of life in the galaxy is sparse, then this mechanism may result in more planets having Oxygen in their atmospheres than those that get it through biotic means, so Oxygen lines in the spectra of a planet’s atmosphere would not be a good indication that it harbors life.

We are familiar through descriptions of the history of life on how the biotic process leads to a breathable atmosphere. This has implications, however. To frame this, I’ll use a model in which planets become habitable at the rate of one per million stars starting nine billion years ago. (The figure I selected is arbitrary. You are welcome to adjust it and see what sort of results you get.) Given that star formation in our galaxy is about one star per year (star formation rates have varied over time but an average of one per year will suffice for this model), this will result in the total of 9000 planets that will be habitable to humans at some point in their lifetime. There may well be many more life-bearing planets than this, but this model is only interested in the ones that become habitable to humans.

If we assume these planets have a similar evolutionary track to Earth, then the youngest 5% of these will be at the prebiotic stage. Until about 2.2 billion years ago Earth was dominated by anaerobic life, so the next 20% will have anaerobic atmospheres full of toxic gasses. Hydrogen Sulfide in particular is lethal, killing at 1000 ppm. Intrepid explorers will have to live in sealed habitats with airlocks and go around on the surface in spacesuits. Does this meet your definition of habitable?

About 2.2 billion years ago on Earth, photosynthetic aerobes got the upper hand in Earth’s chemistry and the surface became oxidized with an atmosphere of 1-2% of oxygen. If their timeline is similar to Earth’s, then 20% of these planets would fit this condition.

These planets would be a far more pleasant place to explore. Toxic gasses would be removed by the Oxygen. You could probably go around with just an oxygen concentrator on your back feeding a tube to your nose. Habitats wouldn’t need airlocks; double doors would do. How would you classify these planets?

Then 500 million years ago Earth became fully habitable when the Oxygen concentration crossed 15% and the air became breathable. This period represents 5% of the sample. However, there’s a side effect to this. Oxygen is not very soluble in water and O2 concentrations fall off rapidly with distance. This is why the macroscopic lifeforms from the Pre-Cambrian age (>500 mya) were either flat leaf-like shapes or sponges, both of which give short diffusion distances throughout the organism. Once the oxygen concentration rose, however, lifeforms could develop thickness, and with thickness, they could develop organs such as hearts and circulatory systems, which could then circulate an oxygenated fluid throughout their bodies. A breathable atmosphere allows for the development of complex macroscopic life.

And, over time, complex macroscopic life gives rise to the second side effect of breathable Oxygen levels – sapience. This has often been considered a rare possibility, a fortuitous combination of circumstance, and in the Drake equation it is assigned a low fractional value, but the idea that intelligent life is rare and unique derives from our historical and religious concept that mankind is something unique and apart from the animal kingdom. However, studies show a steady increase in encephalization over time and its widespread occurrence in different phyla and classes: octopi in the mollusks, parrots and corvids in the birds, and dolphins, elephants and apes in the mammals.

Varying levels of communication signaling have been found in numerous species. Just recently, a troop of Chimpanzees has been found to have a 390-word vocabulary constructed by combining grunts and chirps in various sequences. It therefore seems that our ability with language is merely a development of existing trends rather than something that came out of nowhere. And language is the abstract representation of an object or action, so the manipulation of language leads to abstract reasoning.

Encephalization is a tradeoff between the energy consumption of neurons and the benefits they produce in reproductive fitness. Increasing the number of neurons in an organism is easy. A simple mutation in the precursor cells allowing them to divide one more time will do this; however, organizing those extra neurons into something useful enough to justify their extra metabolic cost is a lot more difficult. But increases in neural complexity can lead to more complex behaviors, which can increase fitness or allow the creature to colonize new niches. In addition, neurons, over time, have evolved to become more efficient. Moore’s law operates, but with a doubling time on the order of 100 million years. Parrots’ neurons are both smaller and three times more energy efficient than human ones. So, not only does encephalization increase with time, but the tradeoff moves in its favor. However, like any increases in biological complexity and sophistication, this does take time.

This points to the conclusion that on planets habitable to humans, the evolution of sentience is not so much a case of if, but when.

An atmosphere breathable to humans is also flammable over most of its range, so a good proportion of these sapients would have access to fire allowing smelting technology to develop. What the model I used implies is that 50% of habitable planets will by now have had intelligent life forms evolve on them, a majority of which could develop technology.

I would support this argument by applying the Law of Universality that states that no matter where you are in the universe the laws of nature operate in the same way. This means that a planet like Earth would produce intelligent life forms. There is a certain contingent element in evolution, so the timing and the resulting life forms would not be identical; however, the broad driving forces of evolution would produce something similar. This can be seen in the many cases of convergent evolution that have occurred on Earth. How different from Earth a planet has to be before it stops producing intelligent life forms is a matter of conjecture, but if these changes cripple the evolution of intelligent lifeforms, there’s a good chance they cripple the formation of a breathable atmosphere.

What these intelligent life forms would do to their planet over the eons is a matter of speculation, but if for some reason intelligent life did not arise, then complex life could thrive and the planet would be habitable for another billion years or more – depending on the star’s spectral type – before the star’s increasing luminosity sets off a runaway greenhouse. This means that of the planets that are habitable for humans at some stage in their life approximately 15-25% will be habitable at any given time. (The upper bound assumes that there are a high proportion of them around lower mass stars with longer lifetimes.)

If, however, intelligent life develops on planets as a matter of course, then the model indicates that for every habitable planet we have now (5% of the total) approximately ten planets had intelligent lifeforms at some stage in their history (50% of the total.) And if intelligent life is a side effect of habitability, then there will be a correlation between the number of habitable planets and the number of exosolar technological civilizations in our galaxy. So, in an inversion of the usual order of things, we can estimate the number of planets habitable for humans from the number of alien civilizations in the galaxy. The model I’ve been using points to them being within an order of magnitude of each other.

Adding in the fact that we have no information on the evolution of intelligent life on non-habitable planets, then calculating the number of habitable planets from evidence of alien civilizations is an upper bound. On the other side of the scales, there’s the number of planets that are habitable through abiotic means. Planetary atmospheric spectra within the next couple of decades may give us some indication of this. If, however, we use Hanson’s estimate where he deduces that from the lack of evidence of alien civilizations in our galaxy that the number of technological life forms is just one – us – then this would also point to the number of habitable planets in our galaxy being just one: Earth.

As a final point I would like to add that while I have not done a full literature search, I have read widely in this field and have not come across as rigorous consideration as Dole’s work on defining habitability for humans and considering the likelihood of finding planets that match that criterion. The field’s general mindset seems to focus on finding the conditions upon which life arises; then it just assumes evolution will automatically lead to a habitable planet for humans. We have learned a lot since Dole wrote his paper, but there does not seem to have been much reexamination of the topic. It is perhaps time we applied our minds to it.


Stephen Dole, Habitable Planets For Man, The Rand Corporation, R414-R

Dave Moore, “’If Loud Aliens Explain Human Earliness, Quiet Aliens Are Also Rare’: A review”

Robin Hanson, Daniel Martin, Calvin McCarter, Jonathan Paulson, “If Loud Aliens Explain Human Earliness, Quiet Aliens Are Also Rare,” The Astrophysical Journal, 922, (2) (2021)

The “Habitability” of Worlds (Part I)

Dave Moore is a Centauri Dreams regular who has long pursued an interest in the observation and exploration of deep space. He was born and raised in New Zealand, spent time in Australia, and now runs a small business in Klamath Falls, Oregon. He counts Arthur C. Clarke as a childhood hero, and science fiction as an impetus for his acquiring a degree in biology and chemistry. Dave has kept up an active interest in SETI (see If Loud Aliens Explain Human Earliness, Quiet Aliens Are Also Rare) as well as the exoplanet hunt. In the essay below, he examines questions of habitability and how we measure it, issues that resonate in a time when we are preparing to evaluate exoplanets as life-bearing worlds and look for their biosignatures.

by Dave Moore

In this essay I’ll be examining the meaning of the word ‘habitable’ when applied to planetary bodies. What do we mean when we talk about a habitable planet or a planet’s habitability? What assumptions do we make? The first part of this essay will look into this and address the implications that come with it. In part two, I’ll focus on human habitability, looking at the mechanisms that could produce a habitable planet for humans and what this would imply.

If you look at the Wikipedia entry on habitable planets, the author implies that “habitability” refers to the ability of a planetary body to sustain life, and this is by far the most frequent use of the term, particularly in the literature of popular science articles.

Europa has sulfate deposits on it, which indicates that its surface is oxidizing. If the hydrothermal vents in the moon’s subsurface ocean are like those on Earth, they would release reducing gases such as H2S, and Methane. A connection between the two would provide an electrochemical differential that life could exploit. So it’s quite plausible that Europa’s ocean could harbor life, and if it does, would this now make it a “habitable” moon? If we find subsurface Methanogens on Mars, does Mars become a habitable planet? Traces of Phosphine in Venusian clouds point to the possibility of life forms there. If that’s so, would Venus now be considered habitable?

Andrew LePage on his website is more careful in defining what a habitable planet is. On his Habitable Planet Reality Check postings, he has the following definition:

…the best we can hope to do at this time is to compare the known properties of extrasolar planets to our current understanding of planetary habitability to determine if an extrasolar planet is “potentially habitable.” And by “habitable,” I mean in an Earth-like sense where the surface conditions allow for the existence of liquid water – one of the presumed prerequisites for the development of life as we know it. While there may be other worlds that might possess environments that could support life, these would not be Earth-like habitable worlds of the sort being considered here.

By Andrew’s definition, a habitable planet is first a body that can give rise to life. He then narrows it by adding that the type of life is “life as we know it,” which is life that needs an aqueous medium to evolve. If life evolved in some other medium, say Ammonia, then this would be life as we don’t know it; and the planet would not be classified as habitable. But this is not the only definitional constraint he makes. The planet must also be Earth-like in a sense that its surface conditions allow for liquid water. Europa would be excluded even if it had life in its oceans as its surface conditions do not allow for liquid water. His definition also implies that the planet must be in the habitable zone as defined by Kopparapu, which is thought to be the zone of insolation that allows for surface water on “Earth-like” planets. Would an ocean world with an ocean full of life fit his definition of habitable? Would a Super-Earth with a deep Hydrogen atmosphere (sometimes called a Hycean world) outside the habitable zone but with both oceans and continents and a temperate surface at moderate temperature be habitable? I do note however that his definition does not include human survivability as a requirement because elsewhere in his post he talks about the factors that have kept Earth habitable over billions of years, and Earth’s atmosphere has only been breathable to humans over the last 500 million years.

I’m not picking on Andrew in particular here; he has put more thought into the matter of defining habitability than most. Why I am using him as an example is to show just how fraught defining habitability can be. It’s a word that is bandied about with a lot of unexamined assumptions.

This may seem picayune, but the study of life on other worlds has very little data to rely on, so hypotheses are made using logical inference and logical deduction. And if your definitions are inexact, sliding in meaning through your logical process, then you are likely to draw invalid conclusions. Also, if the definition of habitable is that of a planet that could have life evolve on it, why include this arbitrary set of exclusions?

The answer becomes obvious from reading articles in the popular press. A habitable planet is not just one that is life-bearing, but a planet in which life gives rise to conditions that may be habitable for humans.

The assumption that life leads to human habitability is strongly ingrained from our historical experience. By the early 19th century, it was known that oxygen was required to survive and plants produced oxygen, hence the idea of life and human habitability became intertwined. Also, our experience of exploring Earth strongly influenced our perception of other planets. We found parts of Earth hot, parts cold, others wet and others dry. Indigenous inhabitants were almost everywhere, and you could always breathe the air. And this mindset was carried over to our imaginings of planets. They would be like Earth, only different.

For instance, H. G. Wells, an author known for applying scientific rigor to his stories, in The First Men in the Moon (1901), postulates a thin but breathable atmosphere on the moon and its native inhabitants. This is despite the lack of atmosphere on the moon being known for over a 100 years prior. Such was our mindset about other planetary bodies. Pulp SF before WWII got away with swash-buckling adventures on pretty much every body in the solar system without the requirement for space suits. Post WWII, until the early sixties, both Venus and Mars were portrayed as having breathable atmospheres, Mars usually as a dying planet as per Bradbury, Venus as a tropical planet as for example in Heinlein’s Between Planets (1951.)

When the first results from Mariner 2 came back in 1962 showing the surface of Venus was hot enough to roast souls, there was considerable resistance in the scientific community to accepting this and much scrambling to come up with alternative explanations. In 1965 Mariner 4 flew by Mars showing us a planet that was a cratered approximation of the moon and erased our last hopes that the new frontiers in our solar system would be anything like the old frontier. Crushed by what our solar system had served up, we turned to the stars.

Our search for life is now two-pronged: the first part being a search for signals from technological civilizations, which we regard as a pretty good indication of life; the second being the search for biomarkers on exosolar planets. We’re searching for biomarkers because, in the near future, characterizing exosolar planets will be by mass, radius and atmospheric spectra. Buoyed by our knowledge of extremophiles, we continue to search the planets and moons of our solar system for signs of life, but now it is in places not remotely habitable by humans. If the parameters for the search for life touch on habitable conditions for humans, they are purely tangential. These two elements once fused together in our romantic past have now become separate.

This divergence has led to a change in goals to the search for life. We look now for the basic principles that govern the emergence of life and under what conditions can life evolve and/or allow for panspermia? This leads to the concept of planetary habitability being secondary. Life, once evolved, in its single-celled form, is tough and adaptable, so it is likely to continue until there’s a really major change in the state of a planet; habitability is a parameter of life’s continuity, not its origins. So when describing planets, terms like life-potential or life-bearing become more pertinent. This latter term is now starting to be used in preference to the description habitable.

If we now look at the other fork, the idea of habitability when applied to humans, we note that the term has been used in a loose sort of way since the 17th century. Even the idea of the habitable zone was first raised in the 19th century, but it was Stephen Dole with his report, Habitable Planets For Man, under the auspices of the Rand Corporation in 1964 that put a modern framework to it by precisely defining what a habitable planet was for humans. The book can be downloaded at the Rand site.

This report has held up well considering it was written at a time (1962) when Mercury’s mass had not been fully established and Venus’s atmosphere and surface temperature were unknown.

Image: PG note: Neither Dave nor I could find a better image of the cover of the original Dole volume than the one above, but Stephen Dole’s Planets for Man was a new version of the more technical Habitable Planets for Man, co-authored by Isaac Asimov and published in 1964. If you happen to have a copy of the earlier volume and could scan the cover at higher resolution, I would appreciate having the image in the Centauri Dreams files.

Dole first defines carefully what he means by habitability (material omitted for brevity):

“For present purposes, we shall enlarge on our definition of a habitable planet (a planet on which large numbers of people could live without needing excessive protection from the natural environment) to mean that the human population must be able to live there without dependence on materials bought from other planets. In other words, a planet that is habitable can supply all of the physical requirements of human beings and provide and environment in which people can live comfortably and enjoyably…”

You’ll note that Dole’s definition contains echoes of the experience of American settlement where initial settlement is exercised with minimal technology and living off the land. There is emphasis on self-sustainment. It’s the sort of place you’d send an ark ship to.

I take a view of habitability as more of a sliding scale on how much technology you need to survive and live comfortably. On some parts of Earth, the level of technology needed to survive is minimal: basic shelter, light clothes and a pair of flip-flops will do the job. Living at the South Pole is a different story. You must have a heated, insulated station to live in, and when you venture outside, you need heavily insulated clothing covering your entire body and goggles to prevent your eyeballs from freezing. Move to Mars and you need to add radiation protection and a pressurized, breathable atmosphere. The more hostile the environment the more technology you need. By stretching the definition, you could say that an O’Neill colony makes space itself habitable.

I contrast my definition to Dole’s to show that even when dealing with what makes a planet “habitable for humans” you can still get a significant variation on what this entails.

Dole does however itemize carefully the specific requirements necessary to meet his definition. They are:

Temperature: The planet must have substantial areas with mean annual temperatures between 32°F and 86°F. This is not only to meet human needs for comfort, but to allow the growing of crops and the raising of animals. Also seasonal temperatures cannot be too extreme.

Surface gravity: up to 1.5 g.

Atmospheric composition and pressure: For humans, the lower limit for Oxygen is a partial pressure of 100 millibars, below which hypoxia sets in. The upper limit is about 400 millibars at which you get Oxygen toxicity, resulting in things like blindness over time. For inert gasses, there is a partial pressure above which narcosis occurs. This is proportional to the molecular weight of the molecule. The most important of these to consider is Nitrogen, which becomes narcotic above a partial pressure of 2.3 bar. For CO2, the upper limit is a partial pressure of 10 millibars, above which acidosis leads to long term health problems and impaired performance. Most other gasses are poisonous at low or very low concentrations.

Image: Original illustration from Dole’s Report. You may notice the lower level of O2 set at 60 mm Hg. This is the blood level minimum not the atmospheric minimum. There is a 42 millibar drop in O2 partial pressure between the atm. and the blood.

Other factors he considered were having enough water for oceans but not enough to drown the planet, sufficient light, wind velocities that aren’t excessive or too much radioactivity, volcanic activity or meteorite in-fall.

Dole then went on to discuss general planetology and how stellar parameters would affect habitability—something we now know in much greater detail–and he finishes up by calculating the likelihood of a habitable planet around the nearest stars in a manner similar to the Drake equation.

You will notice that these requirements listed bear little resemblance to the parameters used when discussing habitability with regard to life. The two have gone their separate ways.

Using Dole’s report as a basis for examining the habitability of a planet, in Part II of this essay, I will note how our current state of knowledge has updated his conclusions. Then I will look at how you could produce a planet habitable for humans and the consequences of those mechanisms.


Wikipedia Planetary Habitability Definition

Andrew LePage: Habitable Planet Reality Check: TOI-700e

Manasvi Lingam, A brief history of the term ‘habitable zone’ in the 19th century, International Journal of Astrobiology, Volume 20, Issue 5, October 2021, pp. 332 – 336.

Stephen Dole, Habitable Planets For Man, The Rand Corporation, R414-R

Reducing the Search Space with the SETI Ellipsoid

SETI’s task challenges the imagination in every conceivable way, as Don Wilkins points out in the essay below. A retired aerospace engineer with thirty-five years experience in designing, developing, testing, manufacturing and deploying avionics, Don is based in St. Louis, where he is an adjunct instructor of electronics at Washington University. He holds twelve patents and is involved with the university’s efforts at increasing participation in science, technology, engineering, and math. The SETI methodology he explores today offers one way to narrow the observational arena to targets more likely to produce a result. Can spectacular astronomical phenomena serve as a potential marker that could lead us to a technosignature?

by Don Wilkins

Finite SETI search facilities searching a vast search volume must set priorities for exploration. Dr. Jill Tarter, Chair Emeritus for SETI Research, describes the search space as a “nine-dimensional haystack” composed of three spatial, one temporal (when the signal is active), two polarization, central frequency, sensitivity, and modulation dimensions. Methods to reduce the search space and prioritize targets are urgently needed.

One method for limiting the search volume is the SETI Ellipsoid, Figure 1, which is reproduced from a recent paper in The Astronomical Journal by lead author James R. A. Davenport (University of Washington: Seattle) and colleagues. [1]

Image: This is Figure 1 from the paper. Caption: Schematic diagram of the SETI Ellipsoid framework. A civilization (black dot) could synchronize a technosignature beacon with a noteworthy source event (green dot). The arrival time of these coordinated signals is defined by the time-evolving ellipsoid, whose foci are Earth and the source event. Stars outside the Ellipsoid (blue dot) may have transmitted signals in coordination with their observation of the source event, but those signals have not reached Earth yet. For stars far inside the Ellipsoid (pink dot), we have missed the opportunity to receive such coordinated signals. Credit: Davenport et al.

In this approach, an advanced civilization (black dot) synchronizes a technosignature beacon with a significant astronomical event (green dot). The astronomical event, in the example, is SN 1987A, a type II supernova in the Large Magellanic Cloud, a dwarf satellite galaxy of the Milky Way. The explosion occurred approximately 51.4 kiloparsecs (168,000 light-years) from the Sun.

Arrival time of the coordinated signals is defined by a time-evolving ellipsoid, with foci at Earth (or an observation station within the Solar System) and the source event. The synchronized signals arrive from an advanced civilization based on the distance to the Solar System or other system with a technological system (d1), and the distance from the advanced civilization to the astronomical event (d2). Signals from civilizations (blue dot) outside the Ellipsoid coordinated with the source event have not reached the Solar System. Stars inside the Ellipsoid (pink dot) but on line between the advanced civilization and the Solar System will not receive the signals intended for the Solar System. However, the advanced civilization could beam new signals to the pink star and form a new Ellipsoid.

The source event acts as a “Schelling Point” to facilitate communication between observers who have not coordinated the time or place of message exchanges. A Schelling point is a game theory concept which proposes links can be formed between two would-be communicators simply by using common references, in this case a supernova, to coordinate the time and place of communication. In addition to supernovae, source events include gamma-ray bursts, binary neutron star mergers, and galactic novae.

In conjunction with the natural event which attracts the attention of other civilizations, the advanced civilization broadcasts a technosignature signal unambiguously advertising its existence. The technosignature might, as an example, mimic a pulsar’s output: modulation, frequency, bandwidth, periods, and duty cycle.

The limiting factor in using the SETI Elliposoid to search for targets is the unavailability of precise distance measurements to nearby stars. The Gaia project remedies that problem. The mission’s two telescopes provide parallaxes, with precision 100 times better than its predecessors, for over 1.5 billion sources. Distance uncertainties are less than 10% for stars within several kiloparsecs of Earth. This precision directly translates into lower uncertainties on the timing for signal coordination along the SETI Ellipsoid.

“I think the technique is very straightforward. It’s dealing with triangles and ellipses, things that are like high-school geometry, which is sort of my speed,” James Davenport , University of Washington astronomer and lead author in the referenced papers, joked with GeekWire. “I like simple shapes and things I can calculate easily.” [2]

An advanced civilization identifies a prominent astronomical event, as an example, a supernova. It then determines which stars could harbor civilizations which could also observe the supernova and the advanced civilization’s star. An unambiguous beacon is transmitted to stars within the Ellipsoid. The volume devoted to beacon propagation is significantly reduced, which reduces power and cost, when compared to an omnidirectional beacon.

At the receiving end, the listeners would determine which stars could see the supernova and which would have time to send a signal to the listeners. The listening astronomers would benefit by limiting their search volume to stars which meet both criteria.

For example, astronomers on Earth only observed SN 1987A in 1987, thirty six years ago. If the advanced civilization beamed a signal at the Solar System a century ago, our astronomers would not have the necessary clue, the observation of SN 1987A, to select the advanced civilization’s star as the focus of a search. Assuming both civilizations are using SN 1987A as a coordination beacon, human astronomers should listen to targets within a hemisphere defined by a radius of thirty-six light-years.

The following is written with apologies to Albert Einstein. The advanced civilization could observe the motion of stars and predict when a star will come within the geometry defined by the Ellipsoid. In the case of the Earth and SN1987A, the advanced civilization could have begun transmissions thirty-six years ago.

The recently discovered SN 2023ixf in the spiral galaxy M101 could serve as one of the foci of an Ellipsoid. 108 stars within 0.1 light-year of the SN 2023ixf – Earth SETI Ellipsoid. [3]

Researchers propose to use the Allen Telescope Array (ATA), designed specifically for radio technosignature searches, to search this Ellipsoidal. The authors point out the utility of the approach and caution about its inherent anthropocentric biases:

“…there are numerous other conspicuous astronomical phenomena that have been suggested for use in developing the SETI Ellipsoid, including gamma-ray bursts (Corbet 1999), binary neutron star mergers (Seto 2019), and historical supernovae (Seto 2021). We cannot know what timescales or astrophysical processes would seem “conspicuous” to an extraterrestrial agent with likely a much longer baseline for scientific and technological discovery (e.g., Kipping et al. 2020; Balbi & Ćirković 2021). Therefore we acknowledge the potential for anthropogenic bias inherent in this choice, and instead focus on which phenomena may be well suited to our current observing capabilities.”

1. James R. A. Davenport , Bárbara Cabrales, Sofia Sheikh , Steve Croft , Andrew P. V. Siemion, Daniel Giles, and Ann Marie Cody, Searching the SETI Ellipsoid with Gaia, The Astronomical Journal, 164:117 (6pp), September 2022,

2. Alan Boyle, How ‘Big Data’ could help SETI researchers intensify the search for alien civilizations, 22 June 2022,

3. James R. A. Davenport, Sofia Z. Sheikh, Wael Farah, Andy Nilipour, B´arbara Cabrales, Steve Croft, Alexander W. Pollak, and Andrew P. V. Siemion, Real-Time Technosignature Strategies with SN2023ixf, Draft version June 7, 2023.

Abundant Phosphorus in Enceladus Ocean Increases Habitability: But is Enceladus Inhabited?

Finding the right conditions for life off the Earth justifiably drives many a researcher’s work, but nailing down just what might make the environment beneath an icy moon’s surface benign isn’t easy. The recent wave of speculation about Enceladus revolves around the discovery of phosphorus, a key ingredient for the kind of life we are familiar with. But Alex Tolley speculates in the essay below that we really don’t have a handle on what this discovery means. There’s a long way between ‘habitable’ and ‘inhabited,’ and many data points remain to be analyzed, most of which we have yet to collect. Can we gain the knowledge we need from a future Enceladus plume mission?

by Alex Tolley 

There has been abundant speculation about the possibility of life in the subsurface oceans of icy moons. Europa’s oceans with possible hydrothermal vents mimicking Earth’s abyssal oceans and the probable site of the origin of life, caught our attention now that Mars has no extant surface life. Arthur C Clarke had long suggested Europa as an inhabited moon in his novel 2010: Odyssey Two. (1982). While Europa’s hot vents are still speculative based on interpretations of the surface features of its icy crust, Saturn’s moon, Enceladus, showed visible aqueous plumes at the southern pole. These plumes ejected material that contributes to the E-Ring around Saturn as shown below.

While most searches for evidence for life focus on organic material, it has been noted that of the necessary elements for terrestrial life, Carbon, Hydrogen, Oxygen, Nitrogen, Sulfur, and Phosphorus (CHONSP), phosphorus is the least abundant cosmically. Phosphorus is a key component in terrestrial life, from energy management (ATP-ADP cycle) and information molecules DNA, and RNA, with their phosphorylated sugar backbones.

If phosphorus is absent, terrestrial biology cannot exist. Phosphorus is often the limiting factor for biomass on Earth, In freshwater environments phosphorus is the limiting nutrient [1]. Typically, algae require about 10x as much nitrogen as phosphorus. If the amount of available nitrogen is increased, the algae cannot use that extra nitrogen as the amount of available phosphorus now determines how large the algal population can grow. The biomass-to-phosphorus ratio is around 100:1. When phosphorus is the limiting nutrient, then the available phosphorus will limit the biomass of the local plants and therefore animals, regardless of the availability of other nutrients like nitrogen, and other factors such as the amount of sunlight, or water. Agriculture fertilizer runoff can cause algal blooms in aqueous environments and may result in dead zones as oxygen is depleted by respiration as phytoplankton blooms die or are consumed by bacteria.

While nitrogen can be fixed by bacteria from the atmosphere, phosphorus is derived from phosphate rocks, and rich sources of phosphorus for agriculture were historically gleaned from bird guano.

A recent paper in Nature about the detection of phosphorus in the grains from the E-ring by the Cassini probe’s Cosmic Dust Analyzer (CDA) suggested that phosphorus is very abundant. As these grains are probably sourced from Enceladus’ plumes, this implies that this moon’s subsurface ocean has high levels of dissolved phosphorus.

The authors of the paper have modeled, and experimentally confirmed the model, and make the claim that Enceladus’ ocean is very rich in phosphorus:

…phosphorus concentrations at least 100-fold higher in the moon’s plume-forming ocean waters than in Earth’s oceans.

around 100-fold greater than terrestrial phosphorus abundance levels. They show that the CDA spectrum [figure 1) is consistent with a solution of disodium phosphate (Na2HPO4) and trisodium phosphate (Na3PO4) (figure 2) The source of these salts on Enceladus is likely from the hot vents chemically releasing the material from the carbonaceous chondritic rocky core and the relatively alkaline ocean. Contrary to intuition, the greater CO2 concentrations in cold water with the hydroxyapatite-calcite and whitlockite-calcite buffer system maintain an alkaline solution that allows for the high phosphate abundance in the plume material that produces the grains in Saturn’s E-ring.

Figure 1. CDA cation spectrum co-added from nine baseline-corrected individual ice grain spectra. The mass lines signifying a high-salinity Type 3 spectrum are Na + (23 u) and (NaOH)Na + (63 u) with secondary Na-rich signatures of (H2O)Na + (41 u) and Na 2+ (46 u). Sodium phosphates are represented by phosphate-bearing Na-cluster cations, with (Na3 PO4)Na + (187 u) possessing the highest amplitude in each spectrum followed by (Na2HPO4 )Na + (165 u) and (NaPO3)Na + (125 u). The first two unlabelled peaks at the beginning of the spectrum are H + and C +, stemming from target contamination 3 (source nature paper). a.u., arbitrary units.

Figure 2. Spectrum from the LILBID analogue experiment reproducing the features in the CDA spectrum. An aqueous solution of 0.420 M Na2HPO4 and 0.038 M Na3PO4 was used. All major characteristics of the CDA spectrum of phosphate-rich grains (Fig. 1) are reproduced at the higher mass resolution of the laboratory mass spectrometer (roughly 700 m/?m). Note: this solution is not equivalent to the inferred ocean concentration. To derive the latter quantity, the concentration determined in these P-rich grains must be averaged over the entire dataset of salt-rich ice grains. (source Nature paper).

Fig. 3: Comparison of observed and calculated concentrations of ΣPO43– in fluids affected by water–rock reactions within Enceladus. a, Relation between ΣPO43– and ΣCO2 at a temperature of 0.1 °C for the hydroxyapatite-calcite buffer system (solid lines) and the whitlockite-calcite buffer system (dashed lines). Constraints on ΣCO2 obtained in previous studies are indicated by the blue shaded area. The area highlighted in pink represents the range of ΣPO43– constrained in this study from CDA data. b, Dependence of ΣPO43– on temperature for the hydroxyapatite-calcite buffer and different values of pH and ΣCO2. A similar diagram for the whitlockite-calcite buffer can be found in Extended Data Fig. 11.

The simple conclusion to draw from this is that phosphorus is very abundant in the Enceladan ocean and that any extant life could be very abundant too.

While the presence of phosphorus ensures that the necessary conditions of elements for habitability are present on Enceladus, it raises the question: “Does this imply Enceladus is also inhabited?”

On Earth, phosphorus is often, the limiting factor for local biomass. On Enceladus, if phosphorus was the limiting factor, then one would not expect it to be detected as inorganic phosphate, but rather in an organic form, bound with biomolecules.

But suppose Enceladus is inhabited, what might account for this finding?

1. Phosphorus is not limiting on Enceladus. Perhaps another element is limiting allowing phosphates to remain inorganic. In Earth’s oceans, where iron (Fe) can be the limiting factor, adding soluble Fe to ocean water can increase algal blooms for enhanced food production and possible CO2 sequestration. On Enceladus, the limiting factor might be another macro or micronutrient. [This may be an energy limitation as Enceladus does not have the high solar energy flux on Earth.]

2. Enceladan life may not use phosphorus. Some years ago Wolfe-Simon claimed that bacteria in Mono Lake used arsenic (As) as a phosphorus substitute. [2] This would have been a major discovery in the search for “shadow life” on Earth. However, it proved to be an experimental error. Arsenic is not a good substitute for phosphorus, especially for life already evolved using such a critical element, and as is well-known, arsenic is a poison for complex life.

3. The authors’ modeling assumptions are incorrect. Phosphorus exists in the Enceladan ocean, but it is mostly in organic form. The plume material is non-biological and is ejected before mixing in the ocean and being taken up by life. The authors may also have wildly overestimated the true abundance of phosphorus in the ocean.

Of these explanations allowing for Enceladus to be inhabited, all seem to be a stretch that life may be in the ocean despite the high inorganic phosphorus abundance. Enceladan biomass may be constrained by the energy derived from the moon’s geochemistry. On Earth, sunlight is the main source of energy maintaining the rich biosphere. In the abyssal darkness, life is very sparse, although it can huddle around the deep ocean’s hot vents.

However, if life is not extant, then the abundance of inorganic phosphorus salts is simply the result of chemical equilibria based on the composition of Enceladus rocky core and abundant frozen CO2 where it formed beyond the CO2 snow line.

While the popular press often conflate habitability with inhabited, the authors are careful to make no such claim, simply arguing that the presence of phosphorus completes the set of major elements required for life:

Regardless of these theoretical considerations, with the finding of phosphates the ocean of Enceladus is now known to satisfy what is generally considered to be the strictest requirement of habitability.

With this detection, it would seem Enceladus should be the highest priority candidate for a search for life in the outer solar system. Its plumes would likely contain evidence of life in the subsurface ocean and avoid the difficult task of drilling through many kilometers of ice crust to reach it. A mission to Enceladus with a suite of life-detecting instruments would be the best way to try to resolve whether life is extant on Enceladus.

The paper is Postberg, F., Sekine, Y., Klenner, F. et al. Detection of phosphates originating from Enceladus’s ocean. Nature 618, 489–493 (2023).


1. Smil, V (2000) Phosphorus in the Environment: Natural Flows and Human Interferences. Annual Review of Energy and the Environment Volume 25, 2000 Smil, pp 53-88

2. Wolfe-Simon F, et al (2010) “A Bacterium That Can Grow by Using Arsenic Instead of Phosphorus. SCIENCE 2 Dec 2010, Vol 332, Issue 6034 pp. 1163-1166

SETI: Asking the Right Questions

Did Carl Sagan play a role in the famous Arecibo message transmitted toward the Hercules Cluster in 1974? I’ve always assumed so, given Sagan’s connection with Frank Drake, who was then at Cornell University, where Sagan spent most of his career. But opinion seems to vary. Artist/scientist Joe Davis, who now has affiliations with both MIT’s Laboratory of Molecular Structure and Harvard Medical School, noted in an email this morning that Sagan’s widow, Ann Druyan, supports the connection, but according to Davis, Drake himself denied Sagan’s role in the composition or transmission of the message.

I mention all this because of Tuesday’s post on the simulated SETI signal being sent via ESO’s Mars ExoMars Trace Gas Orbiter, as a kind of work of art in its own right as well as a test case in building public involvement in the decoding of an unusual message. The idea of doing that irresistibly recalled Joe Davis because in 1988 Davis performed his own act of scientific art involving SETI, one that involved the Arecibo message and raised the question of whether any recipients would recognize it, much less decode it.

Image: A color-coded version of the Arecibo message highlighting its separate parts. The binary transmission itself carried no color information. Credit: Arne Nordmann / Wikimedia Commons. CC BY-SA 3.0.

The project, called “A Message in Many Bottles,” was set up at MIT’s Hayden Library in 1988. Davis used 1679 ‘Boston round’ 16 ounce glass bottles arrayed in a set of partitioned racks that were displayed in stacks. This is remarkably clever stuff: Each of 18 aisles in the library contained racks of bottles mounted, as Davis told me, 23 across. Empty bottles served as 0s in this digital message, while bottles filled with water represented the 1s. The whole thing reproduced the 1974 Arecibo message.

Now remember, this is MIT. You would think that if there is any place where a population of scientists, academics and students might puzzle out an enigmatic artifact like this, it would be here. Davis puts it this way in his email:

Hayden Library is MIT’s science library and contains all of the information needed to decode the message, all information the message refers to, and supposedly, better-than-average terrestrial intelligence. To the best of my knowledge, nobody decoded it. Instead, there were arguments…about whether or not the racks of bottles constituted works of art.

Image: An evidently baffled student contemplates the “Message in Many Bottles.” Credit: Joe Davis.

In 1997, a year after Sagan’s death, Davis reinstalled the display at MIT’s then new biology facility (Building 68), dedicating the work to the memory of Sagan. A short article on the matter in Nature (27 March 1997) noted the project as an homage to Sagan that accurately reproduced the Arecibo signal, going on to note:

Philip Sharp, chairman of MIT’s biology department, describes the exhibit as a “fitting tribute” to Sagan’s work. “It brings the abstraction of a radar message into an accessible, physical form,” says Sharp. He says he sees “numerous benefits” in having an artist who approaches issues from an unorthodox perspective working alongside more formally trained scientists.

Labeled as a tribute to Sagan and explained so that viewers could decode the message, “A Message in Many Bottles” served as an effective exhibit inhabiting the muzzy borderland where science meets art and creative minds translate research into shapes and forms that interrogate the meaning of our experiments. For that matter, was the Arecibo message itself not a kind of art, given that with a target 25,000 light years away, there was no conceivable way to see it as an actual communication?

Back in 2009 Joe Davis wrote “RuBisCo Stars” and the Riddle of Life for Centauri Dreams, presenting his own work at Arecibo, which wound up, on the 35th anniversary of the Arecibo message, in a new message based on molecular biology that was sent to three nearby stars. How he did this using, remarkably, an analog audio file on his iPhone interfacing with Arecibo’s technology is explained in the second part of his 2009 post, “RuBisCo Stars”: Part II. These two posts are, as everything involving Joe Davis’ work continues to be, invigorating and startlingly thought-provoking.

Image: At Arecibo, Joe Davis ponders transmission options as he holds the possible answer. Credit: Ashley Clark.

In fact, Davis notes in part II, in the midst of explaining to Arecibo’s then interim director Michael Nolan what his project is about, that “projects concerned with the search for extraterrestrial intelligence are really more about a search for ourselves; that they make us look much more intensely at ourselves than we look away into space and that nobody seems to see that part of it.” Nor could the myriad well-trained minds who encountered the Arecibo message in “A Message in Many Bottles” decode its meaning.

Science is so often about asking the right question. What are we staring at right now that we are not seeing? Are we asking the right questions about SETI?