Astronomy Now Home

The frugal alien's beacon
Posted: 8 June 2010

Bookmark and Share

August, 15 1977: a pulse of radio waves at 1,420MHz radiates down from space to be received by the Big Ear radio telescope in Ohio for 72 seconds. Then: nothing. Sporadic searches of the area since have failed to find this interstellar radio chorus. It’s origins remain a mystery.

We are of course talking about the fabled ‘Wow!’ signal, the SETI detection that never was. Critics argued that because it switched off after a short time, never to heard from again, it could not be a real alien signal. There was no message contained within it, no structure, no signature of intelligent design.

The computer print out that shows the ‘Wow!’ signal detected in 1977. Image: Ohio State University Radio Observatory/NAAPO.

Now there is a new explanation that raises the credibility of the ‘Wow’ signal’s extraterrestrial hypothesis, an idea we’ll call ‘Benford Beacons’. Developed by the Benford family of scientists – James, Dominic and the science fiction author Gregory – it is a powerful argument against the expectation of a continuous, omnidirectional transmitter built by altruistic aliens that has held SETI in its sway for much of the last fifty years. The basic point of the Benford Beacons is that ET will not be omnipotent, but will face a cost for any actions they decide to take. “A beacon is limited by its power budget,” writes Louis K Scheffer of Caltech in the SETI 2020 review. Therefore, ET civilisations will want to optimise their costs, limit waste, and make their signalling apparatus more efficient. They won’t be blasting out signals in all directions continuously, but will ‘ping’ world after world, over and over again, with short bursts to try and grab our attention. It would be more akin to Twitter than ‘Encyclopaedia Galactica’.

What is the logic behind frugal aliens? It stems from the pioneering work of the Soviet scientist Nikolay Kardashev, who in 1963 created a method of classifying extraterrestrial civilisations based on their ability to harness available resources. A type I civilisation would have the entire resources of a single planet at hand; type II civilisations would be able to utilise the entire power output of a star (perhaps with an immense engineering project such as a Dyson Sphere) while a type III civilisation would dwarf everything else around it, creating an empire that runs off the energy of an entire galaxy, or at least a good chunk of it. For the record, human civilisation doesn’t even rate as type I – we’ve been assessed as coming in at a level of about 0.7. It is perhaps a measure of our wastefulness that large quantities of wind, wave and geothermal power remain untapped whilst we drill endless holes into the ground for oil.

If Kardashev’s civilisations hold true (and why not? We can certainly envisage type I and II civilisations, and with a little imagination type III as well, although the fact that the Galaxy does not appear to have been colonised places type III into question) then a type I civilisation can hope to utilise 1016 watts, type II 1026 watts, and type III a whopping 1036 watts. These are not set limits – there’s a graduation scale and just as humans are at 0.7, other civilisations could be somewhere between type I and II or type II and III – but Kardashev’s scale helps to act as a rough guide. Because the notion of cost – in terms of resources if not money – must surely be a universal one in any finite system, is it fair to say that ET won’t devote every last scrap of power into saying ‘hi’ to the human race? We can’t be certain; one could imagine cosmic missionaries attempting to convert the entire Galaxy to their religion or way of thinking, or perhaps a dying civilisation attempting to send all their knowledge to the stars during their twilight hours, devoting the majority of their resources to this one project. However, we can just as easily imagine frugal aliens that are faced with multiple demands on their resources. Perhaps an interstellar beacon to them would be the equivalent of a Large Hadron Collider or the Apollo Moon-landings, as much an act of prestige as of science. James Benford estimates that an ET civilisation could donate one-thousandth of their total power generation to the construction, maintenance and running of a beacon. With this in mind, what could they achieve?

Cost-optimised beacons would flash at us for mere seconds at a time, and be located close to the plane of the Galaxy within the Milky Way, particularly in the direction of the galactic centre. Image: ESO/S Guisard.

The trick to understanding how Benford Beacons work is appreciating that their signals, designed to capture our attention, need only be very short – seconds or less. Otherwise the alternative – large beacons running constantly – is going to require a little more than 50 pence in the meter to keep switched on.

“Omnidirectional beacons are big-time and expensive, but easily noticed,” James Benford tells us. “But we haven’t seen any, so the observational test result is that they don’t exist.”

Which is a problem for SETI, because many of our detectors are designed to spot exactly these kinds of signals, rather than the shorter beacons. The characteristics of Benford Beacons are a short duration and a rapid revisit time. All they have to do is pique our curiosity and, once detected, we may study the location of the beacon with more scrutiny to try and detect the real message content in the form of a fainter, lower power signal. So we’re searching for what may appear at first glance to be transient radio pulses, and we’ve found plenty of these.

Test case
In 2002 a puzzling and unidentified transient radio source, designated GCRT J17445-3009, was discovered by the Very Large Array (VLA) radio telescopes in New Mexico, USA. It emitted, in bursts, radio waves at 330MHz. There’s nothing to suggest this is an extraterrestrial beacon; it could a precessing pulsar, a quirky flare star, maybe a pair of neutron stars or white dwarf binary that is accreting material. Although it was observed in the direction of the galactic centre, we have no idea how far away GCRT J17445-3009 really is, but the Benfords use it as a great example of ‘what if’ to illustrate why ET would use cost-optimised beacons.

Let’s imagine that it were a beacon located 1,000 light years away. With a bandwidth of at least 30MHz, its total power would be 1.4 x 1026 watts – practically the entire output of a type II civilisation. To make the signal reach even further would need even more power.

Artwork depicting the dishes of the Square Kilometre Array, which may be capable of detecting cost-optimised beacons.

Now, suppose instead that it were a cost-optimised beacon directed towards only us rather than signalling many star systems all at the same time. As the burst lasted ten minutes we would say that this beacon has a ‘dwell time’ of ten minutes; in other words it was pointed at us for that length of time before moving on to the next star. The signal then reappeared after 77 minutes – this would be its revisit time. From this we can measure its ‘duty cycle’, which is the proportion of time during which the beacon is aimed at us, and is the product of the dwell time of the pulse and its revisit time, e.g. 10/77=0.13; in other words it would be directed towards us 13 percent of the time. At a 1,000 light years its power would be 3.4 x 1020 watts – it’s still a lot but falls within the capabilities of a type II civilisation using far less than a thousandth of their total power resources, even if it is not the most efficient beacon one could imagine. The Benfords suggest that we should search for cost-optimised beacons at frequencies above 1GHz and below 10–15GHz because the higher frequencies are more efficient in terms of power to cost ratio (it would also have the advantage of being far above the typical frequency range of pulsars, eliminating the possibility of misidentification; pulsars also have large bandwidths, compared to the narrowband signals of a cost-optimised beacon). Such frequencies are, of course, could potentially take us far above the traditional water hole frequencies between 1.42 and 1.66GHz. The crucial implication, say the Benfords, is that we have been looking in the wrong places for the wrong signals – little wonder that SETI has turned up nothing thus far.

The vast majority of SETI experiments over the years integrate their observations over minutes or hours, and so they will be blind to signals just milliseconds in length. SETI@home can detect signals as short as 13 seconds , and the ‘Fly’s Eye’ experiment built at the University of California, Berkeley, to be used on the Allen Telescope Array, does have the ability to detect signals over millisecond durations but with greatly reduced sensitivity (incidentally, optical SETI observations also have their home at Berkeley, and these are tailored to detect millisecond laser pulses – see our previous article). So things are improving, and it is hoped that the Square Kilometre Array with its planned million square metres of collecting area will be able to integrate over short timescales in order to catch short transient bursts from millisecond pulsars and the like, although the specifics of its capabilities have still to be set in stone.

How to build a beacon
A Benford Beacon would likely be constructed using a phased array – an array of transmitters that are adjusted so their signals are in phase. The benefits of a phased array are in its combined efficiency. Imagine a single transmitter with a voltage V and a power density P (the amount of power across a given unit of area at a right angle to the direction the signal is travelling in). Then imagine another transmitter on the other side of the country. Together they have a voltage of 2V, but because the power density is proportional to the square of the voltage in a phased array, the power density is 4P. Similarly, if you had three transmitters the power desity would be 9P, and so on. A single dish, however large, would not be as efficient or have the same range as a large number of smaller dishes – for the single dish to reach 4P you would have to raise its voltage to 4V.

Searches for beacons should be targeted towards the galactic centre, and in the opposite direction for messages beamed towards the centre. Image: NASA/JPL–Caltech/R Hurt (SSC–Caltech).

To keep the array of transmitters in phase despite the planet rotating (unless the transmitters are located in space, perhaps at the L1 Lagrangian point, which would avoid any attenuation by a planetary atmosphere, at least on the transmitting end) each dish would have to be constantly adjusted to keep the signal coherent. One could imagine a satellite with a global view of the array, using an oscillator to send out a master signal that is received by all the transmitters in the array, corrected for phase decoherence, amplified and transmitted back into space.

If ET has a lot of resources to devote to a beacon, then that’s good, because although the greater the range of the beacon the more it costs, it also reaches more stars and increases the chances of success. But success can only be achieved if those on the other end searching for the signal – i.e. us – are looking for it. So, how should we plan to detect a cost-optimised Benford Beacon?

First, say the Benfords, let’s revisit the locations of all the powerful, one-off transient bursts that have been detected in past surveys. The reason why such bursts, like the Wow signal, have been dismissed, is that they have never been seen to recur, but seeing as SETI searches revisit the Wow location for a few hours each year on average, it is little wonder we’ve not seen it again if it is a cost-optimised beacon. Instead we need to get onto these signals and watch, patiently, for some time.

Towards the galactic centre should be the focus of our surveys, add the Benfords, and already this features in many SETI surveys simply because towards the centre, in the constellation of Sagittarius, is the greatest concentration of stars as we look into the densest part of the Milky Way (and the Wow signal was very close to the direct line towards the galactic centre). We should also look in the opposite direction, towards the anti-centre in Auriga, for anyone sending messages back to the galactic centre, and also look along our nearest spiral arm. Additionally, movement of stars around the Galaxy is not restrained to planar orbits; they also oscillate up and down out of the galactic plane. Why is this important? Being above the plane of the Galaxy may not be a good place for life to find itself. Analysis of periodic extinctions on Earth every 62 million years by Mikhail Medvedev and Adrian Melott of the University of Kansas show that these extinctions coincide with the Earth being above the galactic plane. Medvedev and Melott surmise that planets will experience a larger influx of higher energy cosmic rays accelerated by the bow shock on the northern side of the Milky Way Galaxy as we plough through the intergalactic medium at 200 kilometres per second. This would result in a five-fold increase in cosmic rays on planets high above the galactic plane, so the Benford’s suggest keeping SETI searches focused on a narrow patch of sky that extends just above and below the galactic plane.

The Life Plane also provides another test for Benford Beacons. If unidentified transient radio signals roughly match the distribution of stars above and below the galactic plane, then they’re probably not beacons but natural stellar phenomena that could occur anywhere in the Galaxy. On the other hand, if they are clustered close to the galactic plane then this would be circumstantial evidence that they could be artificial because that is where we expect to find life as we know it, rather than in the exposed regions high above the plane.

Perhaps the sky has seemed silent only because we lack the appropriate detection capability. One particular survey in the 1990s led by W T Sullivan of the University of Washington, spread over two-and-a-half hours with 1.2 minute integrations, turned up a large number of “intriguing, non-repeatable, narrowband signals, apparently not of manmade origin and with some degree of concentration toward the galactic plane.” Maybe when we finally open our ears to Benford Beacons, we’ll be surprised to find that everybody has been yelling at us all along.

Find out more:

How can we distinguish transient pulsars from SETI beacons? (James Benford): Read here

Searching for Cost-Optimised Interstellar Beacons (Gregory Benford, James Benford, Dominic Benford): Read here

The ‘Wow!’ signal at the Big Ear Radio Observatory: Read here

The Fly’s Eye: Read here

Medvedev and Melott’s Life Plane paper in The Astrophysical Journal: Read here.

Previous SETI articles can be found here.