SP-419 SETI: The Search for Extraterrestrial Intelligence




1. Alternative Methods of Communication.
2. Notes on Search Space.
3. Parametric Relations in a Whole Sky Search.
4. Stellar Census.
5. Summary of Possible Uses of an Interstellar Search System for Radio Astronomy.
6. SETI Related Scientific and Technological Advances.
7. A Preliminary Parametric Analysis of Search Systems.
8. Radio Frequency Interference.
9. Protection of a Preferred Radio Frequency Band.
10. Responses to a Questionnaire Sent to Leading Radio Observatories.
11. The Soviet CETI Report.
12. Searches to Date.
13. The Maintenance of Archives.
14. Selected Annotated Bibliography.
15. Workshop Members, Workshop Meetings.

[102] Blank page




Prepared by:
John H. Wolfe
SETI Program Office
Ames Research Center


[104] Blank page




[105] Several methods of achieving contact with intelligent life beyond our solar system have been proposed. These include actual interstellar space travel, the dispatching of interstellar space probes, and the sending or detection of signals of some form. Many other suggestions involving as yet unknown physical principles have also been made but are not considered here.




The classical method of interstellar contact in the annals of science fiction is the spaceship. With our development of spacecraft capable of interplanetary missions, it is perhaps not amiss to point out how far we still are with our present technology from achieving practical interstellar space flight, and indeed how costly such travel is in terms of energy expenditure even with a more advanced technology.

Chemically powered rockets fall orders of magnitude short of being able to provide practical interstellar space flight. A vehicle launched at midnight from a space station orbiting the Earth in an easterly direction and having enough impulse to add 10.5 mi/sec to its initial velocity would escape Earth with a total orbital speed around the Sun of 31.5 mi/sec. This would enable the vehicle to escape the solar system with a residual outward velocity of 18 mi/sec, or about 10-4 c . One could also use Jupiter in a swingby gravity assist maneuver to escape the solar system (as Pioneer 10 has done) and achieve about the same outward velocity with a somewhat lower launch energy requirement. Since, however, the nearest star, a [alpha]-Centauri, is 4 light-years away, the rendezvous, if all went well, would take place in 40,000 years. Clearly we must have at least a thousandfold increase in speed to consider such a trip and this means some radically new form of propulsion.

Spencer and Jaffe (ref. 2) have analyzed the performance attainable from nuclear-powered rockets using (1) uranium fission in which a fraction Greek letter epsilon= 7 x 1O-4 of the mass is converted to energy and (2) deuterium fusion for which Greek letter epsilon= 4 x 10-3. The mass ratios, µ , required for a two-way trip with deceleration at the destination are given in table 1 for various ratios of the ship velocity v to the velocity of light c . The mass ratio, µ = Mi/Mo , where Mi is the total initial mass of the rocket (including fuel) and Mo is the mass after burnout, i.e., Mi minus the mass of the fuel. From these figures we would conclude that with controlled fusion we might make the trip to a[alpha]-Centauri and back in 80 years, but that significantly shorter times are out of the question with presently known nuclear power sources.

Let us ignore all limitations of present day technology and consider the performance of the best rocket that can be built according to known physical law. This is the photon rocket, which....





µ for uranium fission

µ for deuterium fusion



3.8 x 104

8.1 x 101


2.3 x 109

6.2 x 103



1.1 x 106



1.5 x 108

....annihilates matter and antimatter converting the energy into pure retrodirected radiation. The mass ratio µ required in such a rocket is:


mathematical equation, µ=square root of (1+v/c)/(1-v/c) = veff/c + square root of (1+(veff/c) to the power two)(1)



mathematical equation, veff = v divided by square root of (1-v (to the power 2) /c  (to the power 2)).= coordinate distance travelled per unit of ship's proper time


If we choose veff/c = 1, then to reach the a[alpha]-Centauri system, explore it, and return would take at least 10 years' ship time. It is hard to imagine a vehicle weighing much less than 1000 tons that would provide the drive, control, power, communications and life support systems adequate for a crew of 12 for a decade. To accelerate at the start and decelerate at the destination requires a mass ratio of µ2, and to repeat this process on the return trip (assuming no nuclear refueling) requires an initial mass ratio of µ4. For veff/c = 1, µ=1+ square root of two, and µ to the power 4 is equivalent to 34.

Thus the take-off weight would be 34,000 tons, and 33,000 tons would be annihilated enroute producing an energy release of 3 x 1024 J. At 0.1 cent per kWh this represents $1 million billion worth of nuclear fuel. To discover life we might have to make many thousands of such sorties.

Even disregarding the cost of nuclear fuel, there are other formidable problems. If the energy were released uniformly throughout the trip, the power would be 1016 W, But the ship is µ3 times as heavy for the first acceleration as for the last and the acceleration periods are a small fraction of the total time; hence, the initial power would be about two orders of magnitude greater, or 1018 W. If only one part in 106 of this power were absorbed by the ship the heat flux would be 1012 W, A million megawatts of cooling in space requires about 1000 square miles of surface, if the radiating surface is at room temperature. And, of course, there is the problem of interstellar dust, each grain of which becomes a miniature atomic bomb when intercepted at nearly optic velocity.

[107] We might elect to drop veff/c to 0.1 and allow 82 years for the trip. But this would undoubtedly require a larger payload, perhaps a 10,000 ton ship, SO our figures are not changed enough to make them attractive.

Ships propelled by reflecting powerful Earth-based laser beams have been proposed, but these decrease the energy required only by the mass ratio of the rocket they replace, and they require cooperation at the destination to slow them down. In addition, the laser powers and the mirror sizes required for efficient transmission are fantastically large.

Bussard (ref. 3) has proposed an ingenious spaceship that would use interstellar hydrogen both as fuel and propellant. After being accelerated by some other means to a high initial velocity, the ship scoops up the interstellar gas and ionizes and compresses it to the point where proton-proton fusion can be sustained, thereby providing energy for a higher velocity exhaust. Essentially the ship is an interstellar fusion powered ramjet. Bussard's calculations show that, for a 1-g acceleration, a 1000-ton ship would require about 104 km2 of frontal collecting area. No suggestions are given as to how a 60-mile diameter scoop might be constructed with a mass of less than 1000 tons. (104 km2 of 1 mil mylar weighs about 250,000 tons.) Solutions to this and other engineering details must await a technology far more advanced than ours. For example, it has been suggested that an interstellar ramjet might work by pre-ionization and magnetic scooping using superconducting flux pumps (Sagan, ref. 8).

A sober appraisal of all the methods so far proposed forces one to the conclusion that manned interstellar flight is out of the question not only for the present but for an indefinitely long time in the future. It is not a physical impossibility but it is an economic impossibility at the present time. Some unforeseeable breakthroughs must occur before man can physically travel to the stars.




Bracewell (ref. 4) has suggested that advanced societies might build interstellar probes, possibly artificially intelligent and self-reproducing, that expand through the space about their planet of origin, patrolling all star systems likely to develop intelligent life. In this scenario a probe enters a system and may detect stray radio emissions similar to those of our own. Bracewell's postulated response for the probe is to first record signals and then retransmit them to the planet of their origin where the time delay could be interpreted as an indication of the probe's presence and location. Communication with the home civilization then begins via the probe.

Bracewell also suggests we be alert for such probes in our own solar system. Villard (ref. 5) has suggested that long delayed echoes, which are in fact occasionally heard, conceivably could originate from such a probe. Until the source of such echoes can be definitely ascribed to some other mechanism such as slow propagation in the ionosphere near the plasma cutoff frequency, this will continue to be an intriguing, albeit an unlikely, possibility. The phenomenon deserves further study.

[108] An interstellar monitor probe could be much smaller than a spaceship and could take longer in flight. But although there would be no crew to face the psychological barriers or physiological problems of generations spent in space, there are still good reasons to require a short transit time. If the probe were to require 1000 years (or even only a century) to reach its destination, serious doubt would exist that it would not be obsolete before arrival. Thus, even if probes should be capable of velocities of the order of that of light, probe weights in excess of a ton would almost certainly be needed.

To "bug" all the sun-like stars within 1000 light-years would require about 106 probes. If we launched one a day this would take about 3000 years and an overall expenditure well over $10 trillion. Interstellar probes are appealing as long as someone else sends them, but not when we face the task ourselves.

The simple fact is that it will be enormously expensive, even with any technological advance we can realistically forecast, to send sizable masses of matter over interstellar distances in large numbers. This is not to say that unmanned probes to one (or a few) nearby stars for the purpose of scientific exploration would not be a worthwhile endeavor; in all likelihood such an attempt will be made at some future date.




We cannot rule out the possibility that we might stumble onto some evidence of extraterrestrial intelligence while engaged in traditional archeological or astronomical research, but we feel that the probability of this happening is extremely small. Not everyone shares this view. Dyson (refs. 6, 7), for example, has suggested supercivilizations of various sorts whose activities can be detected even if they are not actively engaged in an effort to contact and communicate with other societies. Dyson argues that a few societies may be unable to avoid a Malthusian population rise and so are forced to redesign their planetary system, then the nearby star system, and finally the entire galaxy in order to gain increased living space and energy resources. In particular, he has imagined a society that disassembles its planets and rebuilds them into a sphere of structures orbiting the central star. These structures are so numerous and so densely distributed that they effectively capture all the stellar radiation and reradiate it in the infrared region of the electromagnetic spectrum. From afar, such civilization would be a strong infrared source that showed other peculiar properties, such as radio or laser emission.

While there is a lack of general support for these and even more imaginative suggestions, some further consideration should be given such scenarios. Indeed, their implicit advantage is that each carries little cost or planning obligation-one only has to go about his normal business and wait either to stumble across evidence of supercivilizations or to be discovered by them. This is an obviously passive approach as compared to the active approach following logically from the orthodox view.




Although no one can deny the excitement that would accompany a physical visit to another inhabited world, most of the real benefit from such a visit would result from communication alone. Morrison has estimated that all we know about ancient Greece is less than 1010 bits of information; a quantity he suggests be named the "Hellas." Our problem therefore is to send to, and to receive from, other cultures not tons of metal but something on the order of 100 Hellades of information. This is a vastly less expensive undertaking.

Fundamentally, to communicate we must transmit and receive energy or matter or both in a succession of amounts or types of combinations that represent symbols, which either individually or in combination with one another have meaning-that is, can be associated with concepts, objects, or events of the sender's world. In one of the simplest, most basic, types of communication the sender transmits a series of symbols, each selected from one of two types. One symbol can be the presence of a signal, the other can be the absence of a signal, for example. This type of communication is called an asymmetric binary channel. For the receiver to be able to receive the message, or indeed, to detect its existence, the amount of energy or number of particles received when the signal is present must exceed the natural background. Suppose, for example, we knew how to generate copious quantities of neutrinos and to beam them. And suppose we could capture them efficiently in a receiver. Then, with the signal present, our receiver would have to show a statistically significant higher count than with no signal.

Even if the natural background count were zero, the probability of receiving no particles when the signal is in fact present should be small. Since the arrivals during a signal-on period are Poisson-distributed, the expectation must be several particles per on-symbol. Thus to conserve transmitter power we must seek particles having the least energy. The desirable properties of our signaling means are:


1. The energy per quantum should be minimized, other things being equal

2. The velocity should be as high as possible.

3. The particles should be easy to generate, launch, and capture.

4. The particles should not be appreciably absorbed or deflected by the interstellar medium.


Charged particles are deflected by magnetic fields and absorbed by matter in space. Of all known particles, photons are the fastest, the easiest to generate in large numbers, and the easiest to focus and capture. Low-frequency photons are affected very little by the interstellar medium, and their energy is very small compared with all other bullets. The total energy of a photon in the microwave region of the electromagnetic spectrum is one ten billionth the kinetic energy of an electron travelling at half the speed of light. Almost certainly electromagnetic waves of some frequency are the best means of interstellar communication - and our only hope at the present time (see Section II-4).




1. Oliver, B. M.; and Billingham, J.: Project Cyclops, A Design Study of a System for Detecting Extraterrestrial Intelligent Life. NASA CR-114 445, 1972.

2. Spencer, D. F.; and Jaffe, L. D.: Feasibility of Interstellar Travel. NASA TR32-233, 1962.

3. Bussard, R. W.: Galactic Matter and Interstellar Flight. Astronautica Acta, vol. VI, Fasc.4, 1960.

4. Bracewell, R. N.: Communications from Superior Galactic Communities. Nature, vol.186, no. 4726, May 28, 1960, pp. 670-671.

5. Villard, O. G., Jr.; Fraser-Smith, A. F.; and Cassan, R. T.: LDE's, Hoaxes, and the Cosmic Repeater Hypothesis, QST, May 1971, LV,5, pp. 54-58.

6. Marshak, R. E., ea.: Perspectives in Modern Physics, John Wiler & Sons, 1966, p. 641.

7. Cameron, A. G. W., ea.: Interstellar Communications, W. A. Benjamin, Inc., New York, 1963, pp. 111-114.

8. Sagan, C.: Direct Contact Among Galactic Civilizations by Relativistic Interstellar Space Flight, Planetary and Space Science. Science, vol. 11,1963, pp. 485-498.


*Most of the information presented here has been taken from the Cyclops report, pp. 33 - 36 (ref. 1).