by Wm. Robert Johnston
updated 30 October 2007
This interesting question came up recently: which pulsars are used in the Voyager record pulsar map? Answering this question turned into an interesting exercise in interpreting this map, which I review here.
During planning of the Pioneer 10 and 11 missions to the outer solar system, a group of astronomers proposed including a message for any intelligent extraterrestrials that might discover the probes. These two probes would be the first manmade objects to leave the solar system. While the probability of such beings discovering the probes is infinitesimally small (even assuming they exist), a message was included on a metal plaque on the outside of each probe prior to their launches in 1972 and 1973.
For Voyagers 1 and 2, outer solar system probes launched in 1977 and also destined to leave the solar system, a more extensive message was incorporated into a record containing sounds and digital pictures from Earth. The cover of the record included a message modified from the Pioneer message.
All four messages included in particular a pulsar map. This map was designed by Frank Drake to shown the location of the Earth relative to 14 pulsars. Pulsars are rapidly rotating neutron stars which emit beacons of radio waves. The precise timing of the resultant radio pulses provide markers observable in radio wavelengths across broad stretches of the galaxy.
A quick search failed to turn up a readily accessible list of the pulsars used in the map, so I decided to experiment with interpreting the map. If the map could really be interpreted to show our location, it would then be possible to identify the pulsars from the map. This would serve to test at least a few elements (only a few) involved in its potential interpretation.
The map identifies the distance and direction of the 14 pulsars relative to the direction and distance of the center of the galaxy, along with--most importantly--their periods. The periods are expressed in binary notation as multiples of the hyperfine transition period of hydrogen (7.04024183647 x 10-10 sec). This transition occurs in hydrogen atoms and involves a change in the relative spins of the proton and electron; the resultant radio-frequency radiation, with a wavelength of 21.1 cm, is observed in interstellar gas.
Using a diagram from p. 58 of Murmurs of Earth by Sagan et ali (which describes the Voyager record message), the data in table 1 is derived (listing the pulsars as on the map, clockwise from the line to the center of the galaxy). (Note: some measurements may not be accurate due to the quality of the diagram.) Pulsar period is given in the units of the hydrogen hyperfine transition (in both binary and base 10) and converted to seconds. The direction and distance information is less precisely indicated on the diagram. Distance in particular to most pulsars is poorly determined, even today.
|psr #||period (binary, H transition units)||period|
(base 10, H transition units)
|period (seconds)||direction angle from|
galactic center (°)
(to galatic center=100)
As an initial attempt, the derived periods were compared to a list of pulsars known as of June 1975. Using this sample of 147 would narrow down the possibilities, since it would at least include the 14 used to create the map around 1970. The list used was from Taylor and Manchester, "Observed properties of 147 pulsars," (1975), The Astronomical Journal, (80:794-806). Matches were found for all 14 pulsars, with data given below in table 2. Included is the difference (in parts per million) between the periods in the map and the presumed matches in Taylor and Manchester; note that most are quite good.
|psr #||PSR (B1950 ID)||period (seconds)||difference (ppm)||distance (parsecs)|
Finally, a modern pulsar listing was used to check that the pulsars can be unambiguously identified. Using the ATNF Pulsar Catalog (2002), which lists 1,480 pulsars, the data in table 3 was obtained for the above pulsars. Celestial coordinates (RA and DEC) and galactic coordinates (l and b) are given along with period, P dot (the rate of decrease of the period), the epoch (time of period measurement), and distance.
|psr #||PSR (J2000 ID)||R A||DEC||l (°)||b (°)||period (seconds)||P dot|
All 14 pulsars could be distinguished from others, although not by period alone in one or two cases. In these cases the direction made the choice clear. Table 4 compares direction and distance information from the pulsar map to that derived from the ATNF Catalog data.
|psr #||PSR (J2000 ID)||direction, angle from|
line to galactic center (°)
|from map||from ATNF data||from map||from ATNF data|
The direction information is somewhat helpful, but the distance information is not.
The precise nature of pulsar periods largely extends to the rate of decrease of these periods. If the pulsar map periods are accurate, one can in principle determine the time corresponding to those measurements of the pulsar periods. Comparing the data from the ATNF Catalog to that from the pulsar map, this is used in table 5 to estimate the date of origin of the pulsar map.
|psr #||PSR (J2000 ID)||derived date|
Discarding the discordant value from PSR J1645-0317, this gives a date of origin of 1969.7 ± 1.2 year! This is not a bad result. (A note: I don't know whether the pulsar data was updated for the Voyager pulsar map.) Now suppose some ETI reads the map and manages to determine which pulsars we were referring to. Reading the map is not so simple in this case: the solar system is moving through space, as are the pulsars (generally with higher space velocities than main sequence stars); adjusting for the change in the pulsar periods must take into account the light travel time from the pulsars, which is hundreds to tens of thousands of years. But if these factors were successfully accounted for, the map could be used to identify the location of the solar system to within tens of light years or less.
Of course, my task of reading the map was far less daunting than it would be for some ETI that finds the map. Just a sampling of issues for interpretation that did not enter this analysis:
Epilogue: After completing this page, I received a reply from the Planetary Society which referred me to a 1972 article by Carl Sagan, Linda Salzman Sagan, and Frank Drake identifying the pulsars. The article uses yet another type of pulsar identifier in a table, data from which is given below (note: this was read from a poor reproduction, so there may be errors in my table). The last column gives the pulsar identifier from a 1970 article by B. Y. Mills, giving some indication of the then non-standardized cataloguing system.
|psr #||ID in Sagan et ali||given period,|
1970/71 epoch (sec)
|period, H transition units||ID in Mills (1970)|
Some statements of interest from the Sagan et ali article:
A time interval of only 3 weeks existed between the formulation of the idea of including a message on Pioneer 10, achieving NASA concurrence, devising the message, and delivering the draft message for engraving.This occurred before final assembly of the Pioneer 10 spacecraft.
The large number of digits is the key that these numbers indicate time intervals, not distances or some other quantity... There are no other conceivable quantities that we might know to ten significant figures for relatively distant cosmic objects.Actually, angular position as well can be determined to ten significant figures.
With 14 periods, almost all of which are accurate to nine significant figures in decimal equivalent, a society which has detailed records of past pulsar behavior should be able to reconstruct the epoch of launch to the equivalent of the year 1971. If past records of pulsar "glitches" (discontinuities in the period) are not kept or reconstructable from the physics it should still be possible to reconstruct the epoch to the nearest century or millennium... If either [of the two pulsars nearest the Earth] is correctly identified, it can be used to place the position of our solar system in the galaxy to approximately 20 pc, thereby specifying our location to approximately 1 in 103 stars.This does not address the fact that light travel time from the pulsars must be taken into account. If this can be done (and, as noted above, if discontinuities in pulsar periods are not problematic), the location can be determined to within a few light years, nearly uniquely identifying the Sun.
(In 2007 Rick Nungester alerted me to a few errors, particularly my calculation of the period of pulsar #1; this is corrected above.)
© 2003, 2007 by Wm. Robert Johnston.
Last modified 30 October 2007.
Return to Home. Return to Astronomy and Space.