The Wow! signal

The Ohio State University Radio Observatory was built by John Kraus between 1956 and 1963 and used for the Ohio Sky Survey until 1971; from 1973 it was given over to SETI, a programme led by Robert Dixon that ran until the telescope was demolished in 1998 to make way for a golf course and housing⁠[1]⁠[2]. The receiver recorded 50 channels of 10 kHz across the hydrogen line at 1420 MHz and the output went to an IBM 1130 computer and a line printer; the paper was collected every few days and delivered to Ehman for review⁠[2].

The signal entered the receiver at about 23:16 Eastern Daylight Time on 15 August 1977⁠[2]. Ehman's own 1997 and 2010 reports give the recorded parameters: observing frequency 1420.4556 ± 0.005 MHz in channel 2; a peak at about 30 sigma; a flux density inside the 10 kHz band of either 54 or 212 Jy, depending on which horn saw it; six data points at 12-second intervals, so about 72 seconds in all; and two possible J2000 positions, RA 19h25m31s ± 10s or 19h28m22s ± 10s, both at Dec −26°57′ ± 20′, in Sagittarius⁠[2]. The intensity profile matched the antenna pattern of a point source drifting through the beam within measurement error⁠[2].

The telescope's two horns produce a positive and a negative response offset by about 2 minutes 52 seconds in right ascension; a steady source produces both. The Wow! signal produced one. Either the source turned off (or on) within three minutes, or something blocked it in one horn; the staff could not say which horn saw it, and later checks of the horn wiring did not resolve the question⁠[2]. The Big Ear went on to log more than 40,000 transient narrowband events in 25 years of SETI operation; none has the Wow! signal's combination of strength and shape⁠[1].

There are no eyewitnesses in the ordinary sense; the record is the printout. Ehman's account of finding it a few days after the detection, probably on 19 August, is given in his own words in the 20th and 30th anniversary reports, which also reconstruct every number on the printout line by line⁠[10]⁠[2]. Dixon, the programme's director, and the chief observer Russell Childers have described the observatory's operation and the signal's place in it in a 2026 conference paper with Méndez⁠[1]. The original printout, which Ehman kept and annotated, is held by the Ohio History Connection in Columbus; a colour copy scanned several years after 1977 appears in Ehman's reports⁠[2].

Terrestrial interference is the default for any single narrowband event and must be considered first. Against it: the frequency lies in a band where transmission is prohibited worldwide; the signal's rise and fall matched the beam pattern of a fixed point in the sky; aircraft and satellites cross the beam too fast or too slowly to produce that profile⁠[2]. For it: nothing in the data proves the signal was not a reflection of a ground transmitter from a piece of debris that happened to drift at the sidereal rate; Ehman regarded this as unlikely but not excluded⁠[2]. The 2025 reanalysis argues from the frequency shift that an astrophysical origin is more likely than interference⁠[9].

The comet hypothesis of Antonio Paris proposed that hydrogen clouds around comets 266P/Christensen and P/2008 Y2 (Gibbs), which were in the general region in 1977, produced the emission⁠[6]. Ehman and Dixon rejected it: the comets were off the recorded position by about an hour of right ascension (266P) or three degrees of declination (P/2008 Y2) against a beam three minutes wide in RA; a cometary hydrogen cloud would be degrees across while the signal came from a point; there is no known mechanism for strong 21-cm emission from comets; and a comet would have been seen in both horns⁠[7]. The hypothesis is not accepted by the Big Ear staff or by later analysts⁠[8].

The hydrogen-cloud hypothesis of Méndez, Ortiz Ceballos and Zuluaga rests on Arecibo drift scans from 2020 that found several narrowband signals near the hydrogen line, about a hundred times weaker than the Wow! signal, which they identify as small cold hydrogen clouds (kinetic temperature under about 100 K). They propose that the Wow! signal was such a cloud brightening suddenly, by stimulated emission or Dicke superradiance, after a burst of radiation from a transient such as a magnetar flare behind it⁠[8]. The mechanism would explain the narrow band, the hydrogen frequency, the single appearance and the point-like source. It does not yet have a second example, and the triggering transient was not observed; the authors say so⁠[8]. The 2025 revision of the frequency to 1420.726 MHz is consistent with the proposal and with a Galactic source⁠[9].

The extraterrestrial-transmitter hypothesis is the one the search was built to test. The signal has every property the Ohio programme expected an artificial beacon to have: narrow band, at the hydrogen line, fixed on the sky, strong⁠[2]. What it lacks is a repeat. A beacon that sweeps or repeats at intervals longer than the follow-ups cannot be excluded by the observations so far⁠[11]. No modulation or information content was recorded, because the receiver could not have recorded any⁠[2].

A repeat would settle most of it. A second detection at the same position and frequency, with a modern receiver recording the waveform, would distinguish a hydrogen-cloud flare (a broadened, thermal line with no structure) from a transmitter (a spectral line narrower than any natural process, or modulation). The repeating-beacon analysis says what that costs: about 62 days of accumulated observation on the two positions to exclude a random repeater at 3 sigma⁠[11]. A second signal of the Wow! type anywhere on the sky, caught with a telescope able to follow up within minutes, would test the Méndez mechanism, which predicts that such events are rare but not unique and that they sit on cold hydrogen clouds⁠[8]. Short of a repeat, the Ohio archive is the one unexamined resource: 25 years of consistent data in which the signal's statistical company has never been fully characterised⁠[1].