The first series of negotiations in Geneva on a comprehensive ban on nuclear explosion testing among the then nuclear weapon states (U.S.S.R., U.K., and U.S.A.) began in the late 1950s. Seismologists from a number of countries, including Canada, were included in the technical delegations of the 1958 "Conference of experts to study the methods of detecting violations of a possible agreement on the suspension of nuclear tests." The experts agreed that underground nuclear explosions in the range 1- 5 kilotons could be detected and identified if seismograph facilities were established in 170 land-based control posts. Although later studies revised these initial recommendations, research on the appropriate type of seismograph facilities began almost immediately.

The U.K. experimented first with a small seismological array installed in Wyoming, U.S.A., which demonstrated that seismic waves from nuclear explosions at distances around 3000-10 000 km could be relatively efficiently detected. In contrast to a standard seismograph station, which houses one or more detectors (seismometers) at a single location, an array consists of a number of seismometers spread out over an area. Computer processing of the recorded data allows the array to be steered like an antenna not only to enhance detection of seismic signals but also to estimate independently the locations from which they came.

The U.K. therefore focused its research attention on this far-distance (teleseismic) range (3000-10 000 km), and re-designed its arrays to have 20 seismometers spread over a region of 25 km diameter. Four such arrays were installed in Scotland, Canada, Australia and India in the early 1960s, and these arrays are still in operation (1989).

In April 1962 the British Ministry of Defence approached the Canadian Defence Research Board about the possibility of locating a seismic array in Canada. Agreement was reached whereby the U.K. would supply and install all equipment and Canada would provide the site, do the necessary construction and, through the Department of Mines and Technical Surveys (now Natural Resources Canada), supply the personnel required to operate the array.

The Yellowknife area was selected for the array because of its location with respect to known nuclear test sites, its remoteness from coastlines, urban areas and other sources of cultural seismic noise, its good communications facilities, and its location on the stable Canadian Shield. Installation was completed in late 1962.

The original array had 19 seismometer vaults in the form of a cross, with a distance of 2.5 km between vaults. Each seismometer output was used to modulate the amplitude of an audio tone that was transmitted to the control centre on cables suspended from wooden tripods. In the control centre, the signals were recorded on 24-channel FM magnetic tape, each reel holding 3 days of data. Power from the control centre to the vaults was supplied via the same cables.

A major problem encountered by this original array was cable maintenance. Severe lightning strikes could cause enough cable breaks and equipment damage to put the array out of operation for several days. This problem was solved in 1971 by replacing the signal cables with VHF radio links between each vault and the control centre, and by installing a propane-fuelled thermo-electric power generator at each vault.

Various improvements in processing the seismic data from the FM tapes were made during the early years. However, a major breakthrough came in the early 1970s with the declining cost and increasing power of small computers. In 1974 the FM tape recording system was replaced by an on-line computer which was programmed to automatically detect seismic signals, steer the array to locate the source of the signals, and store the data on digital tape. The array operated in this configuration until the refurbishing completed in mid-1989.