Sir Francis Graham-Smith, who has died aged 102, was the last of the generation that created modern radio astronomy, the branch of astronomy that studies the universe with radio waves, in the 1940s and 50s. His PhD thesis, on the first Cambridge radio survey, carried out between 1948 and 1950, with reasonably accurate positions for the brightest sources, paved the way to demonstrating that the majority of celestial radio sources are distant galaxies with massive black holes in their nuclei.
Following the discovery of pulsars, pulsating radio sources associated with rapidly rotating neutron stars, in 1967 by Antony Hewish, Jocelyn Bell and others, Graham-Smith used the Jodrell Bank Mark I telescope to study pulsars in detail. He and Andrew Lyne wrote the definitive book on the subject, Pulsar Astronomy (1990).
Graham-Smith’s first radio survey was carried out with a pair of captured German Wurzburg radar antennae used as an interferometer, where the differences in the signals from the two antennae are used to construct an image of the sky. There were four really bright sources, named for the constellations they fell in: Cassiopeia A, Taurus A, Cygnus A and Virgo A, and Graham-Smith gave the first accurate positions for these in a 1951 paper in the journal Nature. The importance of these positions for the identification of the powerful radio galaxies Messier 87 and Cygnus A launched his career.
However the survey was not an unmitigated success. Most of the rest of the 50 sources listed eventually turned out to be spurious. In addition, in their 1950 survey paper, Martin Ryle, Bruce Elsmore and Graham-Smith, while considering the possibility that the radio sources were extragalactic, opted for the interpretation that they were some kind of active star. By the time of the second survey in 1955, the Cambridge group had changed their minds about this and now believed the sources were mostly extragalactic.
This 1955 survey gave extraordinarily steep counts of sources as they looked towards fainter fluxes. Ryle claimed that this proved that we must live in an evolutionary universe, in contradiction to the steady-state theory advocated by Fred Hoyle and others. Unfortunately the survey was also strongly affected by spurious sources generated by the interferometer, and it was only with the third Cambridge survey in 1962 that the correct source-counts were measured. Graham-Smith was involved with the development of the more accurate interferometers used for the second and third surveys, and with the third survey, the revolution initiated by Graham-Smith’s 1951 paper on interferometric radio positions could now begin.
When the first ever satellite, the Soviet Sputnik, was launched in 1957, Jodrell Bank famously tracked the launch rocket casing, but it was Graham-Smith who used the Cambridge interferometer to track the satellite itself and was able to demonstrate the precession of its orbit due to the Earth’s equatorial bulge.
In 1964 Graham-Smith moved to a professorship at Jodrell Bank, Manchester, taking with him a new PhD student, Andrew Lyne, who was to be his key collaborator for the next 40 years. This turned out to be a very fortuitous move, because three years later Antony Hewish and Jocelyn Bell discovered pulsars at Cambridge. However the ideal telescope for detailed study of pulsars was the 250ft Mark I telescope at Jodrell Bank, and Graham-Smith, Lyne and the Jodrell team threw themselves into this work, becoming world-leading experts in this field. They worked on the relativistic beam (“lighthouse”) model of the pulses, monitoring the long-term slowing-down of the rotation speeds of the neutron stars.
In 1971 the Jodrell Bank Mark I telescope was given a new surface so that it could operate at higher frequencies, and Graham-Smith procured a laser device to map its shape. Lyne said: “He was able to tell the engineers the number of turns required on each of the several hundred adjusting nuts in the backing structure to perfect its shape. This was the radio-astronomer’s equivalent of polishing the mirror of an optical telescope and gave us a wonderful radio instrument. It was a spectacular feat of measurement at that time.”
Born in London, Francis was the son of Claud Smith, a civil servant, and Cicely (nee Kingston). In early scientific papers he was FG Smith and hyphenated his name only later. He attended Epsom college and Rossall school in Fleetwood, Lancashire, and then went to Downing College, Cambridge, in 1941, to study natural sciences. In 1943 he was seconded to work on radar at the Telecommunications Research Establishment at Malvern, returning to complete his degree in 1946.
He began a PhD in Ryle’s radio astronomy group, working at first with simple dipole antennae to study the size of solar flares, and then with increasingly complex radio interferometers to measure the positions of radio sources, culminating in the first Cambridge Radio Catalogue in 1950 and the accurate positions of the four brightest radio sources in the sky in 1951.
The dominance of radio astronomy on the British astronomical scene in the 1960s and 70s was reflected in Graham-Smith’s appointment in 1975 as director of the Royal Greenwich Observatory at Herstmonceux, East Sussex. During his tenure he oversaw the development of the UK observatory on La Palma in the Canary Islands, the shipment of the 2.4m Isaac Newton Telescope to La Palma in 1979, and the plans for what became the 4.2m William Herschel Telescope, which eventually saw first light in 1987.
In 1981, on the death of Sir Bernard Lovell, Graham-Smith moved back to Manchester to become director of Jodrell Bank, in which post he continued till his retirement in 1988. He secured funding for a major upgrade of the Merlin interferometric array, including the construction of the 32m telescope at Cambridge. These endeavours set the scene for Jodrell Bank to play a leading national role in the future international Square Kilometre Array (SKA) telescope, which is now being built in Australia, but has its headquarters at Jodrell Bank.
Graham-Smith delivered the 1965 Royal Institution Christmas Lecture jointly with Lovell, Ryle and Hewish. He was made a fellow of the Royal Society in 1970 and was awarded the society’s Royal Medal in 1987. From 1975 to 1977, he was president of the Royal Astronomical Society and, from 1982 to 1990, he was Astronomer Royal, following Ryle in the post.
In 1986, the year he was knighted, he was unlucky enough to be involved in a live television broadcast of the flypast of Halley’s comet by the European Space Agency’s Giotto mission, during which none of the astronomers present could work out what was happening with the images. This broadcast was said to have angered Margaret Thatcher and resulted in cuts to UK space science funding.
Graham-Smith was the author of several books, including Pulsar Astronomy, which went through five editions, and Introduction to Radio Astronomy (1997), with Bernie Burke, which went through four editions. He was still writing highly cited papers in his 90s.
Throughout his life Graham-Smith was a keen gardener, and for many years an avid bee-keeper. He was looked after by his daughter, Helen, in his last few years.
His wife, Elizabeth (nee Palmer), whom he married in 1945, died in 2021. He is survived by Helen and three sons.