The low radio frequency survey that imaged the sky in one day

[Nature India Special Issue: Lighting the way in physics]

The low radio frequency sky survey was one of the important research projects at the Raman Research Institute (RRI) during the early 1980s. Its main purpose was to image the radio emission from the Milky Way and from the extragalactic sources as accurately as possible. By the end of the 1980s, it was possible to produce a survey of the sky at low radio frequencies by using the Gauribidanur radio telescope working at 34.5 MHz (a wavelength of about 8.7 metres).

Radio astronomy began at low frequencies in the early 1930s when Karl Jansky used an antenna that received radio waves at 20.5 MHz. The quest for better angular resolution soon saw radio astronomers using much higher radio frequencies – about 1,000 MHz – since it was easier to achieve higher angular resolutions with relatively smaller apertures at shorter wavelengths.

There were many attempts to image large regions of the sky at different sites around the world, including an array at 38 MHz in Cambridge in the UK during 1966, the Penticton array in Canada at 10 MHz during 1976 and the Clark Lake array in the United States at 26 MHz during 1988. These observations were plagued with problems such as varying instrumental and ionospheric conditions and interference from terrestrial radio signals over the days, weeks and months of observing the sky. They were also unable to image the extended radio emission from the Milky Way.

The sky survey at RRI set out to overcome these shortcomings. The dipole array situated at Gauribidanur was 1.4 km along the east-west arm and 0.5 km along the south arm. Special multi-channel receivers were built to record signals from each of the 90 dipoles along the south array and were multiplied by the signal from the east-west array. This novel technique allowed the entire observable sky to be imaged in just one day.

The effects of solar and ionospheric activities and terrestrial radio frequency interference were minimised by choosing days that were most suitable for the observations. Furthermore, a novel technique adopted in analysing the images minimised the instrumental response in the images.

This survey was a unique opportunity for students at the RRI to learn and master radio astronomy. The students built and successfully used all the hardware and software needed for the survey, which gave them invaluable experience that went way beyond what they could have learnt from books. The students who worked on this project went on to lead research activities in diverse areas of astronomy all over the world.

This survey firmly established low frequency radio astronomy in India and led to RRI’s involvement in building the Mauritius radio telescope working at 150 Mhz and in contributing to building the Murchison Widefield Array in Western Australia working in the range of 80–250 MHz.

The RRI survey produced a reliable image of the sky at low frequencies with an angular resolution of about half a degree by half a degree (about the size of the full moon). It has become one of the reference points in the study of radio emission from the Milky Way at low frequencies.

As an illustration of how low the frequency used in the survey was, I recall a conversation with an astronomer who worked at the Karl G Jansky Very Large Array (VLA) in the United States in the early 1990s. When I told him that I had worked at low frequencies, he assumed that I was an expert working in the VLA L-band. The L-band referred to radio frequencies around 1,500 MHz. I sheepishly told him that I had worked at 30 MHz — only 50 times lower than the VLA L-band.