Telescope for a digital age
Astronomers are exploring the largely untapped low-frequency end of the radio spectrum by combining
thousands of antennas with cutting-edge computer technology. Edwin Cartlidge reports on their
“software telescope”
Think of a radio telescope and what is
likely to come to mind is a huge white
dish pointing skyward, such as the
76 m Lovell Telescope at Jodrell Bank
in the UK or the 100 m Green Bank
Telescope in West Virginia. Indeed,
such dish telescopes have been the
mainstay of radio astronomy for the
last 50 years, allowing astronomers to
discover new kinds of celestial objects,
such as pulsars and masers, not visible
at optical frequencies. However, such
dishes do not work well when it comes
to observing the lowest frequency
radio waves – below about 100 MHz –
because their diameter is too small
relative to the wavelength to provide
adequate resolution.
ASTRON
eight different directions at the same
time. As the various stations are
spread out over Western Europe,
combining the signals from all these
different locations provides LOFAR
with its high resolution.
Overcoming the ionosphere
LOFAR so good
To open up this largely unexplored
section of the electromagnetic spectrum, astronomers are developing a
number of facilities consisting of dipole antennas – simple wires that
work just like FM receivers. Since
they are very cheap to make, many
such antennas can be placed together
to create a huge and therefore sensitive telescope at low cost. The only
catch is that these arrays rely on huge
amounts of computing power, with
the signals from the antennas being
digitized, sent to a central processor
and then combined using software
so that they emulate the output of a
conventional telescope.
Radio astronomy was in fact born
at very low frequencies. In 1931 Bell
Telephone engineer Karl Jansky was
investigating the origin of interference in shortwave communications
across the Atlantic when he discovered radio waves at 20 MHz coming
from the centre of the Milky Way. But
because these low-frequency signals
are heavily distorted by the Earth’s
ionosphere, radio astronomers have
since concentrated on the higher end
of the spectrum, where such distortions are less important
More than 50 000
antennas spread
across Europe will
make up the 7100m
Low Frequency Radio
Array (LOFAR).
LOFAR will use antenna divided
into two kinds – low-band antennas
operating from 10–90 MHz and high-band devices operating between 110
and 250 MHz. Several hundred of
each type will be placed together at
a single station, and there will be 18
such stations distributed over several
square kilometres at the core site, currently being built in Drenthe, a province in the north of the Netherlands.
There will be a further 18 stations
within 50 kilometres of this site and
then several other stations in France,
Germany, Sweden and the UK. The
long baselines that result will provide a
high enough resolution that radio
sources can be identified with visual
objects, even at the lowest frequencies.
The largest such telescope that is
currently under construction is the
7100m Low Frequency Radio Array
(LOFAR), which, when complete,
will consist of 50 000 antenna stations
throughout the Netherlands and
nearby countries such as Germany,
Sweden and the UK. Being developed
by a consortium under the leadership
of ASTRON, the Netherlands Institute for Radio Astronomy, LOFAR
will allow scientists to look back to the
formation of the first stars in the universe, scan the skies for rare transient
phenomena, and study high-energy
cosmic rays.
Since the early 1990s, astronomers
have been using computer programs
to correct for such ionospheric interference for an array of dishes.
However, such corrections are harder
to make when radio signals are combined using interferometry from
arrays of dishes separated by up to
several hundred kilometres – a technique that is needed to improve the
otherwise poor resolution of radio
astronomy. It is in order to achieve
high resolutions while also overcoming the blurring effects of the ionosphere that astronomers are now
turning to arrays of dipole antennas.
Michael Garrett, general director of
ASTRON, points out that this project
represents a fundamental change in
the development of radio telescopes.
“It places most of the costs not in some
huge steel parabolic antenna,” he says,
“but in commercially available back-end digital electronics.”
Dish telescopes bring incoming
waves from a particular direction to
a focus at a single point by virtue of
their parabolic geometry. An antenna
array, in contrast, works by calculating
the difference in arrival times of a
wave at neighbouring antennas and
then introducing an equal delay when
combining the signals from these antennas. In fact, by carrying out many
such calculations simultaneously,
LOFAR can observe “beams” from
It places most
of the costs not
in some huge
steel parabolic
antenna, but in
commercially
available back-end digital
electronics
Looking back in time
LOFAR was originally an international consortium, comprising astronomers from Australia, Europe and
the US. However, after disagreements
over where to site the project, the consortium split, with the result that the
Murchison Widefield Array (MWA)
is now being built in Western Australia, with separate plans for similar
facilities in the US and China.
One of LOFAR’s specific aims
will be to study “reionization” – the
period that began a few hundred thousand years after the Big Bang when
the neutral hydrogen filling the universe at that time began forming into
stars and galaxies. These stars and
galaxies then ionized their environment, destroying neutral hydrogen in
the process and leaving a marker for
the evolution of the universe. LOFAR
can observe this neutral hydrogen
because its redshifted emission probably lies at about 140 MHz. However,
more sensitive instruments will be
needed to chart the evolution of this
star-forming period by mapping neutral hydrogen as a function of frequency (i.e. the redshift).
In addition, the fact that LOFAR
can observe eight different patches of
the sky at the same time (and retro-
