Figure 1: Some interferometric data for 3C99.
Click on the image for a full size version.
Figure 2: A dirty map of 3C99.
Click on the image for a full size version.
Figure 3: The dirty beam for observations of 3C99.
Click on the image for a full size version.
Figure 4: Cleaning the map of 3C99.
Click on the image for a full size animation of the cleaning
process.
Figure 5: A close-up of the final cleaned map of 3C99.
Click on the image for a full size version.
Figure 6: The optical image of 3C99.
Click on the image for a full size version. |
Introduction
The dish of a radio telescope works much like the mirror of an
optical telescope. The radio waves arrive from space, are bounced
off the surface of the dish and are focussed onto a piece of
electronic equipment called the receiver. The receiver converts
the radio waves into electrical signals that can be measured with
other electronic devices.
Astronomers want to look at objects in the sky that may be both
very small and very faint. By increasing the collecting area of
a radio dish, i.e. making it bigger, we can increase the amount
of radio waves it reflects onto the receiver. This improves the
sensitivity of the telescope so we can see fainter objects. The
amount of detail that a telescope can discern is called its
resolution. Technically, the resolution is the minimum separation
that two points in the sky have to be in order for the telescope
to be able to distinguish them. The resolution of a telescope also
depends on the size of the dish. So astronomers can increase both
their sensitivity and resolution simply by building larger and
larger telescopes. However, there is a limit to the size we can
build telescopes because we need to be able to point them and steer
them very accurately. A telescope much bigger than the 76-m Lovell
telescope at Jodrell Bank would be too cumbersome to move around.
However, radio astronomers, including those at Jodrell Bank, have
devised sophisticated methods of improving resolution without having
to build cumbersome radio dishes.
Interferometry
The technique is known as interferometry. It involves linking small
radio dishes together to act like a bigger dish. To understand how
this works requires a knowledge of how a single radio dish operates.
Imagine the surface of the dish split up into separate segments.
Each segment reflects the radio waves to the focus where they are
superimposed. In effect the signal received is the combined signal
from each of the segments. If we have only two segments we could
simulate the effect of a larger dish by moving one segment around
the other and adding together all the combinations of signals. Now
imagine that instead of separate segments of a single dish we use
separate dishes. If we move these dishes and add together all the
signals we can synthesise the signal that would be achieved with a
single large dish. However, instead of moving each dish we can allow
the rotation of the Earth to change their orientation with respect
to the object we are observing. This technique is known as
Earth-rotation aperture synthesis. The signals from each pair of
dishes are multiplied and accumulated in a process called
correlation. If we sample the signal with dishes in every possible
position we would produce the image that would be obtained with a
completely filled-in dish. The result, however, is never that good
because there are always gaps in the synthesised dish. However, the
resolution, like other telescopes, is dependent on the size of the
synthesised dish. This size is equal to the largest separation, or
baseline, of all the individual dishes.
Cleaning
An interferometer consists of two or more antennas acting in tandem
with one another. Each antenna receives radio waves which vary
periodically just like waves on the sea. The signals received from
each pair of antennas are multiplied together. This gives another
varying signal or interference pattern which consists of an amplitude
and phase. As the Earth rotates each separate pair of antennas
measures the amplitudes and phases of the interference pattern for
many different orientations. It is these measurements, the
interferometric data, which can be used to create an image of the
object being observed. Figure 1 shows some
example interferometric data taken with the MERLIN instrument for
the radio galaxy 3C99. The structure of the interference pattern,
and how it changes with time as the Earth rotates, reflects the
structure of radio sources on the sky. To get the image of the
radio source the interferometric data are Fourier transformed,
that is, a complex mathematical procedure is carried out on them.
When an interferometer samples the interference pattern it provides
data from a synthetic aperture. But because we have not sampled the
interference pattern at every possible position for each pair of
telescopes we do not have all the information we need to get a good
sharp image. If we transform the data into an image we get a dirty
map which is distorted because of the missing data.
Figure 2 shows the dirty map of 3C99 that if formed by
transforming the interferometric data. Dirty maps have
characteristic ring or striping features on them. However, it is
possible to correct for this effect using a technique called
cleaning. It turns out that the sampling of the interferometer, when
transformed into what radio astronomers call the dirty beam, can
be removed from the data. Figure 3 shows
the dirty beam for the observations of 3C99. Cleaning results in
a sharper image called the clean map which is free of distortions.
It is as if we have filled in all the gaps in the synthetic image
by interpolating the data we have. This process of cleaning radio
interferometric images is an iterative process; that is, it is
repeated many times, each time improving on the previous iteration.
Eventually the sharp image of the radio object is revealed as the
gaps in the data are filled in. Figure 4
shows an animation of the cleaning process on the image of 3C99.
3C99
The final cleaned image of the radio source 3C99 shows a one-sided
jet from the core to the north-east. Such jets are thought to be
formed by matter racing away from the core of the host galaxy at
speeds close to that of light. To power such jets astronomers believe
that supermassive black holes exist in the cores of these active
galaxies. A close-up of the final cleaned image is shown in
Figure 5. The compact feature to the
north-east is a knot in the jet. These are formed when the jet
hits dense regions of matter on its journey from the core of
the galaxy. The base of the jet to the south-west runs into the
core. The total length of the jet is about 1.5 arcseconds or about
15 kpc at a redshift of 0.5. In optical light 3C99 is an elliptical
galaxy at a magnitude of 19.4 and a redshift of 0.426. The optical
image of 3C99 taken with the UK Schmidt telescope is shown in
Figure 6. Note the vast difference between
the optical and radio images. The radio structure lies within the
optical extent of the galaxy.
|