The Role of Images in Astronomical Discovery Page 24
early 1930s. It is a fine story.
The world was at the beginning of the radio communication age. Italian electrical
engineer Guglielmo Marconi (1874–1937) had developed the first radio transmitters and
receivers in the 1890s. Marconi shared the 1909 Nobel Prize in Physics with Karl Ferdinand
Braun for the development of “wireless telegraphy.” The propagation of radio waves was
poorly understood. Jansky was investigating the atmospheric static that was affecting long-
distance radio communications. Lightning from local storms or thunderstorms occurring
in distant tropical regions produces a static radio noise that outbalances man-made signals.
Jansky was searching for quieter periods (and directions in the sky) and wavelength win-
dows that would improve and optimize transatlantic radio communications. He was also
testing for directions of antenna where the reception of the man-made radio signal would
be enhanced. However, there were other radio signals. In addition to the thunderstorm static
that he recorded carefully, Jansky found a mysterious “hiss.”
Observing at the long radio wavelength of 14.6-m (frequency of 20.5 MHz), Jansky
established that the radio hiss had a periodicity of 23 hours 56 minutes. He noticed that
this corresponded with the sidereal period of the Earth’s rotation, the time it takes the Earth
for its full rotation with respect to the stars. Certain enough of the celestial origin, Jansky
referred to an extraterrestrial origin for this signal in his 1933 paper. By 1935, he had
accumulated more data and was able to attribute the source of “interstellar interference”
to the Milky Way.6 The radio signal with the sidereal period was the strongest when his
merry-go-round antenna pointed in the direction of the Sagittarius constellation where the
center of the Milky Way, highest concentration of stars, is located (Fig. 7.2). For the first
time in human history, electromagnetic signals from the cosmos other than light were being
recorded.
Jansky’s story is fascinating. Like several other discoveries in radio astronomy, his find-
ing had been serendipitous. Two other remarkable serendipitous findings are the cosmic
5 J. S. Hey, Solar Radiation in the 4–6 Metre Radio Wavelength Band, Nature, 1946, Vol. 157, pp. 47–48.
6 K. G. Jansky, Electrical Disturbances Apparently of Extraterrestrial Origin, Proceedings of the Institute of Radio Engineers, 1933, Vol. 21, p. 1387; Karl G. Jansky, Note on the Source of Insterstellar Interference, Proceedings of the Institute of Radio Engineers, 1935, Vol. 23, pp. 1158–1163.
14:06:02, subject to the Cambridge Core
.009
154
Part II – Images as Galaxy Discovery Engines
Fig. 7.2 Transformational Image: Karl Jansky’s Radio Map of the Sky. Karl Jansky pointing to
a portion of the Milky Way in the direction of the constellations of Cassiopeia and Cygnus where he
detected enhanced cosmic noise shown as long wiggly lines. Credit: Image courtesy National Radio
Observatory/AUI.
background radiation by Arno Penzias and Robert Wilson in 1964 and pulsars by the
Northern Irish astronomer Jocelyn Bell and British astronomer Antony Hewish in 1967. As
noted by British historian John North, “only gifted observers make chance discoveries.”7
The radio engineer Jansky was one of these gifted researchers and he had opened a new
and extraordinary window into the universe. Within decades, many other sidereal objects,
7 J. North, Cosmos: An Illustrated History of Astronomy and Cosmology, Chicago: University of Chicago Press, 2008, p. 661.
14:06:02, subject to the Cambridge Core
.009
7. A Symphony of Waves
155
galaxies in particular, turned out to be rich sources of radio emission. Curiously, radio
astronomy, as a new discipline of astronomy, would take a bit of time to emerge.
Radio Astronomy Slowly Gains Acceptance: Reber the Pioneer
Although publicized in the New York Times on 5 May, 1933, Jansky’s discovery did not draw
any attention from professional astronomers in the least.8 His work remained in the shadows
for a decade. The revelations of the great optical telescopes made in the 1920s and 1930s
had been so transforming that “there was little reason to believe that significant contribu-
tion to our knowledge of the universe could come from other parts of the electromagnetic
spectrum.”9 Nobody thought the radio domain would be significant for astronomy, and even
less that astronomers would one day record spectacular images at radio wavelengths.
It was not until 1944 that the first radio astronomy publication appeared in a professional
astronomy journal. This was the year Walter Baade had published his seminal paper, in the
same volume of the journal, on the resolution of stars in the Andromeda Galaxy and its
companions (Chapter 6). Grote Reber, who was not an astronomer but an electrical engi-
neer and radio amateur, was the author of the groundbreaking paper. He reported on his
mapping of cosmic static in the prestigious publication, The Astrophysical Journal.10 This
time astronomers did pay attention.
Reber’s paper of 1944 had been the result of several years of patient work carried out
in complete independence and isolation. Reber had built a 31-ft-diameter antenna with
2 × 4 lumber and metal sheets in his backyard, as well as the receiver equipment which
he operated from his house (Figs. 7.1 and 7.3). With this rudimentary forerunner of
future large radio telescopes, Reber produced maps of the radio sky at a wavelength near
1.87 m, much shorter than that used by Jansky. Reber observed at night to avoid the strong
man-made radio interferences during the day, which came especially from automobile
engine sparkplug firing.
Reber’s work and findings were intriguing, challenging professional astronomers who
were still on the defensive. In an unusual investigative move, the young astronomers Jesse
Greenstein (1909–2002) and Gerard Kuiper (1905–1973) of the University of Chicago
went to Reber’s house in Wheaton. Following their visit and inspection of the facility, they
reported back to Otto Struve, then editor of The Astrophysical Journal, at the University
of Chicago. As faithful reporters in the field, Greenstein and Kuiper commented that the
equipment “looked modern” and that Reber’s work appeared “genuine.”11
Reber’s radio maps of the sky were in the form of isolines, which trace the constant-
intensity levels in terms of 10−22 watts/square centimeter, per circular degree and per MHz
band (Fig. 7.4). It is quite remarkable that Reber had carefully quantified the intensities (the
8 R. Smothers, Commemorating a Discovery in Radio Astronomy, The New York Times, June 9, 1998.
9 Sir B. Lovell, The Story of Jodrell Bank, New York: Harper & Row, 1968, p. 21.
10 Grote Reber, Cosmic Static, The Astrophysical Journal, 1944, Vol. 100, pp. 279–287.
11 Cited in K. I. Kellerman, Grote Reber’s Observations of Cosmic Static, in The Astrophysical Journal: American Astronomical Society Centennial Issue, 1999, Vol. 525, p. 372.
14:06:02, subject to the Cambridge Core
.009
156
Part II – Images as Galaxy Discovery Engines
Fig. 7.3 Grote Reber’s radio telescope in his backyard at Wheaton, Illinois. Credit: Image courtesy
National Radio Observatory/AUI.
<
br /> 14:06:02, subject to the Cambridge Core
.009
contour
asnw
sho
are
atad
The.
ay
Wy
ilk
M
the
of
I.
Maps
U
adioR atory/A
s
Observ
Reber’
te
adio
ro
R
G
National
Image:
courtesy
ormational
Image
ransfT redit:C7.4
Fig.
maps.
14:06:02, subject to the Cambridge Core
.009
158
Part II – Images as Galaxy Discovery Engines
units must have baffled the optical astronomers of the time). The numbers on the central
horizontal line indicate the right ascension (east–west direction) and the numbers around
the circle perimeters, the declination (north–south direction). To emphasize the seriousness
of his experiment, Reber included photographs of the radio dish, the power supply and the
automatic recorder in the first two illustrations of his article. Then followed three pages
of chart recordings of the 1.87-m radio signal corresponding to different declinations. The
chart speed was 6 inches per hour as Reber’s antenna moved only in declination.
Making Radio Waves
What caused the radio emission that was detected by Jansky and mapped by Reber? It is
produced by free electrons moving through space at speeds close to the velocity of light as
they spiral around and along the lines of the magnetic field threading its way through the
tenuous interstellar gas clouds (Chapter 6). As the magnetic field and gas are concentrated
in the plane of the Milky Way, it is then not very surprising that cosmic radio emission
coincides with the luminous band of the Milky Way. It is also where the sources of energetic
particles, novae, supernovae and evolved stars, are more abundant.
Electromagnetic radiation produced by relativistic electrons is called synchrotron radia-
tion; relativistic refers to the speed of particles being close to the speed of light. This unusual
type of radiation was identified in a General Electric laboratory accelerator in 1947.12 A few
years later, Swedish physicists Hannes Alfvén (1908–1995) and Nicolai Herlofson (1916–
2004) showed that electrically charged particles trapped in the magnetic field surrounding a
star could be accelerated in a similar way.13 Later in the 1950s, well after Jansky’s discovery
and Reber’s observations, Soviet astrophysicists Vitaly Ginzburgh (1916–2009) and Iosif
Shklovsky (1916–1985) developed a detailed mechanism explaining synchrotron radiation
and its properties.14
This non-thermal emission is distinct from the black-body emission that arises from
any physical object emitting naturally by just being at a given temperature. Synchrotron
radiation is a continuum of electromagnetic emission; the wavelength where it peaks in
the spectrum depends on the energy of the particles and the strength of the field (Plate
7.2). The behavior of synchrotron radiation intensity as a function of wavelength is differ-
ent from black-body radiation such as is emitted by the Sun. Hence, by making intensity
measurements at various wavelengths, the synchrotron radiation can be identified from its
characteristic spectral shape. Synchrotron radiation is also strongly polarized, that is, the
observed intensity varies with the angle to the plane of the sky. Cosmic synchrotron radia-
tion is observed most commonly in the radio region, but very energetic electrons spiraling
in strong magnetic fields can also produce synchrotron emission in the visible domain and
even X-rays. For example, powerful man-made light sources produce synchrotron radiation
12 F. R. Elder, et al., Radiation from Electrons in a Synchrotron, Physical Review, 1947, Vol. 71, pp. 829–830.
13 H. Alfvén and N. Herlofson, Cosmic Radiation and Radio Stars, Physical Review, 1950, Vol. 78, p. 616.
14 I. S. Shklovsky, Cosmic Radio Waves, Cambridge, MA: Harvard University Press, 1960.
14:06:02, subject to the Cambridge Core
.009
7. A Symphony of Waves
159
as very intense beams of visible light or X-rays that are used to sample all kinds of earthly
materials.
Greber’s pioneering work came at the time of huge developments in radar and electron-
ics, accelerated by World War II. Benefiting from these developments and pushed by rising
interest from professional electrical engineers and astronomers, the new discipline of radio
astronomy blossomed. Surfing on the post-war technological wave, research teams success-
fully overcame several technical hurdles to use radio waves to explore our universe through
a revealing new cosmic window and to make new kinds of images altogether.
Advantages and Limits of Radio Waves
Radio telescopes are analogous to optical telescopes in the way they operate. Radio waves
coming from the sky are collected by an antenna, often but not necessarily a parabolic
reflector, and sent to a focal plane where a receiver (instead of a photographic plate) regis-
ters them. Nonetheless, there are some important differences. Single-dish radio telescopes
measure the electromagnetic signal originating from only one point at a time. Several point-
ing directions are required to make a meaningful radio map of a region of the sky. The image
of an extended source is then assembled by joining points of equal brightness (isocontours)
of radio signal intensity, as Reber did (Fig. 7.4).
Building radio images this way requires measurements of many contiguous spatial ele-
ments in the sky, each converted to a point measurement, a picture element, i.e. a pixel. The
smaller the pixels, the finer the details of the image, and the greater the number of pixels,
the larger the images are. Thus, making images or maps at radio wavelengths has for a long
time required a huge amount of observing time and data reduction work. At the beginning
of the book, I wrote: we read text; we read images. We also write text, and techniques of
imaging with radio waves get close to what could be described as “writing” images. To
overcome the obstacle of the time-consuming observing process for assembling images,
radio astronomers had to come up with astute tactics. To understand how one gets around
tedious point-by-point mapping, let us first review a few elements of wave physics.
A fundamental property of electromagnetic waves fixes the angular resolution that can
be obtained using a given collector. For a lens, mirror or collector of a given aperture,
the highest angular resolution, or finest detail, that can be resolved is set by the Raleigh
criterion: θ = 1.22 λ/ a, where λ is the wavelength of the wave and a is the diameter of the collecting area, in the same physical unit, θ being the angular resolution expressed in
radians. A radian is a unit of angular measure corresponding to the radius projected as
an arc on a circle: one radian is 57.3 degrees. A 1- degree angle divides into 60 arcminutes (shortened to arcmin) where each arcmin splits into 60 arcseconds (shortened to arsec). For
example, the human eye with a 3–9-mm v
ariable pupil has an angular resolution of about
1 arcmin; for scaling purposes, the Sun and the Moon viewed in the sky are each about
30 arcmin in angular diameter. When using binoculars of, say, 8 × 35, which have lenses
of 35 mm in diameter, an angular resolution of about 10 arcsec can be achieved; with this
simple equipment, the main craters on the surface of the Moon can easily be distinguished.
14:06:02, subject to the Cambridge Core
.009
160
Part II – Images as Galaxy Discovery Engines
The Hubble Space Telescope has a primary mirror that is 2.4 m in diameter; its resolution
is about 0.1 arcsec, which is 600 times finer in angular size than that of the human eye.
From the Raleigh criterion it follows that large beam sizes result in less resolution. Radio
telescopes of tens of meters in diameter appear huge, but they have beam sizes, i.e. reso-
lutions, of a few arcminutes. This is very coarse because radio telescopes operate at much
longer wavelengths than optical telescopes. Astronomers can detect and observe galaxies in
the radio domain, but the large beam sizes (i.e. low resolution) of radio telescopes prevent
structural studies of distant external galaxies.
In order to achieve higher angular resolutions, such as those obtained at optical wave-
lengths, collecting areas of tens of kilometers in size would need to be built, which is
impractical and technically unachievable if one were to do this with a single monolithic
parabolic dish.
Tricks for Making Finer Radio Images
To get around this fundamental limit, radio astronomers and engineers decided to mimic
large apertures by a sort of technical cheat. With their usual knack, they invented an astute
technique, aperture synthesis. The principle is that it is not necessary to work with filled
or monolithic apertures, as with traditional parabolic mirrors; it is only necessary for the
collecting surfaces to occupy patches of a large virtual area and to move them around. To
understand the trick, let us make a detour into a few other fundamental aspects of imaging
with electromagnetic waves.
Waves have interesting and useful properties. For example, waves of all sorts give rise to
interference: amplifying when the peaks of the waves superpose (constructing interference)
and canceling when peaks coincide with minima of the other wave (destructive interfer-