IMAGINE a telescope that could tell the story of the Universe’s past and
future. Peering into deep space, it would reveal exactly when time began, and
how much matter is tied up in stars and galaxies. It would report back on how
much matter is hidden away as invisible, dark matter. Most sensationally of all,
it would end all speculation about whether the Universe will expand forever into
the cosmic night or shrink back down to a “big crunch” at the end of time.
Wishful thinking? Not according to NASA and the European Space Agency, who
have just such space telescopes on their drawing boards. NASA plans to launch
MAP—the Microwave Anisotropy Probe—in 2000. And at a meeting in
November, ESA looks set to give the green light to the COBRAS/SAMBA mission, for
launch in 2004. Both will measure the cosmic background radiation, the faint
afterglow of the big bang fireball in which the Universe was born between 9 and
14 billion years ago, with unprecedented accuracy.
These observations will pin down the principal cosmological parameters that
determine the kind of Universe we live in—parameters such as the expansion
rate of the Universe and its overall density. “Currently, these parameters are
uncertain by a factor of two or more,” says Charles Bennett of NASA’s Goddard
Space Flight Center in Greenbelt, Maryland, leader of the MAP scientific team.
“We think we can nail them to within a few per cent.”
Advertisement
MAP and COBRAS/SAMBA will build on the enormous success of NASA’s Cosmic
Background Explorer (COBE), launched into Earth orbit in 1989 to observe the
background radiation. “The background radiation is an absolute gold mine of
information,” says George Efstathiou of the University of Oxford. “The new
satellites are going to totally revolutionise cosmology.”
Foggy Universe
The background radiation is a relic of the early Universe, a searing fireball
that contained vast numbers of free electrons, nuclei and photons. As the
fireball expanded, the photons were constantly scattered in collisions with the
electrons, forcing them to follow crazy zig-zag paths through space. The light
was effectively trapped, and the Universe would have been like a dense,
impenetrable, glowing fog.
But around 300 000 years after the big bang, the temperature in the expanding
fireball had fallen to about 3000 °C. Atomic nuclei in the cooling fireball
then started to mop up the free electrons to make atoms (see
Diagram).
Without the electrons, the fireball’s light was able to fly through space
unhindered. For the first time, the Universe was transparent—as it has
been ever since—allowing light from the fireball to escape through space.
We see it today as the ubiquitous cosmic background radiation, stretched to long
wavelengths (“red-shifted”) by the expansion of the Universe.FIG-20524001.gif
There are subtle variations of the energy, or temperature, of the photons of
cosmic background radiation, depending on where you look in the sky. Dubbed
cosmic ripples, these were first observed in 1992 by COBE, which found that the
radiation coming from some patches of the sky was about a hundred-thousandth of
a degree cooler or warmer than the average of 2.726 kelvin (This Week, 2 May
1992, p 4).
The cold patches correspond to regions of the early Universe that were very
slightly denser than average at the time the Universe became
transparent—the “epoch of last scattering”, as cosmologists call it. In
escaping the gravitational pull of these regions, photons lost energy or were
red-shifted, lowering the temperature of the radiation. As the Universe
expanded, the regions of enhanced density would have pulled in more and more
matter, and eventually become immense conglomerations of galaxies. By measuring
the fluctuations of background radiation at points in the sky more than 10°
apart, COBE found the seeds of what were to become the giant clusters of
galaxies in today’s Universe.
The curious thing about these background radiation fluctuations, however, is
that they are too big to have been crossed by light in the time since the
beginning of the Universe to the epoch of last scattering. This leads to the
paradoxical conclusion that no process operating since the big bang would have
had sufficient time to give birth to them. Creating the over-dense regions found
by COBE at the epoch of last scattering was as impossible as getting a line of
deaf and blind soldiers to all salute when there has been insufficient time for
someone to run the length of the line and tap each one on the shoulder.
One explanation for this conundrum is that the dense regions found by COBE
originated as quantum fluctuations in the vacuum which were then inflated by an
extraordinarily rapid expansion of the Universe during its first split second. A
tiny region crossable by a ray of light could suddenly be inflated to a
much larger size by this expansion. What COBE saw may well have been the
inflated structures that lived on from the first moments of creation.
MAP and COBRAS/SAMBA will provide further insights into the Universe at a
later time—the processes actually happening at the epoch of last
scattering—by observing cosmic ripples on much smaller scales than COBE.
To see how a smaller scale could improve our view, think in terms of an
imaginary sphere—the so-called Hubble sphere—whose diameter at any
epoch is the distance light could have travelled since the moment of creation.
For instance, when the Universe was 1 second old the Hubble sphere was a light
second across. Events closer together than the diameter of the Hubble sphere can
affect each other, whereas events farther apart cannot.
Last scattering
The Hubble sphere at the epoch of last scattering had a diameter of 300 000
light years, corresponding today to a patch of sky about 1° across—twice
the apparent size of the Moon, but much smaller than COBE’s scale. By observing
at scales down to a fraction of a degree, MAP and COBRAS/SAMBA will focus on
what was happening around the time that the background radiation broke free of
matter.
Both MAP and COBRAS/SAMBA will have a chance to see the footprints of the
changes going on in the Universe at that time. Before then, the scattering of
photons by electrons effectively coupled matter and radiation to make a single
gas of photons and the heavy particles or “baryons”, such as protons and helium
nuclei. The pressure from the photons was tremendous and would have prevented
slightly denser regions of the normal matter in the fireball from collapsing
under their own gravitational pull.
However, once the nuclei had mopped up electrons (and matter and radiation
had parted company) the contribution of photons to the gas pressure vanished.
Clumps of ordinary matter all over the Hubble sphere were suddenly free to begin
shrinking to form the structures we see in today’s Universe. Before this time,
theory suggests that the masses in regions of enhanced density would have
started to collapse, rebound, then collapse again—in other words, they
would have oscillated.
This means that just before the epoch of last scattering, the photon-baryon
gas was sloshing back and forth like water in a bath. Standing waves would have
been set up with fixed peaks and troughs stretching across the Universe.
Crucially, these waves would have left their imprint on the background radiation
as a series of peaks in the temperature variation at small angular
scales—smaller than COBE was able to measure.
Since the biggest standing wave possible will just fit within the Hubble
sphere at the epoch of last scattering, it will be responsible for the first
peak—that is, the one at the largest angular scale (see
Diagram).
Smaller standing waves, with many peaks and troughs across the Hubble sphere,
will create peaks at smaller and smaller angular scales. And imprinted on the
sizes and positions of these peaks is all the information that cosmologists are
looking for.FIG-20524002.gif
Especially interesting is the angular scale at which the first peak appears.
This depends on the size of the Hubble sphere at the epoch of last
scattering—the speed of light multiplied by the age of the Universe at the
last scattering. However, the angle may differ from the theoretical prediction
because light en route from the epoch of last scattering to the present day is
bent by the curvature of space-time. The curvature, in turn, depends on the
total matter content of the Universe—a quantity enshrined in the
cosmological parameter &OHgr; (the Greek letter “omega”), the ratio of the density of
the Universe to the critical density needed to eventually halt its
expansion.
If &OHgr; is less than 1—that is, the density of the Universe is less than
the critical density—the gravity of all the matter in the Universe will be
too weak to ever halt its expansion. The Universe will expand forever, its dying
galaxies becoming increasingly isolated in an ever-growing ocean of space. But
if &OHgr; is greater than 1, the gravity of all the matter in the Universe will be
enough eventually to halt its expansion and turn it into a runaway collapse to a
big crunch—a sort of mirror image of the big bang in which all of creation
is squeezed into a tiny volume.
The third possibility is that &OHgr; is precisely equal to 1, and that the
Universe is balanced on the knife edge between eventual collapse and eternal
expansion. The expansion would slow to a halt at an infinite time in the future.
It turns out that if inflation is correct, then we live in just such a
knife-edge Universe.
But although the value of &OHgr; will be indicated by the angular size of the
first peak, a complication arises if the Universe has a so-called cosmological
constant, a sort of repulsion of empty space. This will also play a part in
bending the path of the big bang fireball photons on their way to us.
Priceless information is also contained in the height, or amplitude, of the
first peak. This depends on two things. The first is the expansion rate of the
Universe, since a rapidly expanding Universe would have accentuated the sloshing
of the photon-baryon gas. The greater the observed amplitude, the greater the
expansion rate must have been. This, in turn, is related to the present-day
expansion rate, or Hubble constant.
Springy matter
But the amplitude of the first peak also depends on the stiffness, or
springiness, of the photon-baryon fluid, since this determines how much it
rebounds in each oscillation. The stiffness, in turn, depends on the mass tied
up in the matter and radiation of the fluid. In fact, the contribution of
radiation can always be deduced—simply by observing how much mass-energy
is tied up in the present-day cosmic background and extrapolating backwards.
This means that the amplitude of the first peak can be used to determine how
much ordinary matter there is in the Universe.
With many of the interesting cosmological parameters being dependent on each
other, the data from the new satellites may be difficult to untangle. But the
first peak is only one of scores of rich veins that MAP and COBRAS/SAMBA will
mine. Efstathiou says that COBRAS/SAMBA will be able to measure 2000 useful
quantities from the other peaks, and MAP will measure a few hundred. He is
hopeful that together they will be enough to pin down the 15 key cosmological
parameters that theorists need to constrain their models of the Universe.
It may even turn out that there are no regularly spaced peaks, as their
existence is dependent on the initial seeds of structure being quantum
fluctuations blown up by inflation. Another possibility is that the seeds of
structure were caused by “defects” in space-time created in the so-called Grand
Unified Theory era (“Cosmic beakers”, New Scientist, 21 September, p
46). It was then that the Universe underwent a phase transition in which
nature’s strong and electroweak forces split apart.
The defects that formed in the phase transition could have taken the form of
strange zero-dimensional “global monopoles”, one-dimensional “cosmic strings” or
three-dimensional “textures”. Since such defects would be scattered randomly
throughout the Universe, their signature would be a far more irregularly spaced
set of peaks.
Nobody knows whether the seeds of structure in the Universe were defects or
quantum fluctuations blown up by inflation. MAP and COBRAS/SAMBA could provide
us with the definitive answer. And if these orbiting observatories live up to
astronomers’ dreams, the biggest questions about the beginning and the end of
time and space could be laid to rest for ever.
The name COBRAS/SAMBA was inspired by two ESA proposals with similar goals:
Cosmic Background Radiation Anisotropy Satellite and Satellite for Measurement
of Background Anisotropies. Any suggestions for a shorter name can be sent to
project scientist Jan Tauber (jtauber@astro.estec.esa.nl).
* * *
A window on the early Universe
The plan is to launch the 700-kilogram MAP on a Delta rocket in 2000 and
the 1.5-tonne COBRAS/SAMBA on Ariane 5 in 2004. Both spacecraft will hover in
the deathly cold of space, about 150 million kilometres beyond the Earth, whose
light would swamp the faint afterglow of the big bang.
MAP’s detectors will be high electron mobility transistors, or HEMTs,
sensitive to microwaves at five wavelengths between 3.3 and 13.6 millimetres.
The HEMTs will amplify the microwaves directly, rather than by the more usual
method of converting them to low frequencies before amplification. “HEMTs are a
spectacular improvement in sensitivity over the detectors used by COBE,” says
Charles Bennett of NASA. “The detectors will make their measurements hundreds of
times faster.”
COBRAS/SAMBA will use 56 HEMT amplifiers sensitive to microwaves between 2.4
millimetres and 1 centimetre. It will also use 56 bolometers, which change their
electrical resistances in response to the heat of far-infrared radiation. This
will extend the satellite’s coverage into the far-infrared region between
wavelengths of 0.3 and 2 millimetres. To maximise the sensitivity, the
bolometers will be cooled by liquid helium to 0.1 kelvin.
“Being sensitive to a wide range of wavelengths is crucial,” says Jan Tauber,
project scientist on COBRAS/SAMBA at the European Space Research and Technology
Centre in the Netherlands. The reason is that the Milky Way also emits in the
microwave and far-infrared region. So strong is the emission in the galactic
plane that it will swamp measurements of the microwave background. But for about
50 per cent of the sky, looking out of the plane, the aim is to predict the
galactic emission and subtract it from the observations. “Subtracting the Galaxy
is the killer,” says Tauber. “If we can’t do this, there’s no way we will be
able to measure the cosmic background.”
COBRAS/SAMBA will probe smaller angular scales than MAP—down to 0.1°
compared to MAP’s 0.3°. This will allow it to measure several thousand features
of the ripples compared to MAP’s few hundred. But this strategy will not be
without problems. “On small angular scales, the satellite’s measurements will be
contaminated by unknown extragalactic sources,” says Bennett. “Subtracting their
contribution to obtain pure background radiation will be a challenge.”
MAP will focus the background radiation onto its detectors with two
back-to-back telescopes, comparing the temperature at points 180° apart in the
sky. “Measurements that involve comparisons are always easier than direct, or
`absolute’ measurements,” says Bennett. “They were also proven to work on
䰿.”
COBRAS/SAMBA will instead use a single telescope to focus the radiation. Its
HEMT amplifiers will make absolute measurements by comparing the temperature of
the sky with a resistor held at a known temperature. “It’s a high-risk path,”
says Bennett. “Their temperature standard will have to be very stable.”
Another difference between the satellites is their cost. MAP is a relatively
low-budget mission at $70 million, while the more complex COBRAS/SAMBA
satellite will cost $350 million and will include various backup systems
in case of failures. “MAP is a cheap and cheerful middle-range Explorer
satellite with no built-in redundancy,” says Bennett. “One system goes and we’re
.”
Several ground-based experiments could also observe the background radiation
on small scales. For instance, the Anglo-Spanish Very Small Array will be built
at a cost of £2.6 million on top of a volcano in the Canary Islands. The
VSA will map the radiation using 15 microwave horns fixed to a turntable just
1.5 metres across.
Ground-based detectors must contend with atmospheric water vapour, which
emits strongly at shorter wavelengths. At longer wavelengths, our Galaxy emits
confusing radiation. The wavelengths chosen for the VSA are a compromise to
minimise these problems. It will be built by astronomers at the University of
Cambridge, Jodrell Bank in Cheshire and the Astrophysical Institute of the
Canaries, and should be working by 2000.