We have discovered a new way to detect the first stars when
the Universe was in its infancy at a mere 1% of its present age. Using
powerful computer models we showed that an expected difference in
the speed of gas and dark matter causes the first stars to clump
together into a prominent cosmic web. The discovery of these web-like
structures now makes it feasible for radio astronomers to detect the
21-cm wavelength light from hydrogen that was heated by the first
stars when the Universe was only 200 million years old.
Expanded outline
The formation of stars is a part of our cosmic history. Astronomers
know that long before there were stars, the early universe was filled
with a hot, very uniform gas. In contrast, today we see a complex
universe of stars and galaxies. A great unknown frontier is the era of
the formation of the first stars, which marked the transformation of
the universe to its current state. Currently, the most distant
reliably detected galaxy is from a cosmic age of around 800 million
years, and it is very difficult to go much earlier with detections of
individual galaxies. Since the universe was filled with hydrogen atoms
at those early times, the most promising method for observing the
epoch of the first stars is using the prominent emission of hydrogen
at a wavelength of 21 cm (corresponding to radio waves). Measuring the
cosmic 21-cm emission is difficult, though, due to the foreground
emission from our own Milky Way and other nearby galaxies. However, if
the cosmic signal fluctuates then it is much easier to distinguish it
from the bright local emission.
Indeed, the first stars and galaxies are expected to show large
fluctuations, so that some regions contain many stars (collected into
mini-galaxies, each much smaller than current galaxies such as the
Milky Way), and other regions are nearly empty. The reason can be
understood from an analogy: Imagine searching on Earth for mountain
peaks above 5000 meters. The 200 such peaks are not at all distributed
uniformly but instead are found in a few distinct clusters on top of
large mountain ranges, with none found over the rest of the Earth's
surface. The point is that given a mountain range, every small hill on
top of it becomes a high mountain peak, while in a valley it would be
just a small hill. Similarly, in order to find the early galaxies, one
must first locate a region with a large-scale density enhancement, and
then galaxies will be found there in abundance; the higher overall
density enhances gravity throughout the region and makes it easier
to form high concentrations of dark matter, into which gas falls and
forms stars.
This idea is made more effective by a recent insight that dark matter
and ordinary matter (gas) move at different velocities in the early
universe. The effect of this velocity difference has been studied over
the last two years with analytical models and numerical simulations.
In our paper we produced the first simulated 3-D maps of the
distribution of the first stars and showed that the relative velocity
effect significantly enhances large-scale fluctuations. In particular,
during the era of the first heating of the intergalactic hydrogen by
X-rays associated with star formation, prominent fluctuations are
expected on a very large spatial scale that corresponds to 400 million
light-years in today's universe. This would be observed on an angular
scale of 2/3 of a degree (the sun and moon subtend about 1/2 a
degree), which makes the signal relatively easy to observe (since
exquisite resolving power is not necessary). Thus, this spatial
structure makes it much more feasible for radio astronomers to detect
early stars from a cosmic age of around 180 million years (1.3% of the
current age of the Universe). The expected signal comes with a
characteristic signature that would mark the existence of small
mini-galaxies at that time and the presence of the velocity effect.
This exciting possibility should stimulate observational efforts
focused on this early epoch.
