- :: Atom & Cosmos
- :: Body & Brain
- :: Earth
- :: Environment
- :: Genes & Cells
- :: Humans
- :: Life
- :: Matter & Energy
- :: Molecules
- :: Science & Society
- :: Other Topics
- :: Science News For Kids
A newly discovered “dark flow” appears to carry clusters of galaxies toward a point in the southern sky, a new study suggests.
As if dark matter and dark energy weren’t confusing enough, researchers detected what they have dubbed dark flow while surveying 700 galaxy clusters — each containing hundreds to thousands of galaxies — within a radius of approximately 1 billion light-years. On average, the clusters appeared to move in a uniform direction at about 1,000 kilometers per second.
While no one knows the cause of the motion, the scientists suggest that whatever it is may no longer lie within the visible universe. The work appears online in two separate papers, one to appear in the Oct. 20 Astrophysical Journal Letters and the other in an upcoming Astrophysical Journal.
“We expected to find something completely different,” says
Alexander Kashlinsky, an astrophysicist at NASA’s
The result flies in the face of one of cosmologists’ most cherished assumptions — backed by a vast wealth of data — that the universe is uniform. That is, its structure and the density of matter in it are about the same in all regions of the sky.
But the findings further complicate the picture of
cosmology, comments cosmologist Glenn Starkman of
Harald Ebeling, an astronomer at the
The researchers’ work built upon a survey of the entire sky in the X-ray spectrum taken by the orbiting telescope ROSAT in the early 1990s. Galaxy clusters are usually suffused in a thin but very hot plasma, which emits X-rays. Back then, Ebeling and others used the ROSAT data to identify hundreds of large galaxy clusters by their X-ray halos, and matched that with optical-telescope data to estimate the clusters’ distance from Earth.
In the new study, the researchers estimated the motion of each cluster with respect to the CMB radiation, which is believed to be “the ultimate reference” of movement on a cosmological scale, says Ebeling.
As CMB radiation crosses a galaxy cluster, it gets scattered by electrons in the intergalactic plasma, Ebeling says. The scattering affects the radiation’s frequency. The frequency goes up if the cluster is moving toward Earth, and down if it’s moving away. This is called the kinetic Sunyaev-Zeldovich effect, analogous to the familiar Doppler shift of sound waves. The Doppler shift explains why the pitch of an ambulance’s siren sounds different depending on whether the ambulance is approaching or moving away from the listener.
The researchers looked for the kinetic Sunyaev-Zeldovich effect in CMB data released two years ago by NASA’s Wilkinson Microwave Anisotropy Probe mission. The effect was extremely small — comparable to a temperature change of millionths of a kelvin, Ebeling says.
For a single cluster, a variation this small easily drowns in the much larger experimental errors. Moreover, each cluster tends to move in its own direction, tugged by clusters nearby. But on average, the velocities showed a clear trend. “The velocity is not only high,” Kashlinsky says, “but it also remains the same velocity as far as you can see.”
“People will be inherently skeptical of any such results,” Starkman says, since they call into question the standard, homogeneous model of the universe. “Even those who have doubts about the model don’t have better alternatives.” But, he adds, researchers should still take the results seriously.
Kashlinsky says that random energy fluctuations in the earliest split second of the Big Bang — the epoch of stupendous expansion called inflation — could have created a large imbalance in the distribution of matter. While the denser regions of the universe would now be forever out of sight, the imbalance could have left its mark on the overall structure of spacetime. Like a dining room table tipped so that all the dinner plates slide off in the same direction, the imbalance may have put the local corner of the universe on a slope.
Such a large-scale imbalance is “absolutely possible,” says
cosmologist Andrei Linde of
The results are something that people will “scratch their
heads over,” says Ethan Vishniac of
Found in: Atom & Cosmos
- Cowen, R. 2008. State of the Universe: Microwave glow powers cosmic insights. Science News 173 (March 15): 163. Available at [Go to]
- Cowen, R. 2008. Before the beginning: Theory suggests a pre-Big Bang universe. Science News Online (June 12). Available at [Go to]
- Kashlinsky, A., F. Atrio-Barandela, D. Kocevski and H. Ebeling. 2008. A measurement of large-scale peculiar velocities of clusters of galaxies: results and cosmological implications. Astrophysical Journal Letters, in press, October 20 (686). Available online at arXiv:0809.3734v1 or by following this [Go to]
- Kashlinsky, A., F. Atrio-Barandela, D. Kocevski and H. Ebeling. 2008. A measurement of large-scale peculiar velocities of clusters of galaxies: technical details. Astrophysical Journal, in press. Available online at arXiv:0809.3733v1 or by following this [Go to]
- Losing Louisiana
- Iron-ic twist deepens cosmic ray puzzle
- Stone Age flutes found in Germany
- Glass beads cluster as they flow
- Concerns over bisphenol A continue to grow

but what about the explanation based on the relativity of motion?
What if the observer moves with respect to the clusters in question
with a speed vector equal minus the vector of the apparent
collective motion of the clusters?
To challenge this explanation, one has to estimate
a possible magnitude of the observer velocity.
The article mentions each cluster moving at its own velocity
every which way. What is the average magnitude of those velocities
from each of which the collective velocity and the
"expanding-universe" velocity has to be subtracted
geometrically before averaging? (Not the geometric average,
which is, of course, 0, but average of magnitudes.)
That number would be the most "typical" velocity
of a galactic cluster and as such it would be an estimate
of the observer velocity because the observer moves
with a galactic cluster, the one our Milky Way belongs to.
If the number happens to be of the same order,
as the geometric average of the velocities which is discussed
in the article and which has the magnitude about 1000 km/sec, then
one should attribute the phenomenon to the relativity of motion.
Maybe what I mentioned above is not discussed
in the article because of this comment triviality.
But then why all this talk about beginning of the universe
etc.? If the galactic clusters typically move at such speeds
and nobody is concerned about the reason for them to do so,
why should be one for ours?
Rog Dog
Please login or register to participate.