Friday, August 31, 2012


The study of extrasolar planets has been receiving a lot of attention recently, and the reasons are clear. Astronomers are searching for planets like Earth and attempting to identify where life may have formed, making this one of the most exciting fields in astrophysics. The results coming from NASA's Kepler mission have been especially interesting.

But, there are bigger mysteries being addressed in astrophysics, concerning the origin and evolution of the entire universe. What is the universe made of? How did it begin? How old is it? How will it change in the future? These are easy questions to ask, but very difficult questions to answer. This post will explain some of the practical challenges that have already been overcome in trying to answer these cosmology questions. A later post will look at future challenges.

A key question is "what is the universe made of?", since the answer affects our understanding of how the universe will evolve. Below is a pie chart showing our current knowledge of the mass and energy content of the universe. The small, red piece of the pie corresponds to normal matter found in our bodies and the planets and stars, and is the only piece that is well understood. The blue piece corresponds to dark matter, first suggested by Fritz Zwicky back in the 1930s. Indirect evidence for this invisible matter has been inferred from its gravitational effects on visible matter, but the particle or particles making up dark matter have yet to be identified in experiments. Finally, the green piece of pie corresponds to dark energy, the biggest, freshest and most mysterious piece of the cosmic pie. It was invented to explain the acceleration in the expansion of the universe, discovered back in the late 1990s.

The mass & energy content of the universe. Credit: CXC/M.Weiss 

This pie chart is a remarkable figure for a couple of reasons. First, we didn't even know that the expansion of the universe was accelerating until 14 years ago, which means that we didn't know dark energy existed. Second, enormous observational challenges have already been overcome to enable this figure to be made, including astoundingly accurate measurements of the brightness of exploded stars - supernovas - many of them located billions of light years away. Years of patient supernova observations and careful calibration were required to make these measurements, involving dozens of astronomers around the world. It was worth it, because the discovery of cosmic acceleration is one of the biggest results in science, and resulted in the 2011 Nobel Prize for Physics being awarded to three astrophysicists, representing two different teams.

Even more challenging than identifying the need for dark energy is to understand what it is. The simplest explanation is that it is an energy associated with empty space - known as vacuum energy - that does not change over space and time. Another possibility is that it is a type of energy called "quintessence" that varies with space and time. A third possibility is that dark energy is not needed at all, and that cosmic acceleration can instead be explained by modifying the theory of gravity, Einstein's General Theory of Relativity, over very large distance scales.

Astronomers are looking for changes in the properties of dark energy over time, to decide between vacuum energy and quintessence. In the case of modified gravity, they are studying the way that galaxy clusters grow to see if this differs from predictions using General Relativity. The observational challenges in deciding between these possibilities are huge, because the observed effects can be very subtle and the objects being studied are often extremely faint. This means that large samples are needed to give interesting results with reasonable statistics. Then there are the dreaded systematic errors, where measured values differ from the true values, sometimes in unexpected ways.

To overcome these problems many different approaches are needed, as explained in the Dark Energy Task Force (DETF) study, commissioned to advise NASA, NSF and the Department of Energy on directions for future dark energy research. Supernova observations are continuing to be used to measure distances and study accelerated expansion. Another technique uses "Baryonic Acoustic Oscillations (BAO)" to trace ripples in the positions of galaxies left by sound waves in the early universe. The size of the ripple pattern can be used to measure distances. A convincing detection of this ripple pattern was first reported in 2005 by Daniel Eisenstein and collaborators. Eisenstein is now at the Harvard-Smithsonian Center for Astrophysics (CfA), where the Chandra X-ray Center is also located (as well as my office in their Education and Public Outreach group). Good improvements have been made with newer projects such as the Baryon Oscillation Spectroscopic Survey, or BOSS. This is a big project, targeting over a million galaxies, with observations extending over 5 years and expected to finish in the spring of 2014. That is what it takes to detect the weak ripple pattern with confidence and use it to study dark energy.

Baryonic Acoustic Oscillations. Credit: BOSS/C.Blake & S.Moorfield

The DETF emphasized the need for techniques that estimate the growth of large structures, such as galaxy clusters, over time. This is a completely independent way to study dark energy and it offers the ability to search for a possible breakdown of General Relativity. One success story has been the use of X-ray observations with NASA's Chandra X-ray Observatory to estimate the decrease in the growth rate of galaxy clusters caused by cosmic acceleration, over the last 5 billion years, in research led by Alexey Vikhlinin, also from CfA. The stifling of the growth of galaxy clusters over time shows excellent agreement with predictions, when it is assumed that vacuum energy is causing cosmic acceleration. No evidence for deviations from General Relativity were seen.  As an aside, because this involved Chandra observations I worked on the publicity effort.

Optical and X-ray image of a galaxy cluster and simulations showing growth of cosmic structure.
A second team, led by Steve Allen from Stanford University, has also used Chandra observations to probe dark energy and test General Relativity. Again, Einstein's theory was found to pass the test.

As with the supernova work, years of work were required before these authors were able to do real cosmology with their observations. Careful attention has been given to simulations, calibration and statistical analysis, including the study of possible systematic errors.

Another technique is "weak gravitational lensing", which measures the distortions of galaxy images by intervening matter. By probing dark matter at different distances, astronomers have another way to estimate the decrease in the growth rate of galaxy clusters with time. The distortions in the shapes of galaxies are subtle, requiring careful statistical analysis and large observing programs. Some promising results have already been reported, including ones using Hubble Space Telescope observations.

So far all of the evidence found by astronomers agrees with the simplest possible explanation for dark energy, namely vacuum energy. However, the other possibilities - quintessence and modifications to gravity - remain in contention, and there may be other possibilities we are not even aware of. What are the prospects for getting a definitive understanding of the nature of dark energy? I will consider that question in a future blog post. As a spoiler, the challenges far exceed the ones that have already been overcome.