Scientists know very little about the mysterious force dubbed “Dark Energy” that seems to be quickening the expansion of the universe but, according to LSU Physics and Astronomy Associate Professor Bradley Schaefer, one thing appears to be true: Dark Energy is not constant; it varies over time, a concept that challenges Einstein’s idea of an unchanging force called a “Cosmological Constant.”

Schaefer discussed this finding Jan. 11 at the 207th meeting of the American Astronomical Society in Washington, D.C.

His analysis is based on 52 distant explosions called gamma-ray bursts, which he says can serve as “mileage markers” through the universe to measure how quickly it has expanded.  Previously, astronomers have used supernovae as standard candles to measure the rate of expansion. Gamma-ray bursts originate from galaxies far more distant than visible supernovae and can serve as probes to the very distant, early universe.

“The most distant gamma-ray bursts appear brighter than they should if Dark Energy does not change with time,” Schaefer said. “It appears that the expansion in the early universe was slowing down more than Einstein would have predicted.”

Not only is the universe expanding, but the rate of expansion is currently accelerating instead of slowing down. This was an unexpected, landmark discovery in 1999.  Independent teams made this discovery by observing certain kinds of star explosions called Type Ia supernovae. These explode and shine with a known energy, like a light bulb of a specific wattage, and are used to measure distances in the universe.  The farther they are, the dimmer they will appear.

The longer variety of gamma-ray bursts, usually lasting about 10 seconds or more, arise from the explosions of very massive stars and are about a thousand times brighter than Type Ia supernovae.  With such high luminosity, bursts are visible across most of the universe and can serve as markers for cosmology.  Most of their light is given off in gamma rays.   The bursts are detected by satellites, such as NASA’s Swift and High Energy Transfer Explorer, or HETE.

A valuable tool to measure the universe is the Hubble Diagram, named after Edwin Hubble, who discovered the expansion of the universe in the 1920s. This diagram is a graph that compares an object’s distance with its “redshift,” or the extent that its light is shifted to lower energies by the expansion of the universe.  The greater the distance, the greater the redshift, as the light is pulled back by space itself on its long journey towards us.  The shape of the plot in a Hubble Diagram reflects the expansion history of the universe and hence of the structure of space-time itself.

Astronomers in 1999 observing supernovae up to redshift 1, or about 7.7 billion light years away, found that the supernovae were dimmer than expected, which meant that the universe was expanding faster than expected, taking us farther away from the original light source.

Schaefer’s Hubble Diagram contains 52 gamma-ray bursts, and 21 have a redshift larger than the most distant Type Ia supernovae known — that is, greater than redshift 1.7, or a light travel time of 9.8 billion years.   The most distant burst is at redshift 6.29, or 12.8 billion light years away.

While gamma-ray bursts are brighter and more distant than Type Ia supernovae, their intrinsic energy varies greatly. Yet, the wattage of each burst can be determined with fair accuracy by measuring any of five specific properties. For gamma-ray bursts, these observed quantities correlate with burst luminosity. There is the spectral lag (the delay between the “bluer” and “redder” light); the variability (the “spikiness” of the brightness changes); the brightest photon energy (analogous to color); the time of the “jet break” (when the afterglow light starts to fade fast); and the minimum rise time (how fast the brightness turns on).

“In a similar task, people are always judging the distance to lights by seeing how bright they are,” said Schaefer. “To do this, we have to know the wattage of the light, and this can be deduced from various observed properties. For example, a driver at night seeing a pair of similar lights appearing on the road ahead will deduce that they are headlights of an oncoming car, and with a known wattage the driver will automatically estimate the distance to the car. Or, last month, if we saw a blinking red light on a tree through the window of some house, we could readily get an idea of the distance to the Christmas tree by seeing how bright the light appears.”

In the case of gamma-ray bursts, any of the five properties give independent information on the wattage of the burst and then an estimate of the distance. Some of the bursts have only one measured property while others have all five measured.  In all, Schaefer has used a total of 172 distance measures from 52 gamma-ray bursts.

When the burst distances and redshifts are plotted onto a Hubble Diagram, the shape of the curve reveals the expansion history of the universe up to 12.8 billion years ago. Schaefer’s analysis shows that the Hubble Diagram is significantly bent down at large distances, compared to the prediction based on the Cosmological Constant.  This argues that the universe is not ruled by the Cosmological Constant.  The Dark Energy is changing with time across the age of the universe.  In the young universe, the expansion was decelerating more than expected.

Gamma-ray bursts have been used for Hubble Diagrams since 2003.

The new result improves by using 10 times more distance measures than those used in previous work. These distance measures are from the various independent luminosity indicators for each burst.

The rejection of the Cosmological Constant is at the 97 percent confidence level, which is to say that there is a small chance — three percent — that normal fluctuations in the data can result in an invalid rejection.   In addition, as part of the inevitable and good course of science, this result will be subjected to intense scrutiny for improvements and complications.  As always, a final rejection of the Cosmological Constant must await the duplication of the result by independent methods.

Schaefer added that his analysis represents countless hours by hundreds of workers who build and operate various burst detectors on satellites, as well as the many observers at telescopes around the world.  Roughly one-third of the bursts were discovered with the Swift satellite, one-third with the HETE satellite and the remainder with four other satellites.

“Over the next two years, Swift should discover another 50 bursts that can be placed on the Hubble Diagram,” said Neil Gehrels, Swift Principal Investigator at Goddard Space Flight Center in Greenbelt, Md.  “This will double the number in the current data set and provide a test of Schaefer’s fascinating result.”

Figures, charts and additional information are available at

During the AAS meeting Jan. 8-12, Schaefer can be reached by leaving a message with the AAS Press Office at 202-745-2108. Otherwise, Schaefer can be reached at 225-578-0015.