The explosions were fainter than the teams expected
In 1998, astronomers studying distant Type Ia supernovae found that the explosions appeared fainter than expected in simple models of a slowing universe. After correcting their brightnesses and comparing them with nearby examples, the High-Z Supernova Search Team inferred larger distances. A second group, the Supernova Cosmology Project, reached the same broad conclusion with a different sample. The two teams had found evidence that the expansion of the universe had accelerated.
The discovery began with a practical aim: use bright stellar explosions to measure how cosmic expansion changes with time. Gravity from ordinary matter and dark matter was expected to slow that expansion. Distant supernovae offered a way to look into the past and compare the expected slowing with the distance scale implied by their light. The surprise emerged from the relation between measured brightness and redshift, a relation that an expansion history predicts.
A useful cosmic lamp needs a correction
A Type Ia supernova is associated with the thermonuclear disruption of a white dwarf, a dense stellar remnant. These explosions can be recognized from their spectra and their changing light over time. They are bright enough to be seen far beyond nearby galaxies. Yet they do not all have exactly the same peak luminosity. Using every Type Ia event as an identical lamp would throw away important information.
Nearby observations showed a useful pattern: brighter normal Type Ia supernovae tend to fade more slowly. Astronomers measure the light curve, the rise and fall in brightness, and use its shape to standardize the peak. Colour information helps estimate dust extinction and other variations. A standardized candle is therefore a corrected distance indicator. Its usefulness depends on how reliably these corrections transfer from nearby explosions to distant ones.
The observing task is demanding. The team needs to find an event early, measure it repeatedly in several bands, identify its type and subtract the host galaxy’s light. At larger redshift, a chosen observing band also corresponds to a different wavelength at the source. Those effects enter the comparison. The original studies devoted substantial work to calibration, light-curve corrections, selection and possible systematic shifts.
Brightness and redshift supply different information
An object of known luminosity appears fainter at a greater distance. In an expanding universe, the brightness relation also includes the stretching of photon wavelengths and the slowing of their arrival rate. Astronomers gather these effects into a luminosity distance, the distance that gives the observed flux for the adopted luminosity. A larger luminosity distance means a fainter standardized supernova.
Redshift measures how much the light’s wavelength has stretched since emission. It tells us about the change in cosmic scale between then and now. The distance reached at a given redshift depends on the expansion throughout the intervening history. Astronomers therefore compare whole distance-versus-redshift curves. A single distant point supplies no direct stopwatch reading of cosmic acceleration. A population of points can favour one expansion history over another.
Two samples pointed away from simple slowing models
Riess, Schmidt and their colleagues reported ten newly observed distant Type Ia supernovae in their 1998 paper. Combined with earlier work, they compared sixteen high-redshift events with thirty-four nearby ones. Their distant objects were, on average, about ten to fifteen per cent farther away than expected in the low-density, zero-cosmological-constant model used for that comparison. Perlmutter and the Supernova Cosmology Project’s 1999 paper analysed forty-two distant events together with a nearby comparison sample.
Dust could make an object fainter, and a population of intrinsically different explosions could shift a distance scale. The teams examined colours, light-curve behaviour, selection effects and other possible biases. Independent samples and analysis methods strengthened the result. Later work expanded the samples and treated shared calibration errors more carefully. Such an error can move many supernova distances together, so it cannot be reduced as if every point were an independent new measurement.
Within the homogeneous expanding-universe models being tested, the data favoured an accelerating phase. A positive cosmological constant supplied one explanation. In general relativity it contributes a smooth energy component with a gravitational effect capable of accelerating the expansion. “Dark energy” names the wider class of proposed causes. The distance observations constrain the expansion history; identifying the underlying physical cause requires additional measurements and theory.
The prize followed a change in the question
The 2011 Nobel Prize recognized Saul Perlmutter, Brian Schmidt and Adam Riess for the discovery through distant supernova observations. The result changed cosmology’s practical agenda. Astronomers now had to determine how the inferred accelerating component behaved over time and compare the supernova distances with evidence from other cosmic measurements. A modest shift in the brightness of distant explosions had opened a much larger question about what controls the expansion of space.