Olbers' paradox, described by the German astronomer Heinrich Wilhelm Olbers in 1823 (but not published until 1826 by Bode) and earlier by Johannes Kepler in 1610 and Halley and Cheseaux in the 18th century, is the argument that the darkness of the night sky conflicts with the supposition of an infinite and eternal static universe. It is one of the pieces of evidence for a non-static universe such as the current Big Bang model. This "paradox" is sometimes also known as the "dark night sky paradox" (see physical paradox).
If the universe is assumed to contain an infinite number of uniformly distributed luminous stars, then:
Looking at trees within a big flat wood in the direction of the horizon shows the effect: The mass of dark trees will hide the horizon (imagine the trees now as lights).
The further away one looks, the older the image viewed by the observer. For stars to appear uniformly distributed in space, the light from the stars must have been emitted from places where the stellar density of the region at the time of emission was the same as the current local stellar density. A simple interpretation of Olbers' paradox assumes that there were no dramatic changes in the homogeneous distribution of stars in that time. This implies that if the universe is infinitely old and infinitely large, the flux received by stars would be infinite.
Kepler saw this as an argument for a finite observable universe, or at least for a finite number of stars. In general relativity theory, it is still possible for the paradox to hold in a finite universe: though the sky would not be infinitely bright, every point in the sky would still be like the surface of a star.
A more precise way to look at this is to place Earth in the centre of a "sphere". If the universe were homogeneous and infinite, then at a distance r away from the earth, the shell of the sphere would have a certain flux (viewed from Earth) due to the individual flux of the stars on the shell (brightness) and also the number of stars in the shell (cumulative flux). When an observer from Earth looks to a farther distance to another shell, r+x, the number of stars increases by the square of the distance, while the flux decreases by the inverse squared. Comparing the total brightness of the first shell to the second shell, one notices that both shells have equal flux, since the flux of each individual star decreases due to distance but is equally made up for by the number of stars. This means that no matter how far away an observer on Earth views the sky, the brightness of each consecutive shell would not diminish; rather, they would be equal. If the universe were infinite (age and volume) and had a regular distribution of stars, then there will be an infinite number of such shells and infinite amount of time for the light to reach Earth (infinite flux) as long as the earth remains, effectively meaning that there would never be night on Earth.
The universe, according to the mainstream theory of the universe, called the Big Bang Theory, is only finitely old; stars have existed only for part of that time. So, as Poe suggested, the earth receives no starlight from beyond a certain distance.
According to the Big Bang Theory, the sky was much brighter in the past, especially in the first few seconds of the universe. All points of the local sky at that era were therefore brighter than the circle of the sun, despite the finite and even more limited range that light could travel in that prehistoric era; this implies that most light rays will terminate not in a star but in the relic of the Big Bang.
The redshift and expanding space hypothesised in the Big Bang model would by itself explain the darkness of the night sky, even if the universe were infinitely old. The steady state cosmological model assumes that the universe is indeed infinitely old and uniform in time as well as space. It is also expanding exponentially, producing a redshift. There is no Big Bang in this model, but there are stars and quasars at arbitrarily great distances. The light from these distant stars and quasars will be redshifted accordingly, so that the total light flux from the sky remains finite and dominated by the nearest light sources. However, the steady state model cannot explain the detailed behavior of distant starlight and the microwave background, since it requires a continuous transformation of the former into the latter at decreasing frequencies; this transformation is not observed.
However, this reasoning alone would not resolve the paradox given the following argument: According to the second law of thermodynamics, there can be no material hotter than its surroundings that does not give off radiation and at the same time be uniformly distributed through space. Energy must be conserved, per the first law of thermodynamics. Therefore, the intermediate matter would heat up and soon reradiate the energy (possibly at different wavelengths). This would again result in intense uniform radiation as bright as the collective of stars themselves, which is not observed.
Mathematically, the light received from stars as a function of distance from stars in a hypothetical fractal cosmos can be described via the following function of integration:
= the minimum distance from which light is received ≠ 0
= the variable of distance
= average luminosity per star at
= number of stars at
The function of luminosity from a given distance determines whether the light received is finite or infinite. For any luminosity from a given distance proportional to , is infinite for but finite for . So if is proportional to , then for to be finite, must be proportional to , where . For , the numbers of stars at a given radius is proportional to that radius. When integrated over the radius, this implies that for , the total number of stars is proportional to .
Mainstream cosmologists reject this type of fractal cosmology on the grounds that studies of large-scale structure in combination with the timeline of the universe have not produced conclusive evidence for it.