Clearing the Cosmic Fog
AIn its early history, the universe was filled with a thin fog of neutral hydrogen gas. This fog absorbed ultraviolet light so effectively that much of the radiation produced by the first stars and galaxies could not travel far. Then, roughly 13 billion years ago, the situation began to change. Energetic radiation gradually stripped electrons from the hydrogen atoms, a process known as ionisation, and the fog lifted, allowing light to move freely across space. Astronomers refer to this transformation as reionisation, and it ranks among the most important events in cosmic history. Yet, as Cody Carr, an astronomer involved in a new study, admits, scientists still do not know exactly which objects caused it or how it evolved.
BFor the fog to clear, two conditions had to be met. First, something had to produce enormous quantities of ionising radiation, and young galaxies full of hot, massive stars are widely regarded as the leading candidates. Second, that radiation had to escape from the galaxies in which it was produced. This second step is far from simple, because galaxies are themselves surrounded by gas and dust that can absorb the radiation before it ever reaches intergalactic space. Understanding how ionising light gets out of galaxies is therefore central to explaining how reionisation took place.
CFor many years, the favoured explanation has centred on supernovae, the violent explosions that end the lives of massive stars. According to this view, the combined energy of many supernovae drives powerful winds of gas outwards through a galaxy. Supernovae certainly seem powerful enough for such a role: a single explosion can briefly outshine an entire galaxy. These winds were thought to clear channels through the surrounding material, rather like holes torn in a curtain, through which radiation could then stream into space. The idea is so widely accepted that, as the astrophysicist Renyue Cen notes, supernovae play a central role in the computer simulations that cosmologists use to model how radiation escapes from galaxies.
DTesting this idea directly is difficult, because the galaxies of the reionisation era are extremely distant and faint. The new study, published in the journal Nature Astronomy, therefore took an indirect approach. A team of nearly 30 scientists from more than a dozen institutions, including researchers at the University of Michigan, used the Hubble Space Telescope to examine nearby compact starburst galaxies, which are forming stars at an exceptional rate. Such galaxies are considered modern stand-ins for the young galaxies of the early universe. Because they are much closer, they can be studied in far greater detail than their ancient counterparts. By analysing dips in the galaxies' spectra caused by absorption, the researchers were able to measure how fast their winds were moving. They then compared these speeds with the proportion of ionising radiation escaping from each galaxy.
EIf the supernova model were correct, the galaxies with the fastest winds should have been the most efficient at releasing radiation. The researchers found exactly the opposite. The galaxies with the fastest winds allowed the smallest fraction of their ionising radiation to escape. One possible explanation is that, instead of clearing a path, the energy from supernovae stirs up the surrounding gas and dust, spreading neutral material more widely so that it blocks more of the radiation. If so, the radiation that does escape probably leaves earlier, released by young, massive stars before supernovae have had time to generate large winds. In Carr's words, the observations show that the ionising photons are escaping before galactic winds form.
FThe findings could have significant consequences for cosmology. Simulations that rely heavily on supernova-driven winds to release radiation may need to be revised. The authors suggest that more attention should be paid to the period before the first supernovae explode, and to the effects of smaller winds and other activity associated with young stars. In other words, the key stage for releasing radiation may come earlier in a galaxy's life than has generally been assumed. At the same time, they are careful to stress that their results do not yet settle the question of what caused reionisation. They are now seeking more observing time on Hubble to find out whether the same pattern appears in other galaxies.
GIn my view, this caution is entirely justified. Nearby starbursts may resemble early galaxies in some respects, but they exist in a universe that is billions of years older, and there is no guarantee that they behave in exactly the same way. Moreover, a pattern observed in one group of galaxies needs to be confirmed in a much larger sample before it can overturn a well-established model. Even so, the result is valuable precisely because it contradicts expectations. An explanation that has become a standard ingredient of computer simulations can easily be accepted without sufficient testing, and observations that challenge it force researchers to re-examine their assumptions. Should the supernova model turn out to be wrong, the story of how the universe became transparent will need to be rewritten, with the youngest stars, rather than the dying ones, playing the leading role. Whatever the final answer, the study is a reminder that even the most familiar explanations in science remain open to question.