A high redshift galaxy identified in the William Herschel Deep field. Top: the same region of sky, but now a 'true' colour image taken at optical wavelengths. The cross indicates the same position as on the infra-red picture -- the galaxy has completely dissappeared. Bottom: an infra-red 'true' colour image showing a bright galaxy indicated by the cross. Photo: U. of Durham

Astronomers at Durham University (UK) have found new evidence that large numbers of galaxies were in existence at even earlier times than previously thought. They have identified many galaxies with redshifts between 4 and 6. This means they were already in existence about 10 billion years ago, when the universe was six times smaller than it is now. With this discovery, researchers may have to rethink their ideas about how galaxies formed. The work will be presented today by Dr Tom Shanks at a symposium during the International Astronomical Union General Assembly in Manchester (UK).

Because of the finite speed of light, pictures of very remote galaxies (identifiable because their redshifts have high values) record them as they appeared many billions of years ago, or even in the process of formation. Observations of this kind directly test cosmological theories about how and when galaxies formed. The most popular theories so far have said that galaxies formed relatively recently and predict that hardly any galaxies should be seen with high redshifts.

Now, new red and infra-red pictures have been taken of small areas of the sky already targetted by the deepest ultraviolet and blue surveys. The new images extend the search for galaxies to higher redshifts than every before. Surprisingly, the new results show such large numbers of galaxies that there seems to be almost as many bright galaxies with redshifts of 5 as there are at low redshifts nearby! This makes the epoch of galaxy formation earlier in the history of the universe than astronomers previously thought.

The ground-based pictures come from the UK/Dutch/Spanish 4.2-metre William Herschel Telescope in the Canary Islands and from the 3.5-metre Calar Alto Telescope in Spain. Tens of hours of exposure time went into a picture taken in red light at the Herschel Telescope with another similarly long exposure made in infrared light at Calar Alto. These pictures have been compared to new pictures in infrared light taken with the Hubble Space Telescope in its Hubble Deep Field North and to new optical and infrared pictures in the Hubble Deep Field South. The Space Telescope exposure was a total of 120 hours in a single tiny patch of sky, observing in ultraviolet, blue and red light. The Space Telescope pictures reach deeper (i.e. can see fainter objects) than the ground-based pictures but they cover a smaller area of sky. However, the basic result is that the counts of high redshift galaxies from both the ground- and space-based experiments agree well, in the range where they can be compared and so both these experiments appear to be giving consistent results.

A false-color view of the William Herschel Deep field through a single infra-red filter (1.6 microns) using the 3.5 Calar Alto telescope in Southern Spain. This exposure took 14 hours to complete and opens up a new, wide-angle view of the Universe at infra-red wavelengths. Photo: U. of Durham

In 1996, the ultraviolet and blue pictures in the Herschel and Hubble Deep Fields revealed so many faint blue galaxies at a redshift of 2 that they already challenged the claims of the most popular cosmological theory, which suggested that galaxies formed around a redshift of 1, when the universe was half as big as it is now. Since then, observations by Charles Steidel and collaborators at the 10-metre Keck telescope confirmed the Durham group's results by finding many galaxies at redshifts of 3 and 4. Now, applying similar techniques as before but to the new red and infra-red pictures, Dr Shanks and colleagues find large numbers of galaxies at the even higher redshifts of 5 to 6. There are as many galaxies at these high redshifts as are found locally.

Dr Shanks says, "Four years ago, we described the galaxies we found at with redshifts of 2 as being at 'The Final Frontier' because we thought that just beyond them we might be looking back to a time before galaxies formed. Now that large numbers of galaxies at even higher redshifts have been found, we feel entitled to describe them as being Beyond the Final Frontier!".

Deep pictures as time machines
Light travels at a speed of 300,000 kilometres per second. Although this speed is high it is finite and this means that light takes about 8 minutes to reach us from the Sun. The huge size of our Milky Way Galaxy means that light takes about 25 000 years to reach us from the Centre of our Galaxy. But the light from the faint galaxies seen in these deep pictures has come from even farther away, so far away that the light has taken about 10 billion years to reach us!! This is close to the estimated age of the Universe and hence the deep pictures are probing not only out in distance but they probe back in time as well, since we can see galaxies as they appeared early in the history of the Universe. In this sense the deep pictures can be viewed as time machines allowing us to see deep into our past. The allow cosmologists directly to observe galaxies in the process of formation. Their theories can then be tested in detail.

The ability to detect galaxies at redshifts as high as z=6 comes from the new red and near-IR Hubble and Herschel deep pictures. Essentially they allow much more accurate measurements of the colours of the faintest galaxies, and the quality of these measurements is so high that the galaxy colours become a very good substitute for galaxy spectra and this allows new estimates of the galaxy redshifts to be obtained.

Could we detect galaxies at even higher redshifts?

The detection of bright galaxies at z=4-6 opens up the question as to whether galaxies at even higher redshift may exist. Though it would make even more problems for theorists, the timespan between z=10 or z=20 and z=5 is incredibly short relative to the timespan between z=5 and z=0. To observe galaxies at these redshifts we need even deeper pictures over a relatively wide field, particularly at infra-red wavebands. The new Wide Field Camera at the UK Infrared Telescope on Hawaii and the UK VISTA telescope in Chile which will be available in a few years will open up these new redshift regimes for observation. A few years later the giant NASA-ESA 6.5-metre Next Generation Space Telescope (NGST) in which the UK has a share will be launched to observe even deeper and further into the infrared with the prime aim of detecting at the highest redshifts the dawn of the age of the galaxies.