Science Nugget—Exploring the Epoch of Galaxy Formation in "3-D"
  Chuck Steidel, TMT Science Advisory Committee Chair

One of the most exciting possibilities enabled by the light-gathering power of TMT is to achieve "tomographic" observations of the distribution of gas both inside and outside of galaxies in the young universe. At early times, most of the normal matter in the universe- mostly in the form of gas- was actually outside of galaxies, tracing what we believe is a "web-like" structure formed by the distribution of dark matter. Galaxies formed where the "cosmic web" is densest, where filamentary structures intersect.  Fresh hydrogen gas from the "intergalactic medium" (IGM) was being rapidly accreted by forming galaxies, providing fresh fuel for the rapid star formation taking place in the densest regions near the centers of galaxies. At the same time, the vast amounts of energy produced by massive star formation and the subsequent supernova explosions, as well as by accretion of material onto super-massive black holes developing at the centers of the galaxies, has a profound (but highly uncertain) effect on the evolution of individual galaxies. This "feedback" of energy into the young galactic systems almost certainly drove large quantities of heavy elements (formed in the most massive young stars and their supernovae) into the IGM, limited the efficiency with which stars could form, and somehow provided a natural thermostat which determined the maximum mass a single galaxy could attain. How these energetic processes worked to shape the universe of galaxies is perhaps the largest unsolved problem in understanding the formation of galaxies. We believe that the most rapid period of galaxy growth occurred at redshifts of z~2-4, when the universe was only 10-20% of its current age (10-12 billion years ago).

Very sensitive spectroscopy of galaxies during this era, using WFOS, IRMS, and IRMOS can be used to catch them in the act of both forming the bulk of their stars and injecting energy, gas, and heavy elements into the IGM. The simultaneous detailed study of both galaxies and the diffuse gas between them, during the most active period in the history of the universe, will be possible within the same survey with TMT.

One-dimensional "core samples" of the high redshift universe have been possible using very bright quasars and high resolution spectrometers on today's 8-10m telescopes, providing most of what we know about the detailed physical properties of the IGM at these epochs. Ideally, one would like to have many background objects for each region of the universe to be studied, so that the "core samples" can be combined into a three-dimensional map of hydrogen and heavy elements. The process of combining such core samples into a 3-D picture is essentially performing "tomography" on the young universe, allowing for a diagnosis of the physical state of all normal matter, its relationship to galaxies and quasars in the same cosmic volumes, and a prognosis for their future evolution, working toward the present day. Historically, the spectra of quasars have been the only available means to study the IGM at high redshifts, since they have been the only suitably bright background sources to allow for sensitive spectroscopic measurements. The problem is that bright quasars are extremely rare, and so information on the IGM is difficult to extend to three dimensions.

The situation is very different when the light-gathering power of TMT is brought to bear on the problem. At a given distance (or redshift), the number of rapidly star-forming galaxies increases extremely steeply with decreasing apparent brightness. At the sensitivity limit of TMT+WFOS, there will be ~2-3 background galaxies per square arcminute of sky (about 1000 for a region the size of the full moon) suitable for obtaining information on the IGM, some 50 times higher density than quasars of the same brightness, and several hundred times the density of probes accessible using existing telescopes. Of course, the same survey will map out the locations of not only the brighter galaxies used as IGM probes, but also much fainter ones for which the spectra will provide redshifts (distances) as well crude physical information. Thus, with TMT, it will be possible to simultaneously obtain a densely sampled map of the distribution of galaxies and the diffuse material between them, in 3-D, providing the most complete possible census of all normal matter, and its relationship to dark matter. For the first time, the empirical picture of the high redshift universe would be of as high fidelity as those that currently exist only inside simulations. In fact, because we will learn not only about galaxies, but also the material in between them, the picture of the distant universe will contain more information than any map ever made, including state-of-the-art surveys of the nearby universe.