The intergalactic medium is most of the volume of the Universe and until recently it contained most of the dark matter and gas. This project is a tight combination of precision measurements of the intergalactic medium through absorption in quasar spectra with a suite of simulations designed to extract the physical parameters from those measurements. It will provide new estimates of cosmologically interesting parameters, including the baryonic, dark matter and dark energy contents, and the amount of clustering of matter on small scales, which are more accurate than those obtained before from the intergalactic gas, and completely independent of those from supernovae, galaxies and the cosmic microwave background. For all parameters the errors will be comparable to the best of any method. The work will also give much more accurate measurements of the temperature of the intergalactic gas, and the intensity of the ultraviolet background radiation. Sophisticated calibrations of spectra will make them suitable for determining which set of parameters makes the simulated spectra most like the real spectra. Some simulations will include radiative transfer so as to interpret absorption near to individual quasars.
This team has a long record of integrating education and diversity with their research. Students receive intensive instruction in astrophysics, cosmology and the methods of precision measurement, observing with the Keck and Lick telescopes, and learn to run a large code on state-of-the-art supercomputers. Most undergraduates will also become co-authors on one or two papers.
The standard way to extract physical information from observations of the hydrogen in the intergalactic medium is to run supercomputer simulations that follow the gravitational clumping and ionization of intergalactic gas. We ran an exhaustive set of simulations for various universes allowed by other types of observation. We found that none of the simulations was able to able to match data from the intergalactic medium, which means that none of the thousands of simulations run since the mid 1990 will have matched data. We found that some simulations can match certain aspects of data, but none match all the data simultaneously. The amount of clumping of the hydrogen has a different dependency on length-scale in the simulations and observations. Moreover, the simulations give too little clumping for a given amount of hydrogen absorption. We also found that the mismatch between standard modern simulations and data is larger at earlier times, a strong clue that the difference is not caused by the ionization of helium, a late event. We explored various ways of changing simulations, including using different types of ionizing radiation, and different starts to the ionization, but these changes were not relevant. We artificially changed temperature values inside the simulation cubes and found that results were more like data when we made the temperature depend on a power of the gas density. This implies that the usual ultraviolet light from young galaxies and QSOs is not the main source for the ionization and temperature of the intergalactic gas. Our findings have broad implications for most results that derive from observations of intergalactic hydrogen. For example, they weaken the upper limits on the masses of the light neutrinos. We also examined the reliability of the data. We measured the properties of the hydrogen absorption in high resolution spectra of 90 QSOs, and we found and corrected two problems with prior measurements, however this did not make simulations and data match. The main conclusion is then that the difference between the simulations and data are not due to problems with the data. Instead, the simulations themselves are the problem. They do not adequately represent the gas in the intergalactic medium, and the most likely problem is with the heating and ionization of the gas. In a second related set of projects, we obtained moderate resolution spectra of pairs of QSOs that are close to each other in the sky so that the light of the background QSO passes near the foreground QSO. We then found absorption arising near the foreground QSO to study how this QSO changed the ionization of nearby gas. We found that strong Mg+ absorption, from lower ionization gas in galaxy halos, is preferentially beside and up to 12 million light-years behind QSOs, while absorption by C3+, Si3+ and N4+, all in higher ionization gas, is more symmetrically distributed. We can interpret these results by having QSOs radiate ten times more ionizing radiation towards us than in the opposite direction. Alternatively, QSOs might emit their current UV luminosities equally in all directions, but for only ~1 million years. This work was led by an undergraduate who went on to graduate school in Astronomy. In support of this work, a second undergraduate, also now in graduate school, found that Mg+ emission line peaks (readily observed at moderate redshift) are on average only 66±5 km/s larger in redshift than the peaks of the [OIII] lines (more accurate but hard to observe). This relatively small difference means that Mg+ gives an adequate measurement of the systemic redshift of a QSO, since random pair-wise velocities that we cannot correct are 100-400 km/s. Any project that requires accurate QSO redshifts can use this finding. We have also used spectra of pairs of QSOs to place limits on the amount of absorbing gas that is moving rapidly, for example in winds flowing out of galaxies. New spectra that we obtained of pairs of QSOs revealed 21 new coincident pairs of absorbers (previously only 2) with 120 - 1200 million-light years separation in the sky and typical redshift separations of only 30 km/s. The new data are sufficient to significantly reduce the upper limit that we placed on the fraction of the gas seen in absorption lines that arise in winds moving quickly out of galaxies. The remote observing room at UCSD that was used for this research was also used to introduce a few hundred non-science major students into the world of astronomical research. Of the five undergraduates who did research for this project, four went to graduate school in science and a fifth decided to work as a science teacher. The PI also worked with two artists who feature cosmic themes in their work.
