Understanding the Dark Universe with 3D Weak Gravitational Lensing

Although we can see a great amount of stars, galaxies and dust in the Universe, many independent observations now agree that the combined mass in these baryonic systems is insufficient to account for more than a few percent of the entire budget. The missing mass is usually explained via an additional but mysterious 'dark matter' component, which interacts with baryons only through gravity. In particular, since dark matter does not emit or even reflect light, it can not be seen by ordinary means. A second, even more mysterious component known as 'dark energy' has recently become apparent. This acts as a repulsive force on very large scales, and is currently accelerating the expansion of the universe away from the trajectory started by the Big Bang. Their existence challenges the standard model of particle physics, and can test General Relativity. As evidenced by the support from governmental agencies and the popular press in the 'Scientific Justification' submitted with this proposal, unravelling these twin puzzles is widely viewed as the outstanding problem in physics. This is also the focus of my proposal. DARK MATTER: Since we can not see dark matter directly, we must measure its indirect effect on the observable universe. Light is deflected or 'gravitationally lensed' by gravitational potentials. This distorts the apparent shapes of distant galaxies as their light passes near mass along our line of sight. Near dense mass concentrations, such as clusters of galaxies, strong lensing distorts individual background galaxies into prominent arcs and Einstein rings. Along most lines of sight, the much weaker distortion cannot be detected from one galaxy image, because the original shape of that galaxy is unknown. However, galaxy morphologies are uncorrelated and, in the absence of lensing, should have no preferred orientation. Using this fact, we can recover the weak lensing signal by averaging the shapes of adjacent galaxies that have been lensed by the same intervening mass. From the degree of correlation between their shapes, it is thus possible to infer the total mass distribution, iincluding dark matter. DARK ENERGY: A gravitational lens system operates like an optical lens in many ways. For example, the lens (mass) is most effective when placed half-way between the source and the observer. The efficiency of a lens at a fixed location depends upon the distance to the source. By observing the distortion in (and redshift of) many galaxies behind one cluster, we can trace the redshift-distance relation. This describes the overall expansion of the universe, for which each candidate for dark energy predicts a different characteristic signal. This simultaneous, dual test of both the growth of structure and of large-scale geometry is unique to gravitational lensing. Unfortunately, the galaxy shape distortions induced by gravitational lensing are typically an order of magnitude smaller than spurious distortions added by imperfections in telescope optics and during transit through the Earth's atmosphere. To increase the precision of weak lensing measurements, these potential biases must be more finely controlled. The onus now placed upon weak gravitational lensing from the wider community is to verify such control, and to reliably ascertain the absolute calibration of the measured signal. The latter is particularly crucial because it can not be internally constrained from observations using current techniques. I have carried out major lensing surveys from the William Herschel, Keck, Subaru and Hubble Space telescopes, attempting to understand their varying instrumental systematics. I have also used this experience to develop advanced techniques for image analysis and for signal interpretation. My goal is to fully exploit the rich potential of large, dedicated weak lensing surveys that have been approved and will be available within five years. However, this will require further, intense development of techniques in advance.