Future cosmological surveys will map the spatial distribution and shapes of hundreds of millions of galaxies, allowing us to constrain fundamental physics, such as the nature of cosmic acceleration, tests of general relativity, neutrino masses, and primordial non-Gaussianity. Extracting information from these data sets faces two key challenges: The quality and size of these data sets require new analysis techniques. Second, the interpretation is limited by systematic uncertainties caused by the observing process, as well as from insufficient understanding of astrophysics.
In this talk I will introduce concepts for the joint analysis of multiple observables tracing the large-scale structure of the Universe. These observables probe different aspects of structure formation and are affected differently by systematic uncertainties. A joint analysis of these observables improves constraints on cosmological parameters significantly. I will describe science prospects of these multi-probe analyses in the context of the Large Synoptic Survey Telescope (LSST), the largest cosmological experiment of the coming decade, and I will give an outlook on synergies between LSST and other ground and space-based surveys.