work. The correct answers are quite likely to surprise us. Time-domain surveys of the sensitivity and scope envisioned in the coming decade will increase by orders of magnitude the number and character of stellar explosions that we can study, allowing us to connect variations in the host galaxies and progenitors to the energy and characteristics of the explosions. Supernovae are critical markers both for mapping out the cosmos and for understanding the formation of heavy elements that are found in all of us, and so these studies are essential for understanding our origins.
By surveying large areas of the sky repeatedly, once every few days, we anticipate the discovery of the wholly unanticipated. Endpoints of stellar evolution we have yet to imagine, and the behavior of ordinary stars outside our experience, could be discoveries that cause us to dramatically revise our cosmic understanding. Exotic objects and events never before anticipated may be revealed. The full realization of time-domain studies is one of the most promising discovery areas of the decade. Advanced gravitational wave detectors will open up a new window on the transient universe, including the last stages of binary neutron star and black hole mergers. Studying electromagnetic counterparts of gravity wave bursts will help illuminate the nature of the sources.
The powerful surveys described above will produce about a petabyte (1 million gigabytes) of data—roughly as much data as the total that astronomers have ever handled—every week. The data must be quickly sifted so that interesting phenomena can be identified rapidly for further study at other wavelengths. Interesting phenomena could also be discovered by cross-correlating surveys at different wavelengths. Vast numbers of images must be accurately calibrated and stored so that they can be easily accessed to look for motion or unusual behavior on all timescales. As daunting as it sounds, the technology and software that enable the accessing and searching of these enormous databases are improving all the time and will enable astronomers to search the sky systematically for rare and unexpected phenomena. This is a new window on the universe that is opening thanks to the computer revolution.
Another way in which computers will enable discovery in the coming decade is through increasingly sophisticated numerical simulations of the complex physical systems that are at the heart of much of astrophysics. The merging of two black holes, the growth of disks and the planets that form within them, the origin of large-scale structures that span the cosmos, and the formation of galaxies from the cosmic web are examples. Such simulations have great potential for discovery because they can illuminate the unanticipated behavior that can emerge from the interactions of matter and radiation based on the known physical laws. Through