The LSST Corporation Data Science Fellowship Program

Northwestern University seeks funding over three years from the Gordon and Betty Moore Foundation to expand a transformational pilot program that will supplement graduate education in astronomy with in-depth training in the necessary skills for doing science with big data: the LSST Corporation Data Science Fellowship Program (LSSTC DSFP). Previous support from the Moore Science Program has increased the capacity of the program establishing it as a preeminent training initiative for data science and infrastructure. The Vera C. Rubin Observatory, which will conduct the Legacy Survey of Space and Time (LSST), is currently the NSF’s flagship telescope project. The Rubin observatory is under construction in Chile, with operations set to begin in 2023. By imaging the entire visible night sky to incredible depth every few nights for a decade, LSST will produce a massive, unprecedented dataset. LSST achieved the highest priority in the 2010 Astronomy Decadal Survey because of its potential to provide insight into a wide variety of outstanding science questions: the population of the Solar System, the structure and evolution of our Milky Way Galaxy, the nature of dark energy and dark matter, and the transient, time-variable universe. LSST’s data will be transformative for the field of astronomy, not only because of its size and complexity, but because these data demand new working practices, and computationally sophisticated methods of data mining and analysis. Unfortunately, these practices and methods are not widely used by astronomers today; currently, most astronomers must teach themselves how to develop software, and with few exceptions, lack the skill set necessary to make efficient use of LSST’s coming data deluge. We propose a tranformative, already-tested, effective program that supplements graduate education in astronomy with in-depth training in the necessary skills for doing science with big data: the LSST Corporation Data Science Fellowship Program (LSSTC DSFP). The DSFP is a two-year training program (consisting of three, one-week sessions per year) designed to teach skills required for LSST science that are not addressed by current astrophysics programs. Each session provides a deep dive on a singular topic necessary for navigating the LSST datascape: