Uncovering Dark Subhalos with Stellar Streams in LSST

Dark matter makes up most of the mass in the universe, yet its fundamental nature remains one of the biggest open questions in physics. This project searches for small, invisible clumps of dark matter orbiting the Milky Way. These clumps contain no stars, so they cannot be seen directly, but they reveal themselves when they pass close to thin streams of stars that wrap around our Galaxy, leaving behind gaps and other signatures of interaction. The team at the University of Washington will use new data from the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) to map these stellar streams in unprecedented detail, applying modern artificial intelligence techniques to identify the imprints of dark matter clumps within them. The results will test the leading theory of dark matter at the scales where its predictions are most distinctive and least tested. The project will also strengthen the regional astronomy community by hosting workshops that connect students across the Pacific Northwest, including at smaller and less-resourced colleges and universities, with the LSST expertise concentrated at the University of Washington. The cold dark matter (CDM) model predicts a large population of dark subhalos, with masses below the threshold of galaxy formation – precisely the regime where alternative dark matter models diverge most strongly from CDM. Dynamically cold stellar streams are the only known probes of this population in the Milky Way: subhalo encounters produce characteristic gaps, density variations, and off-stream features. This program will make the first population-level measurement of these perturbations with LSST. The team will first systematically characterize streams across the Southern sky, incorporating the observational selection function and survey systematics to distinguish real perturbations from variations in data quality. They will then infer the properties of the perturbing subhalos using simulation-based inference, training on forward-modeled streams that include realistic baryonic perturbers and time-evolving Galactic potentials, exploiting the full stream morphology rather than simplified analytic likelihoods. Finally, they will compare the recovered subhalo population against predictions of CDM and alternative models, constraining the mass and self-interaction properties of the dark matter particle. A public database of characterized LSST streams will support applications in dark matter physics, galaxy formation, and Galactic structure. Broader impacts center on building a regional research community: annual workshops at UW will provide Pacific Northwest students with LSST data analysis training and direct connections to Rubin expertise, while a near-peer mentoring program will sustain these ties year-round, broadening participation in LSST science. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.