LEAPS-MPS: Unraveling the Galaxy-Cosmic Web Connection using Monte Carlo Physarum Machine

New Mexico State University, Las Cruces NM

Investigators

Abstract

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2). All of the galaxies in the Universe are embedded in a cosmic web of filaments of material. One of the open questions in galactic astrophysics is how the structure and mass distribution of galaxies might depend on the structure of the filaments in which they are embedded. This work will explore how the cosmic web affects the structure and evolution of galaxies, and in turn how galaxy evolution and the expelling of material might affect the cosmic web. The program will be supported at a Hispanic Serving Institution, and will draw undergraduate researchers from under-represented groups. The team will also initiate and after-school outreach effort at a local high school where high school students will have the opportunity to work on actual science data from the project, and interface with project personnel. The team will deploy a simulation routine called the Monte Carlo Physarum Machine (MCPM), which incorporates filamentary structures while providing environmental density estimates even at the very low densities of the intergalactic medium. The MCPM is a computational model designed to reconstruct physically plausible density fields over sparse discrete data – in this case, these are primarily observed galaxies (positions and masses), and alternatively simulated baryonic/dark matter halos. This model is underlined by the optimal transport theory, obtaining network-like structures that interconnect data according to the principle of least action. Given that exact solutions to optimal transport problems are typically NP-hard, MCPM relies on a biologically inspired solution: emulating three-dimensional growth patterns of Physarum polycephalum ‘slime mold’, which is known to provide good approximations to this class of problems. From a computational perspective, MCPM is a stochastic model based on a combination of a discrete and a continuous component. The discrete component is an ensemble of particle-like agents that freely explore the simulation domain. The continuous component is a pair of 3D scalar lattices: one that represents the concentration of an environmental marker that guides the agents through the input data, and one that accumulates the spatio-temporal density of the agents and represents the actual solution. 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.