Introducing the DREAMS Project: DaRk mattEr and Astrophysics with Machine learning and Simulations (2405.00766v1)

Abstract: We introduce the DREAMS project, an innovative approach to understanding the astrophysical implications of alternative dark matter models and their effects on galaxy formation and evolution. The DREAMS project will ultimately comprise thousands of cosmological hydrodynamic simulations that simultaneously vary over dark matter physics, astrophysics, and cosmology in modeling a range of systems -- from galaxy clusters to ultra-faint satellites. Such extensive simulation suites can provide adequate training sets for machine-learning-based analyses. This paper introduces two new cosmological hydrodynamical suites of Warm Dark Matter, each comprised of 1024 simulations generated using the Arepo code. One suite consists of uniform-box simulations covering a $(25~h^{-1}~{\rm M}_\odot)^3$ volume, while the other consists of Milky Way zoom-ins with sufficient resolution to capture the properties of classical satellites. For each simulation, the Warm Dark Matter particle mass is varied along with the initial density field and several parameters controlling the strength of baryonic feedback within the IllustrisTNG model. We provide two examples, separately utilizing emulators and Convolutional Neural Networks, to demonstrate how such simulation suites can be used to disentangle the effects of dark matter and baryonic physics on galactic properties. The DREAMS project can be extended further to include different dark matter models, galaxy formation physics, and astrophysical targets. In this way, it will provide an unparalleled opportunity to characterize uncertainties on predictions for small-scale observables, leading to robust predictions for testing the particle physics nature of dark matter on these scales.

• The paper introduces the DREAMS project, which uses extensive cosmological hydrodynamic simulations and machine learning to study dark matter and galaxy formation.
• The project uses large simulation suites varying dark matter and astrophysics parameters, analyzed with machine learning like emulators and CNNs.
• Initial results show lower Warm Dark Matter particle masses suppress smaller satellite galaxy formation, providing testable predictions against observational data.

An Overview of the DREAMS Project in Astrophysics

The DREAMS (DaRk mattEr and Astrophysics with Machine learning and Simulations) project represents a significant initiative in the field of astrophysics, aiming to resolve questions concerning the nature of dark matter and its role in galaxy formation and evolution. By leveraging state-of-the-art machine learning techniques, combined with high-fidelity cosmological hydrodynamic simulations, the project seeks to uncover the effects of alternative dark matter models on different astrophysical systems ranging from galaxy clusters to ultra-faint dwarf galaxies.

The cornerstone of the DREAMS project is its robust simulation framework, generating extensive datasets that include thousands of cosmological hydrodynamic simulations. These simulations vary over dark matter physics, astrophysics, and cosmological parameters. The project's current focus has been on Warm Dark Matter (WDM) scenarios, facilitated by the development of two new cosmological hydrodynamical suites using the Arepo code. Each suite contains 1024 simulations—one with uniform-box simulations covering a $(25~h^{-1}~{\rm Mpc})^3$ volume and another consisting of Milky Way-like zoom-ins with resolutions capturing classical satellite properties.

Key Components and Results

1. Simulation Suites: The DREAMS framework involves suites that vary significantly across different assumptions in dark matter models, astrophysical processes, and cosmological parameters. This variability fosters a comprehensive analysis of the influences these factors have on galaxy formation and evolution.
2. Machine Learning Approaches: By employing machine learning models such as emulators and Convolutional Neural Networks (CNNs), the project addresses the challenge of disentangling dark matter properties from baryonic physics effects. For instance, neural network emulators are adept at interpolating within the vast parameter space to predict outcomes like satellite galaxy distributions around a Milky Way-mass halo.
3. Key Findings: Initial results hint at broader trends in the dependence of satellite galaxy numbers on WDM particle mass. These analyses highlight a suppression of smaller satellite galaxy formation with lower WDM particle masses, presenting a direct observational prediction to test against empirical data.

Implications and Future Directions

The DREAMS project holds promise for significantly advancing both theoretical and practical understandings of dark matter and galactic astrophysics. The data-rich simulation suites provide expansive training sets capable of improving machine learning techniques tailored for astrophysical applications.

Potential future developments as noted by the project include expanding analysis frameworks to alternative dark matter models, incorporating additional galaxy formation physics, and refining machine learning models to marginalize over uncertainties more robustly. These efforts will better position the DREAMS project to provide firm constraints on dark matter properties, leveraging astrophysical phenomena as empirical tests for particle physics models of dark matter.

As the project progresses, DREAMS will likely serve as a benchmark for how artificial intelligence and simulation data can harmoniously enhance our comprehension of fundamental cosmic structures and processes. The findings gained from this project will not only elucidate dark matter's nature but could also inform observatories and surveys, directing future observational strategies in cosmology and astrophysics. The collaborative and interdisciplinary nature of DREAMS underlines the power and necessity of integrating computational astrophysics with cutting-edge machine learning methodologies in contemporary scientific endeavors.