Galactic Archaeology with the Subaru `\=Onohi`ula Prime Focus Spectrograph Strategic Program

The Subaru Strategic Program for the Prime Focus Spectrograph (PFS) will conduct a comprehensive Galactic Archaeology survey over 130 nights to investigate dark matter density profiles in dwarf galaxies, compare the assembly histories of the Milky Way and M31, and trace the Milky Way's accretion history by obtaining spectra for approximately 68,000 stars across the Local Group.

The Gist

Imagine the universe as a giant, sprawling city that has been under construction for 13.8 billion years. Most of the time, we look at this city from a very far distance, seeing only the glowing lights of entire neighborhoods (galaxies) but unable to see the individual bricks or the people who built them.

Galactic Archaeology is the science of trying to figure out how this city was built by looking at the individual "bricks"—the stars.

This paper describes a massive new project called the Subaru PFS-SSP Galactic Archaeology Survey. Think of it as a high-tech, super-powered "time machine" and "fingerprint scanner" rolled into one giant telescope instrument called the Prime Focus Spectrograph (PFS), located on top of a mountain in Hawaii.

Here is a simple breakdown of what they are doing and why it matters, using some everyday analogies:

1. The Tool: A Giant Spider Web

The Subaru telescope is like a giant eye. The new instrument, PFS, is like a giant spider web with nearly 2,400 tiny threads (fibers).

• How it works: Instead of looking at one star at a time, the astronomers can drop these 2,400 threads onto 2,400 different stars simultaneously.
• The Result: In a single night, they can capture the "voice" (spectrum) of thousands of stars at once. It's like being able to record the voices of an entire stadium of people in one second, rather than interviewing them one by one.

2. The Three Big Mysteries They Are Solving

The team has divided their 360-night mission into three main detective stories:

Mystery #1: The Ghost in the Dwarf Galaxies (Dark Matter)

• The Scene: There are tiny, faint galaxies called "dwarf spheroidals" that orbit our Milky Way. They are mostly made of invisible Dark Matter (the "ghost" that holds galaxies together).
• The Question: Is the ghost a "cusp" (a sharp, pointy spike of density in the center) or a "core" (a fluffy, flat ball of density)?
  • Analogy: Imagine a snowman. Is the snow packed tight into a hard, pointy nose (cusp), or is it a soft, round ball (core)?
• The Plan: They will measure the speed of 18,000 stars in six of these tiny galaxies. By seeing how fast the stars are moving, they can weigh the invisible ghost. If the stars are moving fast in the center, it's a cusp (standard theory). If they are moving slower, it's a core (suggesting the "ghost" behaves differently or that the stars themselves pushed the ghost around).

Mystery #2: The Family Feud (Milky Way vs. Andromeda)

• The Scene: Our galaxy (the Milky Way) and our neighbor, the Andromeda galaxy (M31), are the two biggest kids in the neighborhood.
• The Question: Did they grow up the same way?
  • Analogy: Imagine two families. One family grew up in a quiet, peaceful house with no fights (Milky Way). The other family grew up in a house that was constantly being remodeled by loud construction crews and moving trucks (Andromeda).
• The Plan: They will look at 30,000 stars in Andromeda's outer edges. By analyzing the chemical "fingerprint" of these stars (specifically the ratio of "alpha" elements like oxygen to iron), they can tell if Andromeda had a violent past with big mergers, or if it was as quiet as our own galaxy. This helps us understand if our quiet history is normal or a lucky fluke.

Mystery #3: The Earthquake Aftermath (The Milky Way's Outer Disk)

• The Scene: Our galaxy isn't perfectly still. It's been shaken by smaller galaxies crashing into it, like a boat rocking after a wave hits it.
• The Question: How does the Milky Way's outer edge react to these crashes?
  • Analogy: Imagine dropping a stone into a pond. The ripples take a long time to settle. The outer edge of our galaxy is like the water far from the center; it's still rocking from crashes that happened billions of years ago (like the "Gaia-Sausage" crash) and recent ones (like the Sagittarius dwarf galaxy).
• The Plan: They will measure the speed and age of stars in the very outer rim of our galaxy. By mapping these "ripples," they can figure out exactly when and how big the crashes were, effectively reconstructing the history of our galaxy's "earthquakes."

3. Why This Matters

Before this project, we were like archaeologists trying to understand a whole civilization by looking at a single, dusty coin. We had to guess the rest.

With this new "2,400-thread web," they are going to find 100,000 new stars and measure their speeds and chemical makeup with incredible precision.

• For Dark Matter: It might prove or disprove our current theories about the invisible stuff that makes up most of the universe.
• For History: It will tell us if our galaxy is a "quiet" outlier in the universe or if violent mergers are the norm.
• For the Future: It creates a detailed map of our cosmic neighborhood, showing us where we came from and how the universe is put together.

In short, this paper is the blueprint for a massive, 6-year expedition to read the "diary" of the universe, written in the light of stars, to finally understand how our cosmic home was built.

Technical Summary

1. Problem Statement

Understanding the formation and evolution of large disk galaxies like the Milky Way (MW) and Andromeda (M31) requires "Galactic Archaeology"—the reconstruction of galactic history through the properties of resolved stars. Key unresolved questions include:

• Dark Matter Nature: Do dwarf spheroidal galaxies (dSphs) possess "cuspy" dark matter density profiles (predicted by standard $\Lambda$CDM) or "cored" profiles (predicted by alternative dark matter theories or baryonic feedback)? Current samples are too small to distinguish between these models with high confidence.
• Assembly Histories: Did M31 and the MW form via similar merger histories? While the MW appears to have had a relatively quiescent recent history (dominated by the Gaia-Sausage-Enceladus event ~10 Gyr ago), M31's history is less constrained.
• Dynamical Response: How do the outer regions of the MW (disk and halo) respond to ongoing perturbations from satellite galaxies like Sagittarius and the Large Magellanic Cloud (LMC)?
• Sample Limitations: Previous spectroscopic surveys (e.g., DESI, APOGEE, Keck/DEIMOS) have been limited by sample sizes (typically <1,000 stars per dwarf galaxy) and depth, preventing robust statistical analysis of sub-populations, bursty star formation, and the outermost tidal regions of galaxies.

2. Methodology

The paper outlines the Subaru Strategic Program (SSP) for Galactic Archaeology (GA) using the Prime Focus Spectrograph (PFS) on the 8.2m Subaru Telescope.

• Instrument Capabilities: PFS utilizes 2,394 reconfigurable fibers over a 1.24 deg$^2$ field of view. It employs three spectrograph arms: Blue (380–650 nm, $R \sim 1900$), Medium-Resolution Red (710–885 nm, $R \sim 5000$), and Near-Infrared (970–1260 nm, $R \sim 3500$). The Medium-Resolution (MR) red arm is critical for measuring precise radial velocities ($\sim 3$ km s$^{-1}$) and chemical abundances.
• Survey Strategy (3 Pillars):
  1. Dwarf Galaxies (dSphs & dIrr): Targeting 6 classical dSphs (Boötes I, Draco, Ursa Minor, Sextans, Sculptor, Fornax) and 1 dwarf irregular (NGC 6822). The goal is to obtain $\sim 18,000$ member star spectra extending beyond nominal tidal radii.
  2. M31 and M33: Surveying 30,000 stars across M31's halo and outer disk (45 deg$^2$) and 2,000 stars in M33's halo to map assembly history via chemical abundance patterns ($[\alpha/\text{Fe}]$ vs. $[\text{Fe/H}]$).
  3. Milky Way Structure: Observing tens of thousands of main-sequence turn-off (MSTO) and giant stars in the outer MW disk and halo (out to $\sim 30$ kpc) to study dynamical responses to mergers and map phase-space substructures.
• Target Selection & Pre-imaging:
  • Deep Hyper Suprime-Cam (HSC) pre-imaging (broadband $g, i$ and narrowband NB515) is used to select candidate members based on color-magnitude diagrams (CMDs) and gravity-sensitive features (separating giants from foreground dwarfs).
  • Machine learning techniques are employed to select metal-rich RGB stars where photometric separation is difficult.
  • Membership probabilities are calculated using stochastic modeling of CMDs (Dartmouth isochrones) combined with Gaia proper motions.
• Observing Strategy:
  • Binarity Mitigation: Key fields (especially in dSphs) are observed twice (separated by months/years) to identify and model binary star orbital motions, which otherwise inflate velocity dispersion.
  • Exposure Times: 3 hours for dSphs and MW fields; 5 hours for M31.
  • Fiber Allocation: A flow-network algorithm optimizes fiber assignment based on membership probability, prioritizing stars that maximize the scientific yield for dark matter modeling.

3. Key Contributions & Technical Innovations

• Unprecedented Sample Sizes: The survey aims for $\sim 100,000$ total stellar spectra, with $\sim 18,000$ in dSphs and $\sim 30,000$ in M31. This represents a 5–10x increase over previous surveys, enabling the splitting of samples into radial bins and kinematic sub-populations.
• Advanced Dynamical Modeling:
  • Moving beyond simple spherical Jeans analysis to non-spherical Jeans modeling and higher-order velocity moments (kurtosis) to break the mass-anisotropy degeneracy.
  • Utilizing GravSphere, MAMPOSSt, and AGAMA to reconstruct dark matter density profiles without assuming spherical symmetry.
  • Implementing Schwarzschild orbit superposition to model complex kinematics.
• Chemical Evolution Precision:
  • Measuring $[\alpha/\text{Fe}]$ to better than 0.1 dex and individual element abundances (up to 18 elements) for thousands of stars.
  • Using these abundances to reconstruct bursty star formation histories (SFHs), which are hypothesized to transform dark matter cusps into cores via baryonic feedback.
• Pipeline Development:
  • Development of a custom template-fitting algorithm for radial velocities that handles low Signal-to-Noise (S/N) data ($S/N \sim 3$) using likelihood stacking and Monte Carlo sampling to accurately characterize error distributions.
  • Two-stage abundance analysis: (1) Global parameters ($T_{\text{eff}}$, $\log g$, $[\text{M/H}]$, $[\alpha/\text{M}]$) via grid fitting; (2) Detailed element abundances via spectral synthesis (SYNTHE/BasicATLAS).

4. Expected Results

• Dark Matter Constraints: Definitive determination of whether the inner density slopes ($\gamma_{\text{DM}}$) of six dSphs are consistent with NFW cusps ($\gamma \approx -1$) or cores ($\gamma \approx 0$). The survey will correlate these profiles with star formation burstiness derived from $[\alpha/\text{Fe}]$ patterns.
• M31 vs. MW Assembly:
  • Testing if M31's disk shows a bimodal chemical structure (thick vs. thin disk) similar to the MW. A lack of bimodality or different chemical patterns would imply a violent, major-merger dominated history for M31, contrasting with the MW's quiet past.
  • Mapping the phase-space structure of M31's streams (e.g., Giant Southern Stream) to constrain the mass ratios and timing of past mergers.
• MW Dynamics:
  • Quantifying the "breathing" and "bending" modes of the MW outer disk induced by the Sagittarius dwarf and LMC.
  • Identifying the "splashback radius" and the edge of the MW stellar halo using Blue Horizontal Branch (BHB) stars and extremely metal-poor (EMP) stars.
  • Characterizing the kinematic wake of the LMC in the distant halo.
• Ancillary Science:
  • Discovery of rare Extremely Metal-Poor (EMP) stars and Ultracool Dwarfs (UCDs) to probe the first generations of stars and the cosmological lithium problem.
  • Chemical abundance measurements of Planetary Nebulae (PNe) in M31 and NGC 6822 to trace enrichment history.

5. Significance

This program represents a paradigm shift in Galactic Archaeology by leveraging the massive multiplexing of PFS to move from studying individual stars or small clusters to statistical populations.

• Dark Matter Physics: It provides the most stringent test to date of the "cusp-core" problem, potentially validating or ruling out specific dark matter models (e.g., SIDM, FDM) or confirming the role of baryonic feedback in shaping halos.
• Galaxy Formation Theory: By comparing the assembly histories of the MW and M31, it tests the universality of galaxy formation scenarios in the $\Lambda$CDM framework, specifically addressing whether "quiet" accretion histories are rare or common.
• Methodological Benchmark: The development of robust pipelines for low-S/N, high-volume spectroscopy sets a new standard for future large-scale surveys (e.g., for the Vera C. Rubin Observatory or future ELTs).

The survey is scheduled to run for six years (2025–2030), with data releases including spectra, radial velocities, and chemical abundances, fundamentally advancing our understanding of the Local Group's evolution.