The MUSE-Faint survey V. The binary fraction of Leo T

Using MUSE-Faint multi-epoch spectroscopy, this study presents the first measurement of the binary fraction in the Leo T dwarf galaxy, finding an overall rate of approximately 55% that varies with stellar age and metallicity but does not significantly inflate the galaxy's inferred velocity dispersion.

The Gist

The Cosmic Detective Story: Unmasking Leo T's Hidden Twins

Imagine you are trying to weigh a ghost. You can't see it, but you know it's there because it's pulling on a nearby swing set. This is exactly what astronomers do with Leo T, a tiny, faint dwarf galaxy. It's so dim and small that it's barely visible, yet it holds a massive amount of invisible "dark matter."

To figure out how heavy Leo T is, astronomers measure how fast the stars inside it are moving. If the stars are zooming around wildly, the galaxy must be heavy (to hold them together). If they are moving slowly, the galaxy is light.

But here's the problem: Binary stars (two stars orbiting each other like a cosmic dance pair) can trick the scale.

The "Dancing Couple" Problem

Imagine you are watching a crowd of people walking down a street. You want to measure the average speed of the crowd.

• Scenario A: Everyone is walking at a steady 3 mph. Easy to measure.
• Scenario B: Most people are walking at 3 mph, but a few couples are holding hands and spinning around each other while walking. To a camera taking a quick snapshot, those spinning couples look like they are moving much faster than everyone else.

If you don't realize they are just dancing in place, you might think the whole crowd is moving incredibly fast. In astronomy, this "fake speed" makes the galaxy look heavier than it actually is. This is the "binary contamination" problem.

The Mission: The MUSE-Faint Survey

The authors of this paper used a powerful telescope camera called MUSE (mounted on the Very Large Telescope in Chile) to take a "movie" of Leo T. Instead of just one photo, they took snapshots over several years (multi-epoch observations).

By watching the stars over time, they could see which ones were wiggling back and forth (the dancing couples) and which ones were walking straight (the single stars).

What They Found

1. The "Twins" are Everywhere
They calculated that about 55% of the stars in Leo T are actually part of a binary system.

• The Analogy: Imagine walking into a party where you expect everyone to be alone. You discover that more than half the guests are actually holding hands with a partner. This matches what we see in our own neighborhood (the Milky Way), suggesting that "dancing couples" are a universal habit, even in the most remote, metal-poor corners of the universe.

2. The Age Gap: Young vs. Old
Leo T is special because it has two distinct groups of stars:

• The Youngsters: Stars born less than 1 billion years ago.
• The Elders: Stars born over 5 billion years ago.

They found that the Youngsters have a higher rate of binary pairs (about 35%) compared to the Elders (about 15%).

• The Analogy: Think of the Youngsters as energetic teenagers who love to pair up and dance. The Elders are like the older generation; over the billions of years, some of their dance partners have drifted away, or the dance floor (the galaxy's gravity) has been too rough, breaking up the pairs. Also, younger stars are generally more massive, and massive stars tend to have partners more often.

3. The Big Surprise: The Scale Wasn't Broken
The biggest question was: Did these dancing couples trick us into thinking Leo T is heavier than it is?

The authors ran a simulation to see if removing the "dancers" would change the weight of the galaxy.

• The Result: Surprisingly, no. Even after removing the binary stars, the calculated speed of the galaxy didn't change much.
• Why? This is the clever twist. When astronomers combined all their snapshots into one final image (a "co-added spectrum"), the telescope effectively took the average speed of the dancing couples.
  • The Analogy: Imagine taking a long-exposure photo of a spinning dancer. Instead of seeing a blur of motion, the camera blurs them into a single, stationary-looking shape in the middle of the spin. The telescope's method of averaging the light over time accidentally "smoothed out" the wobble, hiding the binary motion. So, the previous measurements of Leo T's weight were actually correct, even without knowing about the binaries!

Why This Matters

This paper is like a quality control check for the universe's most mysterious objects.

1. Confirmation: It confirms that even in the smallest, most ancient galaxies, stars love to pair up.
2. Metallicity Clue: It supports the idea that in very "metal-poor" environments (like Leo T, which is made of older, simpler ingredients), close binary pairs are slightly less common than in richer environments, but still very frequent.
3. Reliability: It reassures astronomers that their previous measurements of how heavy these tiny galaxies are (and how much dark matter they hold) are trustworthy. The "dancing couples" didn't ruin the math after all.

In a nutshell: The astronomers watched a tiny, dark galaxy for years, found that half its stars are dancing in pairs, realized that the dance moves were actually hidden by their camera's averaging trick, and confirmed that the galaxy is indeed a dark matter heavyweight champion.

Technical Summary

1. Problem Statement

Ultra-faint dwarf (UFD) galaxies are critical for understanding galaxy formation and dark matter dominance. Their dynamical masses are derived from stellar velocity dispersions ($\sigma$), which are typically very low ($\sigma < 10$ km s$^{-1}$). A major systematic uncertainty in these measurements is the potential inflation of $\sigma$ caused by the orbital motion of unresolved binary stars.

• The Challenge: In systems with low intrinsic velocity dispersion, binary orbital motions can artificially inflate the measured dispersion by over 100%, leading to overestimates of dark matter content.
• The Gap: While binary fractions in the Milky Way and Magellanic Clouds are well-studied, constraints on binary populations in metal-poor UFDs are scarce due to small sample sizes and the difficulty of obtaining multi-epoch spectroscopy.
• Specific Context: Leo T is a unique UFD with recent star formation ($\le$ 1 Gyr) and two distinct stellar populations (young and old). Previous work (Vaz et al. 2023, "MFIV") identified these populations but did not quantify the binary fraction or its specific impact on the derived velocity dispersion.

2. Methodology

The authors utilized multi-epoch spectroscopic data from the MUSE-Faint survey (using the MUSE instrument on the VLT) to perform the first binary fraction measurement for Leo T.

• Data: Five observation blocks (epochs) spanning 2018–2022, yielding high-quality spectra for 55 member stars (44 overlapping with previous studies, 11 newly included).
• Stellar Characterization:
  • Velocities: Measured using spexxy (full-spectrum fitting against PHOENIX models).
  • Masses & Ages: Derived by fitting PARSEC isochrones to HST photometry. Stars were categorized into Young ($\le$ 1 Gyr, 20 stars) and Old ($\ge$ 5 Gyr, 35 stars) populations.
• Forward Modeling Approach:
  • The team employed a forward modeling strategy (based on Giesers et al. 2019 and Arroyo-Polonio et al. 2023) to simulate mock datasets.
  • Simulation Parameters: They generated 10,000 mock samples for binary fractions ($f$) ranging from 0% to 100%.
  • Binary Physics: Simulations incorporated:
    • Orbital Elements: Period ($P$), eccentricity ($e$), mass ratio ($q$), and inclination ($i$), drawn from distributions based on Moe & Di Stefano (2017).
    • Constraints: Periods were limited to $P \le 10^6$ days (wide binaries) and specifically $a < 10$ au (close binaries) to test sensitivity.
    • Flux Ratio: Accounted for the damping of radial velocity amplitude ($v_b$) due to the flux ratio of the binary components.
  • Statistical Comparison: The observed distribution of reduced $\chi^2$ (measuring velocity variability across epochs) was compared against the simulated distributions using a likelihood function based on the Cumulative Distribution Function (CDF).
  • Uncertainty Estimation: A bootstrap approach was used to determine 68% credible intervals.

3. Key Contributions

• First Measurement: This is the first direct measurement of the binary fraction in the Leo T dwarf galaxy.
• Population-Specific Analysis: The study is the first to separate and analyze binary fractions for distinct young and old stellar populations within a single UFD.
• Methodological Validation: The paper validates the use of co-added (multi-epoch) spectra for velocity dispersion measurements, demonstrating that this technique effectively averages out binary orbital motions, thereby mitigating their inflationary effect on $\sigma$.
• Metallicity Constraints: It provides a data point for the relationship between close binary fractions and metallicity in extremely metal-poor environments ($[\text{Fe/H}] \approx -1.6$).

4. Key Results

A. Binary Fractions

• Overall Fraction: The total binary fraction for Leo T is estimated at **$55^{+40}_{-9}$$\sim$50–60% fractions found in other dwarf spheroidals and the Milky Way.
• Close Binaries ($a < 10$ au): The fraction of close binaries is $30^{+34}_{-9}$%. This value aligns with the trend of decreasing close binary fractions in lower metallicity environments (anti-correlation with metallicity), consistent with findings by Moe et al. (2019).
• Population Differences:
  • Young Population: $35^{+40}_{-6}$% (higher fraction).
  • Old Population: $15^{+43}_{-15}$% (lower fraction).
  • Interpretation: The higher fraction in the young population is attributed to the correlation between stellar multiplicity and primary mass (young stars are more massive). The lower fraction in the old population suggests evolutionary disruption (Roche-lobe overflow) and dynamical interactions within the dense dark matter halo of the UFD.

B. Impact on Velocity Dispersion

• No Significant Inflation: The authors tested whether removing likely binary candidates (stars with high $\chi^2$) would significantly lower the velocity dispersion.
  • Full Sample: Dispersion remained stable ($\sigma \approx 6.6$ km s$^{-1}$ vs. $7.2$km s$^{-1}$ after removal).
  • Young vs. Old: While the young population showed a slight deflation in $\sigma$ after removing binaries, the high uncertainties meant the results remained consistent with previous measurements.
• Mechanism: The study concludes that the co-added spectra used in the previous MFIV analysis effectively provided period-averaged velocities. Because the orbital periods of most binaries are long relative to the observation baseline, or because the spectra were averaged over time, the binary orbital motion did not artificially inflate the global velocity dispersion.

5. Significance and Conclusions

• Dark Matter Constraints: The finding that binary motion does not significantly inflate the velocity dispersion in Leo T reinforces the reliability of previous mass estimates for this galaxy. Leo T remains a system dominated by dark matter with a high mass-to-light ratio.
• Evolutionary Insights: The difference in binary fractions between the young and old populations in Leo T offers a unique laboratory to study binary evolution and disruption mechanisms (tidal effects, Roche-lobe overflow) in low-metallicity, dark-matter-dominated environments.
• Future Directions: The study highlights the need for larger samples and more epochs to break degeneracies in period distributions, particularly for wide binaries. It establishes a robust methodology for future MUSE-Faint surveys to characterize binary populations in other ultra-faint dwarfs.

In summary, this paper confirms that Leo T's binary population is consistent with broader galactic trends but exhibits population-dependent variations. Crucially, it demonstrates that for Leo T, the standard practice of co-adding multi-epoch spectra successfully mitigates the systematic bias binaries introduce to velocity dispersion measurements.