🔭 astrophysics

Intermediate-Mass Stripped Stars in the Magellanic Clouds: Forward Modeling the Observed Population Discovered Via UV Excess

By forward modeling the population of intermediate-mass stripped stars in the Magellanic Clouds and simulating the SUMS survey selection process, this study quantifies the survey's low completeness and purity, identifies key observational biases and contamination sources, and demonstrates that the model successfully reproduces the observed population of candidates with masses between 2 and 5 solar masses.

Original authors: Lisa Blomberg, Kareem El-Badry, Bethany Ludwig, Maria Drout, Ylva Gotberg

Published 2026-01-22

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Lisa Blomberg, Kareem El-Badry, Bethany Ludwig, Maria Drout, Ylva Gotberg

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Finding the "Naked" Stars

Imagine a star as a giant, glowing onion. Most stars have a thick, outer layer of hydrogen (the "skin") that hides their hot, helium-rich core. But in some binary star systems (two stars orbiting each other), one star acts like a cosmic peeler. Through a violent dance of gravity, it strips the hydrogen skin off its partner, leaving behind a hot, glowing "naked" core.

Astronomers call these stripped stars.

Low-mass stripped stars are like small, hot embers (known as hot subdwarfs).
High-mass stripped stars are like massive, roaring fires (known as Wolf-Rayet stars).
Intermediate-mass stripped stars (the focus of this paper) are the "Goldilocks" zone—big enough to be rare, but small enough to be hard to find.

The Problem: The "Needle in a Haystack"

A recent survey called SUMS (Stripped-Star Ultraviolet Magellanic Cloud Survey) looked at two nearby galaxies, the Large and Small Magellanic Clouds. They used a special telescope (Swift-UVOT) that sees ultraviolet (UV) light. Because stripped stars are so hot, they glow brightly in UV, making them look "bluer" than normal stars.

The survey found about 820 candidates that looked like stripped stars. However, the astronomers had a big question: How many did they miss? And how many of the 820 are actually imposters?

It's like trying to count rare, glowing fireflies in a forest at night. Some fireflies are hidden behind thick bushes (dust), some are outshined by a bright flashlight (a companion star), and some "glowing" spots are just reflections off a wet leaf (contamination).

The Solution: The "Virtual Forest" Simulation

Instead of just guessing, the authors built a virtual forest (a computer simulation) to see how well their search method works. Here is how they did it:

Building the Population: They used a "recipe book" of how stars are born and evolve to create a fake population of 3,500 stripped stars in the Magellanic Clouds. They knew exactly where these fake stars were, how bright they were, and who their neighbors were.
Injecting the Stars: They took real images from the Swift telescope and digitally "injected" their fake stripped stars into the pictures, just like placing a digital sticker on a photo.
Running the Search: They ran the exact same computer code that the SUMS survey used to find the real candidates. They asked the computer: "Can you find our fake stars in this messy, crowded picture?"

The Results: The "Lost and Found" Report

The simulation revealed some surprising truths about the search:

The "Missing" Majority: The survey only found about 11% to 31% of the stripped stars that actually existed.
- Why? Many were hidden. Some were covered by dust (like a star behind a foggy window). Some were paired with a very bright companion star that drowned out the stripped star's UV glow (like trying to see a candle next to a stadium floodlight). Others were in crowded areas where the telescope couldn't tell two stars apart.
The Bias: The survey is biased toward finding specific types of systems. It mostly finds stripped stars that have small, dim companions or no companion at all. It misses the ones with big, bright partners.
The Imposters: The survey also picked up some "false alarms." These were normal stars that looked like stripped stars because of measurement errors or crowded neighborhoods. The authors estimate that about 10% of the "Very Blue" candidates might be fake.

The Verdict: A Good Start, But Not Perfect

The authors compared their virtual results with the real data:

The Good News: For the most common size of stripped stars (between 2 and 5 times the mass of our Sun), the simulation matched the real observations perfectly. This gives astronomers confidence that their theory of how these stars form is correct.
The Bad News: The simulation found fewer very small and very large stripped stars than the survey did. The authors suspect this is because the survey is picking up too many "imposters" in those size ranges, inflating the numbers.

The Takeaway

This paper is like a quality control check for a treasure hunt. The authors didn't just find the treasure; they built a map of the whole island to see how many treasures were left buried and how many "treasures" were actually just rocks painted gold.

They concluded that while the SUMS survey found a great starting list of candidates, it missed a huge number of real stripped stars and included a few fakes. By understanding these "blind spots," astronomers can now refine their search to find the true population of these mysterious, stripped stars.

Technical Summary: Intermediate-Mass Stripped Stars in the Magellanic Clouds

Problem Statement
Intermediate-mass stripped stars ( $1 - 8 M_\odot$ ), formed when binary interactions remove a star's hydrogen envelope, have only recently been discovered in the Magellanic Clouds (MCs). While low-mass (hot subdwarfs) and high-mass (Wolf-Rayet) stripped stars are well-studied, the intermediate-mass population remains poorly constrained. The Stripped-Star Ultraviolet Magellanic Cloud Survey (SUMS) identified 820 photometric candidates based on UV excess (sources lying blueward of the Zero-Age Main Sequence). However, the completeness and purity of this sample are unknown. Key selection biases include the dilution of UV excess by luminous main-sequence (MS) companions, crowding in dense regions, dust extinction, and the finite spatial coverage of the survey. Without quantifying these biases, the underlying population demographics (occurrence rates, mass distribution, binary fractions) cannot be accurately inferred, limiting tests of binary evolution models and the understanding of stripped stars as progenitors of compact object binaries and stripped-envelope supernovae.

Methodology
The authors employ a forward-modeling approach to simulate the SUMS survey selection process and assess the recoverability of stripped stars.

Synthetic Population Generation: A synthetic population of stripped-star binaries is created using a binary population synthesis model based on MESA evolutionary tracks (Götberg et al. 2018). The model samples from a $10^6 M_\odot$ starburst population, constrained by spatially resolved star formation histories (SFHs) of the SMC and LMC (Rubele et al. 2018; Mazzi et al. 2021). The simulation covers the past 100 Myr and includes systems with MS, post-MS, and compact object companions.
Physical Modeling: The model incorporates 3D dust distributions derived from 3D dust maps (Edenhofer et al. 2024) and Schlegel et al. (1998) to compute line-of-sight extinction for each source. Intrinsic UV excess is calculated based on the relative flux contributions of the stripped star and its companion.
Mock Observations: Synthetic sources are injected into real Swift-UVOT images used by the SUMS survey. The authors utilize The Tractor (Lang et al. 2016) to perform forced PSF photometry, replicating the survey's handling of crowding and source confusion.
Selection Reproduction: The mock photometry undergoes the exact same selection pipeline as the SUMS catalog, including photometric quality cuts, SED consistency checks, and UV-excess thresholds (requiring excess in $\ge 4$ of 9 UV-optical color combinations).
Contamination Assessment: The framework is also used to forward-model a population of MS stars to estimate the rate of false positives caused by crowding and extinction over-correction.

Key Results

Recovery Rates: The survey is incomplete. For the SMC and LMC, the authors recover only 31% and 15% of sources with intrinsic UV excess, respectively. When considering the total population of stripped stars (including those without intrinsic UV excess), the recovery rates drop to 11% (SMC) and 7% (LMC).
Sources of Incompleteness: The majority of missed sources are lost due to:
- Dilution: Luminous MS companions outshine the stripped star, masking the UV excess.
- Crowding: Blending with nearby sources degrades photometric quality or dilutes the signal.
- Extinction: High dust extinction reddens sources, moving them onto the main sequence in color-magnitude diagrams (CMDs).
- Coverage: Sources falling outside the Swift-UVOT footprint.
Selection Biases: The observed sample is heavily biased toward systems with low-mass companions (formed via common envelope evolution) and systems with compact object companions. Systems with massive MS companions are largely undetectable via UV excess.
Purity and Contamination: The study predicts that the observed sample contains contamination from MS stars exhibiting spurious UV excess due to crowding or over-corrected extinction. The simulation suggests that "Blue" candidates (weaker UV excess) are more likely to be false positives than "Very Blue" candidates.
Population Comparison: The population synthesis model matches the number and properties of observed candidates with masses of $(2 - 5) M_\odot$ well. However, it underpredicts the number of lower- and higher-mass candidates, a discrepancy the authors attribute partly to contamination in the observed sample.

Significance and Claims
The paper establishes a rigorous framework for interpreting UV-excess surveys of stripped stars. By quantifying the survey's selection function, the authors demonstrate that the current SUMS catalog represents a highly biased subset of the true stripped-star population. The work provides:

Completeness Estimates: Quantitative recovery rates that allow for the inference of the true underlying population density in the MCs.
Purity Guidelines: Criteria for selecting higher-purity subsamples (e.g., prioritizing "Very Blue" candidates) to minimize contamination from MS stars.
Model Validation: A method to test binary population synthesis models against observed data, accounting for observational biases rather than raw counts.

The authors conclude that while their model successfully reproduces the intermediate-mass range ( $2-5 M_\odot$ ) of the observed sample, the underprediction at mass extremes and the presence of contaminants highlight the need for spectroscopic follow-up of the full SUMS catalog to fully constrain the demographics of intermediate-mass stripped stars.