🔭 astrophysics

Impact of stellar rotation on type II supernova progenitor masses from pre-explosion imaging

This study demonstrates that incorporating the observed distribution of initial stellar rotation velocities into pre-explosion imaging analyses results in only a modest shift toward lower progenitor mass estimates for Type II supernovae, confirming that rotation has a limited impact on current mass determinations within existing uncertainties.

Original authors: L. Martinez (Instituto de Astrofísica de La Plata), O. G. Benvenuto (Instituto de Astrofísica de La Plata, Facultad de Ciencias Astronómicas y Geofísicas - UNLP), M. A. De Vito (Instituto de Astrofísi

Published 2026-03-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: L. Martinez (Instituto de Astrofísica de La Plata), O. G. Benvenuto (Instituto de Astrofísica de La Plata, Facultad de Ciencias Astronómicas y Geofísicas - UNLP), M. A. De Vito (Instituto de Astrofísica de La Plata, Facultad de Ciencias Astronómicas y Geofísicas - UNLP)

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: Weighing the Stars Before They Die

Imagine you are a detective trying to figure out how heavy a person was before they passed away. You can't weigh them directly anymore, so you look at their house, their clothes, and their car to guess their size.

In astronomy, scientists do something similar with Type II Supernovae. These are massive stars that explode at the end of their lives. Before they explode, we often have photos of them (taken years or decades earlier). By looking at how bright these "pre-explosion" stars are, astronomers try to guess their initial weight (mass).

The Problem: The standard way to do this "guessing" assumes the stars were spinning very slowly, like a sleepy cat. But in reality, massive stars are often spinning incredibly fast, like a figure skater doing a triple axel. The authors of this paper asked: "Does the star's spin change our guess about its weight?"

The Analogy: The Spinning Chef

To understand why spin matters, let's use a kitchen analogy.

Imagine a star is a chef cooking a meal (nuclear fusion) in a giant pot.

Non-Rotating Star (The Slow Chef): The chef stirs the pot very gently. The ingredients (fuel) stay in their own layers. The meal cooks at a standard pace.
Rotating Star (The Fast Chef): The chef spins the pot wildly. This creates a whirlpool that mixes the ingredients together. Fresh fuel from the bottom gets dumped into the fire, and the fire gets hotter and brighter.

The Result: Because the "Fast Chef" mixes the fuel so well, the star burns brighter and lives a slightly different life than the "Slow Chef." If you see a very bright star, you might think, "Wow, that must be a huge star!" But if that star was actually spinning fast, it might just be a medium-sized star that was spinning like crazy.

What the Scientists Did

The researchers (Martinez, Benvenuto, and De Vito) decided to stop guessing that all stars were "Slow Chefs." Instead, they looked at real data to see how fast massive stars actually spin in our galaxy.

The Data: They gathered a list of 21 stars that we know exploded as supernovae and had their "before" photos taken.
The Simulation: They used supercomputers to run two types of simulations for these stars:
- Scenario A: The stars didn't spin at all (the old way).
- Scenario B: The stars spun at speeds randomly picked from the real distribution of stars we see in the sky (the new way).
The Comparison: They compared the weight estimates from both scenarios.

The Findings: A Small Tweak, Not a Revolution

Here is the punchline: The spin didn't change the answer as much as they thought.

The Shift: When they accounted for the spin, the estimated weights of the stars did get slightly lower. It's like realizing, "Oh, that bright light wasn't a giant truck; it was a medium car with a really bright headlight."
The Magnitude: However, the difference was small. The estimated weights only shifted down by a tiny bit (about 0.2 to 1 solar mass).
The Uncertainty: The "error bars" (the margin of error) in our measurements are actually much bigger than the effect of the spin. It's like trying to weigh a person with a bathroom scale that is off by 10 pounds; whether they are spinning or not doesn't matter because the scale isn't precise enough to tell the difference.

The Conclusion: We don't need to completely overhaul our math to account for spin. The old "non-spinning" models are still good enough for now.

Solving the "RSG Problem"

There is a famous mystery in astronomy called the Red Supergiant (RSG) Problem.

The Mystery: Theory says stars up to 25 times heavier than our Sun should explode as Type II supernovae. But, in all the photos we've taken, we've never found a "before" picture of a star heavier than about 17–18 solar masses. Where are the heavy ones? Did they disappear?
The New Answer: The authors used their new "spin-aware" models to re-calculate the upper limit. They found the limit is about 20.4 solar masses.
The Verdict: This is still lower than the theoretical 25, so the mystery isn't fully solved, but it's getting closer. The "missing" heavy stars might just be a statistical fluke or hidden by dust, rather than a fundamental flaw in our physics.

The "Spin-Up" Twist (Binary Stars)

The paper also briefly mentions a third possibility: Binary Stars.
Sometimes, two stars dance around each other. One star can steal mass (and spin) from its partner. This "mass gainer" star ends up spinning super fast, even if it started slow.

The authors tested this too. They found that even if we include these "spin-up" stars, the overall weight estimates don't change drastically. The main conclusion holds: Spin is important, but it's not the magic bullet that fixes all our mass-estimation problems.

Summary in One Sentence

The authors checked if the rapid spinning of massive stars changes our estimate of their weight before they explode; they found that while spinning does make stars slightly brighter (and thus makes us guess they are slightly lighter), the difference is small enough that our current, simpler models are still doing a great job.

1. Problem Statement

Type II supernovae (SNe II) are the core-collapse explosions of massive stars ( $\gtrsim 9 M_\odot$ ) retaining hydrogen envelopes. A long-standing method to estimate the initial masses of their progenitors involves converting pre-explosion luminosity measurements into mass estimates using stellar evolution models.

The Issue: Traditionally, these estimates rely on non-rotating models or rotating models with a fixed initial rotational velocity (typically $\Omega_{ini}/\Omega_{crit} = 0.4$ ).
The Gap: Observations show that massive stars at the Zero-Age Main Sequence (ZAMS) possess a broad distribution of rotational velocities. Since rotation significantly alters stellar evolution (via enhanced chemical mixing and mass loss), assuming a fixed or zero rotation rate may bias progenitor mass estimates.
The Goal: To quantify the impact of incorporating the observed distribution of initial rotational velocities on the inferred initial masses of SN II progenitors compared to traditional non-rotating models.

2. Methodology

A. Stellar Evolution Modeling

The authors computed a comprehensive grid of single massive star models using the MESA (Modules for Experiments in Stellar Astrophysics) code (v24.08.1) at solar metallicity ( $Z_\odot = 0.0154$ ).

Parameters:
- Initial Mass ( $M_{ZAMS}$ ): Ranged from $9 M_\odot$ to $30 M_\odot$ (depending on rotation).
- Initial Surface Velocity ( $v_{ini}$ ): Varied from $0$ to $600 \text{ km s}^{-1}$ in steps of $50 \text{ km s}^{-1}$ .
- Physics: Included rotationally induced instabilities (shear, Solberg–Høiland, Eddington–Sweet, etc.) using the diffusion approximation. Internal magnetic fields were modeled via the Spruit–Tayler dynamo.
- Mass Loss: Used updated prescriptions for hot stars (Björklund et al. 2023) and cool stars (Wen et al. 2024).
Rotation Effects: The models demonstrated that rotation enhances internal mixing, supplying fresh fuel to the core. This produces more massive helium cores, leading to higher luminosities at the end of evolution compared to non-rotating stars of the same initial mass. Consequently, for a fixed observed luminosity, a rotating star implies a lower initial mass.

B. Observational Sample

The study analyzed a sample of 21 SN II progenitors detected in pre-explosion images with direct luminosity estimates (excluding those with only upper limits or uncertain identifications, e.g., SN 2003gd, SN 2005cs, SN 2020jfo).

Data: Pre-SN luminosities ( $\log L/L_\odot$ ) were taken from the literature (e.g., Davies & Beasor 2018; Van Dyk et al. 2019, 2024).

C. Statistical Inference (Monte Carlo Approach)

The authors developed a Distribution-Based Rotating (DBR) method:

Sampling: For each SN, they drew a luminosity value from a Gaussian distribution centered on the observed value (incorporating uncertainty).
Velocity Sampling: They drew an initial rotational velocity from the deconvolved observed distribution of Galactic O-type stars (Holgado et al. 2022), derived using the Lucy (1974) deconvolution algorithm.
Interpolation: Using a $k$ -dimensional tree (cKDTree), they interpolated the MESA grid to find the corresponding initial mass for the sampled luminosity and velocity.
Repetition: This process was repeated 10,000 times per SN to generate a posterior probability distribution for the initial mass.
Comparison: Results were compared against estimates derived solely from non-rotating models.

3. Key Contributions

First Application of Observed Velocity Distributions: Unlike previous studies that assumed a fixed rotation fraction, this work explicitly integrates the full, observationally derived distribution of initial rotational velocities into progenitor mass inference.
Systematic Grid Expansion: Generated a high-resolution grid of rotating models spanning a wide range of initial masses and velocities, specifically tailored for SN II progenitor analysis.
Statistical Rigor: Employed a robust Monte Carlo framework to propagate uncertainties in both luminosity and rotation into the final mass estimates, allowing for a direct statistical comparison between rotating and non-rotating scenarios.

4. Results

A. Individual Progenitor Masses

Systematic Shift: The DBR models consistently yielded lower initial mass estimates compared to non-rotating models. This is because rotating stars are more luminous for a given mass; thus, a specific observed luminosity corresponds to a lower initial mass in a rotating model.
Magnitude of Difference: The shift is modest. The median mass differences typically range from $0.2 M_\odot$ to $1.1 M_\odot$ .
Statistical Significance: Despite the systematic shift, the mass distributions from DBR and non-rotating models largely overlap. The median values from DBR models remain within $1\sigma$ of the non-rotating estimates for all SNe in the sample.

B. Population Statistics

Probability Density Functions (PDFs): The overall PDF of progenitor masses derived from DBR models is slightly shifted toward lower masses but remains very similar to the non-rotating PDF. The Hellinger distance between the two distributions is $H = 0.06$ , indicating they are nearly identical.
Upper Mass Limit ( $M_{up}$ ):
- Non-rotating: $M_{up} = 20.8^{+2.8}_{-1.3} M_\odot$ .
- DBR (Rotating): $M_{up} = 20.4^{+2.3}_{-1.9} M_\odot$ .
- The difference is smaller than the statistical uncertainty.
Lower Mass Limit: The analysis suggests a minimum initial mass of $9 M_\odot$ , though some low-luminosity progenitors (e.g., SN 2022acko) fall below the model grid's minimum luminosity, clustering at the $9 M_\odot$ floor.

C. The "RSG Problem"

The "RSG problem" refers to the apparent lack of SN II progenitors with initial masses $> 18\text{--}25 M_\odot$ , despite the existence of such stars in the galaxy.

Using the DBR models, the study finds an upper mass limit of $20.4^{+2.3}_{-1.9} M_\odot$ (68% confidence) with a 95% upper limit of $28.5 M_\odot$ .
This result maintains the statistical significance of the RSG problem at the $\sim 2\sigma$ level, consistent with recent studies (e.g., Davies & Beasor 2020; Beasor et al. 2025). Rotation does not resolve the tension.

D. Secondary Channels (Mass Gainers)

The authors briefly explored the impact of binary mass gainers (stars that accrete mass and spin up to critical rotation). Including a 14% contribution from mass gainers slightly lowered the inferred mass for SN 2025pht ( $13.7 M_\odot$ vs $14.3 M_\odot$ ), but the result remained consistent within $1\sigma$ of the non-rotating estimate.

5. Significance and Conclusions

Modest Impact of Rotation: While stellar rotation physically alters evolutionary tracks, its impact on inferred progenitor masses from pre-explosion imaging is modest when the observed distribution of velocities is accounted for.
Validation of Current Practices: The study concludes that, given current observational uncertainties and model limitations, non-rotating models remain sufficient for estimating SN II progenitor masses. Explicitly incorporating the full velocity distribution does not significantly improve precision or alter the statistical conclusions regarding the progenitor mass distribution.
Future Directions: The authors note that uncertainties in low-mass progenitor modeling (the "faint end" of the luminosity distribution) and metallicity effects remain larger sources of error than the inclusion of rotation. Future work should explore metallicity variations and binary interaction channels more comprehensively.

In summary, the paper demonstrates that while rotation is a critical physical process, the assumption of non-rotating models in SN II progenitor mass estimation is a robust approximation within current error bars, and the "RSG problem" persists regardless of whether rotation is modeled via a fixed fraction or an observed distribution.