🔭 astrophysics

The initial conditions and initial mass functions of Alpha Persei, Pleiades and Praesepe

Using Gaia DR3 data combined with N-body simulations and machine learning, this study reconstructs the initial mass functions and structural properties of the Alpha Persei, Pleiades, and Praesepe clusters, revealing a top-light, broken power-law IMF with significant high-mass scatter that challenges the universality of the initial mass function across different environments.

Original authors: L. Hobart, H. Baumgardt, S. Sweet

Published 2026-07-21

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: L. Hobart, H. Baumgardt, S. Sweet

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

Imagine the universe as a giant, bustling construction site where stars are the bricks being laid down. For decades, astronomers have wondered if the "blueprints" for these bricks are the same everywhere. Do star factories in dense, crowded neighborhoods build the same mix of giant, heavy stars and tiny, lightweight ones as factories in quiet, open fields? This question is at the heart of the Initial Mass Function (IMF). Think of the IMF as a recipe card that tells us how many stars of each size are born in a cluster. If the recipe is universal, every star cluster should have the same ratio of massive, bright giants to small, dim dwarfs. But if the recipe changes based on the environment—like how a baker might use more flour in a humid kitchen versus a dry one—then the universe is far more diverse than we thought. Understanding this matters because the number of massive stars dictates how galaxies shine, how heavy elements are forged, and what kind of "stellar corpses" (like black holes) are left behind.

In this new study, a team of astronomers from the University of Queensland decided to investigate this recipe by looking at three famous "star nurseries" in our own cosmic backyard: Alpha Persei, the Pleiades, and Praesepe. These are open clusters, which are like loose groups of stars that formed together from the same cloud of gas. The researchers acted like cosmic detectives, using the incredible precision of the Gaia space telescope (which maps star positions and movements) to count the stars and figure out how heavy they are. They had to be very careful, though. Just like a photographer might accidentally blur two people standing close together into one giant blob, telescopes often see two stars orbiting each other as a single, brighter star. This can trick astronomers into thinking a star is heavier than it really is. The team used clever computer simulations to "un-blur" these pairs and get a true count.

Once they had the clean data, they didn't just stop at what the clusters look like today. They knew that over millions of years, gravity acts like a chaotic dance floor. Heavier stars tend to sink toward the center, while lighter ones get kicked out to the edges or even thrown out of the cluster entirely. To see the original recipe, the team ran thousands of N-body simulations—essentially, high-speed movies of how these clusters would evolve over time. They used machine learning to speed up these movies and work backward from the current state of the clusters to guess what they looked like when they were just babies.

What they found suggests that the universe's star-formation recipe might not be as standard as we hoped. The team discovered that these three clusters seem to have a "top-light" IMF. In everyday terms, this means they are missing some of the heavyweights. Compared to the standard "Salpeter" recipe (which predicts a certain number of massive stars), these clusters have fewer stars heavier than 1 solar mass ( $1 M_{\odot}$ ) than expected. The data shows a best-fitting initial mass function that follows a three-stage broken power-law distribution. The "breaks" in the recipe—where the slope changes—happen at masses between 0.24–0.50 $M_{\odot}$ and 0.91–1.20 $M_{\odot}$ .

The slopes of this recipe are quite specific. For the intermediate-mass stars, the slope is $\alpha_{med} = 1.72 \pm 0.09$ , and for the high-mass stars, it is $\alpha_{high} = 2.98 \pm 0.22$ . The high-mass slope is notably steeper than the classic Salpeter value of 2.35, reinforcing the idea that these clusters are "top-light" (lacking the heaviest stars). However, the team also found that the recipe isn't perfectly identical across all three clusters. There is some "scatter" or variation in the high-mass slope between them, measured as $\sigma_{high} = 0.29 \pm 0.16$ , while the intermediate-mass slope seems much more consistent, with a variation of $\sigma_{med} = 0.00 \pm 0.14$ .

The study also had to account for the "hidden" population of binary stars (pairs of stars orbiting each other). By using photometry and Monte Carlo simulations in a Bayesian framework, they calculated that the current unresolved binary fractions in these clusters are between (20.0 ± 0.8)% and (23.8 ± 1.2)%. When they corrected for these pairs, the picture of the star masses became even clearer.

The researchers also looked at how the clusters have changed since their birth. They found that the initial number of stars in these clusters was likely between 1,100 and 2,100 systems, with initial binary fractions ranging from 24% to 35%. The clusters were born with 3D half-mass radii between 4 and 6 pc. Over time, the older cluster, Praesepe, has lost a significant fraction of its low-mass stars (below 0.6 $M_{\odot}$ ) due to dynamical evolution, while the younger Alpha Persei and Pleiades have retained more of their original population.

So, what does this all mean? The paper suggests that the Initial Mass Function in these open clusters is indeed different from the "universal" recipe often taught in textbooks. It appears to be deficient in the most massive stars compared to what is seen in the wider galaxy or in denser globular clusters. However, the authors are careful to note that this isn't a slam-dunk proof of a universal rule change. The low-mass part of the recipe is still a bit fuzzy because it depends heavily on the theoretical models used to convert star brightness into mass. Furthermore, the small number of clusters studied and the chaotic nature of star formation mean that some of these differences could just be random luck (stochastic scatter) rather than a fundamental environmental rule.

In short, the universe might be a bit more of a "custom baker" than a "factory line." While the middle of the star-formation recipe seems consistent, the top end (the heavy stars) might vary depending on where the stars are born. To be absolutely sure, we need to look at many more clusters, but this study has taken a big, playful step toward understanding why some star nurseries make more giants than others.

Technical Summary: The Initial Conditions and Initial Mass Functions of Alpha Persei, Pleiades and Praesepe

Problem Statement
The stellar Initial Mass Function (IMF) is a fundamental concept linking star formation to galactic evolution, yet its universality and dependence on environmental conditions remain debated. While a "canonical" IMF is often assumed, recent evidence suggests variations, particularly in the low-mass regime and across different stellar populations (e.g., globular clusters vs. field stars). Measuring the IMF in open clusters is complicated by dynamical evolution (mass segregation, tidal stripping) which alters the present-day mass function (PDMF), and by unresolved binary stars which bias photometric mass estimates. This paper addresses the need to reconstruct the initial conditions and IMFs of three nearby open clusters—Alpha Persei, Pleiades, and Praesepe—by rigorously correcting for these observational and dynamical effects.

Methodology
The authors employed a multi-stage approach combining high-precision astrometry, photometry, and advanced simulation techniques:

Membership Determination: Cluster members were identified using Gaia DR3 proper motions, parallaxes, and photometry, supplemented by UKIDSS (near-infrared) and HIPPARCOS (optical) data to ensure completeness down to stellar masses of 0.10–0.17 $M_\odot$ . A $\chi^2$ filtering technique was used to distinguish members from field stars, followed by photometric filtering to remove outliers.
Mass Estimation: Stellar masses were derived using three methods: PARSEC v2.0 and MIST isochrones, and an empirical mass-luminosity relation based on eclipsing binaries. The authors found that while high-mass estimates were consistent across models, low-mass estimates varied significantly, leading to the selection of PARSEC isochrones for the primary analysis.
Binary Correction: Using a Bayesian framework with Monte Carlo simulations, the authors corrected the observed PDMFs for unresolved binaries. They modeled the "stick-out" binary fraction (stars appearing brighter than the main sequence) and inferred the total binary fraction and the underlying system mass function.
Dynamical Evolution and Initial Conditions: To recover the initial IMF, the authors utilized N-body simulations (NBODY7) to model the dynamical evolution of the clusters. A machine learning emulator (trained on a grid of N-body simulations using Gaussian processes) was coupled with Hamiltonian Monte Carlo (HMC) algorithms. This allowed them to efficiently sample the posterior distribution of initial parameters (initial number of systems, binary fraction, half-mass radius, and IMF slopes) that best reproduce the current observed state of the clusters, assuming gas-free initial conditions.

Key Results

Binary Fractions: The present-day unresolved binary fractions were determined to be $(23.8 \pm 1.2)\%$ for Alpha Persei, $(20.0 \pm 0.8)\%$ for the Pleiades, and $(22.1 \pm 1.0)\%$ for Praesepe.
Initial Mass Function (IMF): The best-fitting initial mass function for all three clusters is described by a three-stage broken power-law distribution. The break masses are found in the ranges $0.24\text{--}0.50\,M_\odot$ and $0.91\text{--}1.20\,M_\odot$ .
IMF Slopes: The average IMF slopes derived are:
- Low-mass ( $\alpha_{med}$ ): $1.72 \pm 0.09$ (Note: The abstract labels the intermediate slope as $\alpha_{med}$ and high as $\alpha_{high}$ ; the text defines the low-mass slope as $\alpha_l$ , intermediate as $\alpha_m$ , and high as $\alpha_h$ . The abstract's $\alpha_{med}$ corresponds to the intermediate mass range $0.24\text{--}0.91\,M_\odot$ and $\alpha_{high}$ to $>0.91\,M_\odot$ . The summary values in the abstract are $\alpha_{med} = 1.72 \pm 0.09$ and $\alpha_{high} = 2.98 \pm 0.22$ ).
- High-mass ( $\alpha_{high}$ ): $2.98 \pm 0.22$ .
- Low-mass ( $\alpha_{low}$ ): $1.24 \pm 0.29$ (highly sensitive to the adopted mass-luminosity relation).
Cluster-to-Cluster Variation: There is evidence of scatter in the high-mass IMF between clusters ( $\sigma_{high} = 0.29 \pm 0.16$ ), whereas the intermediate-mass slope shows no significant variation ( $\sigma_{med} = 0.00 \pm 0.14$ ).
Dynamical Evolution: The analysis confirms that Praesepe has undergone significant dynamical evolution (losing $\approx 49\%$ of its initial mass, primarily low-mass stars), while Alpha Persei and the Pleiades have experienced minimal evolution ( $\approx 4\%$ and $9\%$ loss, respectively).

Significance and Claims
The paper claims to provide additional evidence for a "top-light" IMF in open clusters compared to the Salpeter slope ( $\alpha = 2.35$ ) and the canonical Kroupa IMF, indicating a relative deficiency of high-mass stars ( $>1\,M_\odot$ ) in these metal-rich, low-density environments. The results suggest that while the intermediate-mass regime may be universal, the high-mass IMF may vary with environmental conditions, potentially due to the lower gas surface densities in open cluster progenitors compared to denser star-forming regions.

The authors modestly conclude that while their findings point to environmental dependence, the small sample size and the inherent stochasticity of star formation and dynamical evolution make it difficult to definitively quantify these relationships. They assert that the derived initial conditions provide a robust starting point for future simulation work and that expanding the sample of well-characterized clusters is essential to distinguish between genuine environmental dependencies and stochastic scatter. The study also highlights that improved constraints on mass-luminosity relations at low masses are critical for reducing systematic uncertainties in low-mass IMF measurements.

More like this