🔭 astrophysics

X-Shooter survey of disk accretion in Upper Scorpius II. A lack of correlation between accretion rates and disk properties

This study of 127 Upper Scorpius stars with protoplanetary discs reveals a lack of correlation between mass accretion rates and disc properties, suggesting that the evolutionary links observed in younger regions fade with age due to the decoupling of inner and outer discs.

Original authors: A. Empey, C. F. Manara, R. Garcia Lopez, A. Natta, R. Claes, F. Zagaria, J. M. Alcalá, R. Anania, G. Beccari, J. Carpenter, S. Facchini, D. Fedele, G. Lodato, K. Mauco, A. Miotello, B. Nisini, I. Pasc

Published 2026-06-19

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: A. Empey, C. F. Manara, R. Garcia Lopez, A. Natta, R. Claes, F. Zagaria, J. M. Alcalá, R. Anania, G. Beccari, J. Carpenter, S. Facchini, D. Fedele, G. Lodato, K. Mauco, A. Miotello, B. Nisini, I. Pascucci, L. Piscarreta, G. Rosotti, A. Scholz, L. Testi, M. Vioque

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 a giant cosmic nursery where baby stars are born, surrounded by swirling disks of gas and dust. These disks are the "construction sites" where planets are built. For a long time, astronomers believed these construction sites followed a predictable rule: the bigger the pile of building materials (the disk), the faster the baby star would gulp down fuel (accretion) to grow. It was like thinking a bigger construction site always meant a faster construction crew.

This new study, led by A. Empey and colleagues, decided to check if this rule still holds true for "teenage" stars in a region called Upper Scorpius. While younger star nurseries are about 1 to 3 million years old, Upper Scorpius is older, roughly 5 to 10 million years. The researchers wanted to see if the relationship between the disk size and the feeding rate had changed as the stars got older.

Here is what they found, explained simply:

The "Teenage Rebellion" of Star Disks
The team looked at 127 stars in this older region. They used a powerful telescope instrument called X-Shooter to measure how much material each star was eating (accretion) and combined this with data from the ALMA radio telescope to measure the size of the dust disks around them.

In younger star regions, there is a clear link: big disks usually mean fast feeding. But in Upper Scorpius, that link completely broke down.

The Analogy: Imagine a classroom of 127 students. In a younger class, you might notice that students with bigger backpacks (more dust) are also the ones running the fastest (accreting faster). But in this older class, you see a chaotic mix. Some students have huge backpacks but are walking very slowly. Others have tiny backpacks but are sprinting. There is no pattern. The "big backpack = fast runner" rule has vanished.

The "Silent" Half
About half of the stars in their sample were so quiet that the team couldn't even tell if they were still eating at all. Their feeding rates were so low that they were indistinguishable from the natural "hiccups" or activity of the stars themselves.

The Analogy: It's like trying to hear a whisper in a noisy room. For 50% of the stars, the "whisper" of them eating gas was either too quiet to hear or completely gone. This suggests that for many of these older stars, the inner part of their construction site has already been cleared out, even if the outer dust is still there.

Why the Confusion?
The researchers checked if this chaos was caused by specific groups of stars, stars with binary partners (two stars orbiting each other), or special types of disks with holes in the middle. They found that none of these factors could explain the mess. The lack of connection between the disk size and the feeding rate is a general feature of this older region.

The Big Picture
The study concludes that as these star systems age, the inner part of the disk (where the star eats) and the outer part (where the dust sits) seem to stop talking to each other. They become "decoupled."

The Analogy: Think of a river. In a young river, the water flow (accretion) is directly tied to the size of the riverbed (disk). But in this older region, it's as if the river has split. The water flowing past the star has no obvious connection to how much mud is sitting in the riverbed further downstream. The smooth, predictable flow of the past has turned into a scattered, unpredictable pattern.

What This Means for Theory
Current theories about how stars and disks evolve (like the "viscous" theory, which assumes a smooth, slow draining of the disk) predicted that the relationship between disk size and feeding rate should get tighter and clearer as stars age. This study shows the exact opposite: the relationship gets looser and disappears.

The authors suggest that the current "rulebooks" for how these cosmic nurseries work need a rewrite. The simple models don't explain why some old disks are still feeding fast while others have stopped, or why the size of the dust pile no longer predicts the feeding speed. It seems that by the time these stars reach their "teenage" years, the process of building planets has become much more complex and less uniform than previously thought.

Technical Summary: X-Shooter survey of disk accretion in Upper Scorpius II. A lack of correlation between accretion rates and disk properties

Problem Statement
The evolution of protoplanetary discs is intrinsically linked to planet formation, yet the fundamental mechanisms driving disc dispersal remain debated. Two primary theoretical frameworks exist: viscous evolution, where angular momentum is transported via turbulence, and magnetohydrodynamic (MHD) wind-driven evolution, where angular momentum is removed via winds. While population synthesis models suggest these mechanisms produce different dispersions in accretion-disk relations over time, observational constraints are lacking for older star-forming regions (ages > 5 Myr). Most existing demographic studies focus on younger regions (1–3 Myr), leaving a gap in understanding how accretion rates ( $\dot{M}_{acc}$ ) correlate with stellar and disc properties as systems age. Upper Scorpius (USco), with an estimated age of 5–10 Myr, represents a critical testbed for these models, but previous surveys of the region have been limited by small sample sizes or a lack of direct accretion tracers.

Methodology
This study presents a spectroscopic analysis of a large sample of 127 stars with protoplanetary discs in Upper Scorpius. The sample was constructed by combining 36 targets from a previous pilot study (Manara et al. 2020) with 91 new targets selected from recent Gaia-based membership lists and ALMA continuum detections.

Observations: High-resolution optical spectra were obtained using the X-Shooter spectrograph on the ESO Very Large Telescope (VLT). Observations utilized both narrow and wide slit configurations to correct for slit losses and maximize signal-to-noise (S/N).
Data Analysis: Stellar and accretion properties were derived using the FRAPPE code, which self-consistently fits the UV continuum excess (accretion shock) against the photospheric and chromospheric emission. The code utilizes a grid of Class III templates and isothermal hydrogen slab models.
Upper Limit Determination: A novel method was applied to evaluate upper limits for accretion luminosity ( $L_{acc}$ ). Objects were classified as "definite accretors" or "upper limit accretors" based on the S/N of the Balmer continuum and whether the measured $L_{acc}$ exceeded the statistical estimate of chromospheric emission for their spectral type.
Disc Properties: Disc dust masses ( $M_{disc}$ ) were derived from ALMA Band 7 continuum flux densities (Carpenter et al. 2025), assuming optically thin emission and a standard gas-to-dust ratio. CO disc radii ( $R_{CO}$ ) were incorporated from recent ALMA data (Zagaria et al., in prep).
Statistical Analysis: Power-law relations between $\dot{M}_{acc}$ and stellar mass ( $M_\star$ ), stellar luminosity ( $L_\star$ ), disc mass ( $M_{disc}$ ), and disc radius ( $R_{CO}$ ) were fitted using the linmix Python package, which employs a hierarchical Bayesian approach to handle upper limits and measurement errors.

Key Contributions

Expanded Sample: This work triples the size of the spectroscopically characterized sample of disc-bearing stars in Upper Scorpius, providing a statistically robust dataset of 121 targets with derived stellar and accretion parameters.
Systematic Upper Limits: The study introduces a rigorous criterion for distinguishing between definite accretors and upper limits, revealing that approximately 50% of the sample falls into the upper limit category due to low S/N or low accretion rates indistinguishable from chromospheric activity.
Comprehensive Correlation Analysis: It provides the first large-scale statistical assessment of the $\dot{M}_{acc}$ vs. $M_{disc}$ and $\dot{M}_{acc}$ vs. $R_{CO}$ relations in a 5–10 Myr old region, explicitly testing the persistence of correlations observed in younger regions.

Results

Accretion-Stellar Relations: A weak positive correlation was found between accretion luminosity and stellar luminosity ( $L_{acc} \propto L_\star^{2.1 \pm 0.4}$ , $\rho \approx 0.5$ ) and between mass accretion rate and stellar mass ( $\dot{M}_{acc} \propto M_\star^{2.4 \pm 0.5}$ , $\rho \approx 0.5$ ). However, the dispersion in these relations is significantly larger ( $\sigma \approx 1.3-1.4$ dex) than in younger regions like Lupus and Chamaeleon I.
Accretion-Disc Relations: Crucially, no significant correlation was observed between the mass accretion rate and disc dust mass ( $\dot{M}_{acc}$ vs. $M_{disc}$ , $\rho \approx 0.4$ ) or gas disc radius ( $\dot{M}_{acc}$ vs. $R_{CO}$ , $\rho \approx 0.0$ ). The scatter in $\dot{M}_{acc}$ at a given disc mass spans approximately three to four orders of magnitude.
Subgroup and Structural Effects: The large dispersion in accretion rates could not be explained by membership in specific USco subgroups, nor by the presence of binary systems or transition discs. While transition discs showed high disc masses, their accretion rates remained within the broad scatter of the general population.
Comparison with Younger Regions: While the slopes of the power-law fits for $\dot{M}_{acc}$ vs. $M_\star$ are similar to those in younger regions, the correlation strength ( $\rho$ ) is significantly weaker in USco. The dispersion in USco is larger than in younger regions, suggesting a "fading" of the tight correlations seen in early disc evolution.

Significance and Conclusions
The paper concludes that the lack of correlation between accretion rates and disc properties, combined with the large dispersion in $\dot{M}_{acc}$ , challenges current evolutionary models.

Decoupling of Disc Regions: The results suggest a decoupling between the inner disc (probed by accretion) and the outer disc (probed by dust mass/radius) by the age of Upper Scorpius. The amount of material in the outer disc no longer dictates the accretion rate onto the star.
Model Constraints: Pure viscous evolution models, which predict that scatter should decrease over time as systems converge to isochrones, fail to explain the observed large dispersion. Wind-driven models (MHD) reproduce the scatter better but still struggle to fully account for the lack of correlation and the specific evolution of these relations with age.
Future Directions: The authors assert that current theoretical frameworks require development to explain the observed properties. They emphasize the need for further modeling that incorporates external photoevaporation and dust evolution, as well as more complete gas property measurements, to disentangle the mechanisms driving disc dispersal in older systems.

The study underscores that Upper Scorpius is not merely an "older version" of younger regions like Lupus but exhibits distinct evolutionary characteristics that necessitate a re-evaluation of standard disc evolution paradigms.

More like this