PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation

arXiv:2608.10409 · astro-ph.EP, astro-ph.SR · Submitted 2026-08-11 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio.

Vera: Next we'll be talking about the paper "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation".

Jocelyn: The paper was written by I. Mosqueira from San José State University.

Vera: Stay tuned as we take you through the paper and discuss its implications.

Title and Authors: Vera: We're opening our discussion of a new paper, "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation," by I. Mosqueira at San José State University. Even the title signals something unusual, because it pairs a directly observed planet with a distant companion and asks what both can tell us about how moons are made. And right away, the author is working with a sample of just two objects, so every measurement has to carry a lot of weight.

Jocelyn: For listeners meeting these objects for the first time, PDS 70 is a young star about 112 parsecs away, famous for two directly imaged planets still embedded in their birth disk. PDS 70 c is the outer one, and it shows a compact source of millimeter light around it — a circumplanetary disk. SR 12 c is a wide companion, something like a thousand astronomical units from its host, and it also has a detected cold disk.

Vera: Just two objects with secure cold submillimeter detections around planetary-mass bodies — those are the only two we know of. The paper is working with an extremely tight dataset, and that makes every modeling assumption visible. It also means the two detections have to be argued about with real care, because there is no third example to lean on.

Subrahmanyan: And the author's history matters here. Mosqueira has spent years developing a specific model for the formation of the regular satellites of Jupiter and Saturn — a quiet, solids-enhanced circumplanetary disk. This paper asks whether the first two observed circumplanetary disks are consistent with that framework, and the answer it reaches is yes, provided you account for the fact that both PDS 70 planets sit in a shared gap. That shared-gap geometry turns out to be the hinge for everything.

Jocelyn: The central parallel is that PDS 70's two giants look like a young Jupiter–Saturn pair, with both planets controlling one common radial region of the parent disk. And SR 12 c separates two questions that are easy to conflate: how long it takes to finish building the planet itself, and how long a disk around it can survive to build moons. That separation is what makes the second object more than just another disk detection.

Vera: The timescale numbers are genuinely striking — SR 12 c's mass-growth timescale comes out to about two billion years, which means it has essentially stopped growing. Next we should walk through the paper's summary of its findings, because that is where the dust mass, the system age, and the moon-formation chronology all come together.

The Paper's Summary: Vera: We have introduced "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation" and its author, and I just teased the SR 12 c timescale. Now let's go through the paper's summary of its own findings, because the abstract packs in the key numbers. It is dense, but every number earns its place.

Jocelyn: The summary states that PDS 70 c and SR 12 c are the only two bound planetary-mass objects with secure cold submillimeter disk detections. The headline measurement is the 855-micron flux of PDS 70 c. In the optically thin limit, that flux implies 0 point 007 to 0 point 031 Earth masses of dust, assuming a temperature of 26 Kelvin.

Vera: Callisto alone is 0 point 018 Earth masses. So the radiating grains we see around PDS 70 c already add up to something like a full-sized regular satellite. And if the emission is optically thick instead, the flux still demands a minimum emitting area — a coplanar disk at least 0 point 58 to 0 point 66 astronomical units across, depending on the temperature you choose.

Subrahmanyan: The second object provides the evolutionary measurement. SR 12 c's accretion rate gives a mass-growth timescale of 1 point 9 billion years, with an uncertainty of half a billion. Over another million years, it would add barely five hundredths of a percent to its mass. The companion is essentially finished growing, and yet gas and solids still surround it — that is the cleanest separation we have between planetary assembly and the survival of a moon-forming reservoir.

Jocelyn: Then the paper connects this to the satellite-formation model, where gas drag clearing of the building blocks gives about a million years to make Callisto and about ten million years to make Iapetus. PDS 70's age, 5 point 4 million years with a one-million-year uncertainty, sits between those two values. So the system lands exactly in the window where the inner moon reservoir would be processed and the outer one would still be active.

Vera: Which means we could be looking at a system caught mid-process — the inner reservoir already cleared, the outer one still supplying material. That is exactly the kind of snapshot you would want if you were designing an observation to test theories of satellite formation. And the direction of the argument runs from the observations to the model, not the other way around.

Subrahmanyan: The paper is careful about the epistemic status of that claim. The dust mass is a measurement with assumptions about opacity and temperature, and the formation timescales are outputs of a specific model. The match between the system age and those timescales is an interpretation, but it is one that lands precisely where the model predicts.

Jocelyn: One more layer in the summary: the observed disk scale. The paper reports that the continuum source is smaller than about 1 point 2 astronomical units, yet much larger than the compact size you would expect if material fell in before the planets carved their gap. That radial scale is the clue that leads into the angular-momentum machinery, which is what the paper improves and what we should turn to next.

Improvements Suggested by the Paper: Vera: We have covered the summary of "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation" — the dust masses, the SR 12 c timescale, and the 5 point 4-million-year age landing between Callisto and Iapetus. Now for the improvements the paper brings to the modeling side, starting with how it treats the two planets' shared gap. These are the parts of the paper a quick read might skip, but they carry the argument.

Jocelyn: The first improvement is treating the common gap as a finite reservoir. The two PDS 70 planets torque the surrounding gas, pushing some of it inward and some outward, and that depletes the reservoir over time. With the paper's adopted closure, the e-folding depletion time for the PDS 70 architecture is a few thousand years, so the supply feeding the circumplanetary disks is a declining inflow, not a steady one.

Subrahmanyan: The second improvement is the angular-momentum bookkeeping, and this is the elegant part. The size of a circumplanetary disk is set by the specific angular momentum of the material entering the Hill sphere. The paper writes that as a dimensionless parameter, lambda, and shows that the circularization radius is lambda squared over three times the Hill radius — so the disk scale depends quadratically on the inflow's angular momentum, not on the Hill radius alone.

Vera: Before a gap forms, the standard local estimate gives lambda equal to a quarter, so the disk would be about a fiftieth of the Hill radius — very compact. But once the planets open a gap, gas enters through the L1 and L2 regions with low relative velocity, lambda jumps to about seven eighths, and the disk scale lands near a quarter to a third of the Hill radius. For PDS 70 c, that means roughly one to two astronomical units — right where the observations put it.

Jocelyn: So the disk's physical size is itself a fingerprint of its formation history. A compact disk would suggest early, local infall, while the observed extended disk indicates late-stage, gap-fed delivery. The paper even runs a planar ensemble of ballistic trajectories to map the distribution of circularization radii, and the flux-weighted mean comes to about 1 point 3 astronomical units for the fiducial planet mass.

Subrahmanyan: And there is a methodological improvement that deserves its own mention. The paper includes an appendix documenting a failure mode in eye-assisted scientific reasoning: when the author asked an assistant to summarize the evidence on disk turbulence, it repeatedly treated turbulence as the default and demanded that the observations rule it out. The paper argues that the settling measurements should invert that burden — quiescence is the empirical prior, not the exception.

Vera: That connects to the paper's whole attitude: the observations lead, and the models follow. We have the abstract and the improvements; next we should go back to the opening pages, where the system parameters and the quiescent-disk evidence are set out in detail. That is where the physical stage gets built.

The First Page: Vera: We have discussed the summary of "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation" and the improvements it makes to the modeling framework. Now let's look at the first page, where the paper establishes the physical stage and the observational case for a quiet disk environment. This is where the system architecture and the evidence for weak turbulence come together.

Jocelyn: The introduction gives the architecture: PDS 70 b orbits near 22 astronomical units, PDS 70 c near 34 point 5, both inside a large dust-depleted cavity, and both detected in H-alpha emission — so both are actively pulling in gas. The star is 0 point 76 solar masses, the distance is 112 point 4 parsecs, and the system age is 5 point 4 million years. The paper calls it a Jupiter–Saturn analog in the most relevant sense: two neighboring giant planets with their circumplanetary environments embedded in one shared gap.

Subrahmanyan: With those parameters, the Hill radius of PDS 70 c comes to roughly four to six astronomical units, depending on the planet mass, and the observed continuum source is smaller than about 1 point 2 astronomical units. It sits well inside the Hill sphere. So the millimeter emission is genuinely tied to the planet, not to the circumstellar ring, which is the foundation of the entire argument.

Vera: The first page also lays out the evidence for quiescence in planet-forming disks more broadly. The paper cites the thin dust layers inferred in HL Tau, the dramatic settling seen in edge-on disks like Oph 163131, and a uniform analysis of thirty-three disks where the inferred stirring parameter alpha is typically below ten to the minus three. Molecular-line surveys put upper limits on nonthermal broadening at a fraction of the sound speed — the empirical picture is a quiet disk, not a churning one.

Jocelyn: And on the theoretical side, the first page recalls the Balbus–Hawley result: in a Keplerian shear flow, transient disturbances decay through epicyclic motions, so turbulence is not self-sustaining without a driving mechanism. That matters because a quiescent, solids-enhanced circumplanetary disk is precisely the environment the author's satellite-formation model requires. Strong turbulence would stir the building blocks, prevent settling, and make it hard to assemble moons on the million-to-ten-million-year schedule.

Subrahmanyan: So the first page is establishing the stage: a quiet parent disk, two giants sharing a gap, and a compact reservoir of solids around the outer planet. And the introduction states the central claim we have been circling all episode: the planets' torques deplete the shared reservoir, limit the final masses, and eventually terminate circumplanetary delivery, while that same delivery sets the compositional budget available for moons. Everything else in the paper is the working out of that claim.

Vera: That is the thread we can pull together in our conclusion — what this paper leaves us with, and which open questions it hands to the next round of observations. We will also say our goodbyes to the paper and get ready for the next one on our list.

Conclusion: Vera: We have spent the episode with "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation" by I. Mosqueira, and it is a paper that connects very faint, very small sources of millimeter light to an old question: how did the regular satellites of Jupiter and Saturn form? It pulls observations of two objects into a single physical story.

Jocelyn: What it comes down to is this. PDS 70 c hosts a compact circumplanetary disk with a dust mass comparable to Callisto, on a scale that only makes sense if material was delivered through a developed planetary gap. SR 12 c shows that a planetary-mass companion can be essentially finished growing while gas and solids still remain around it. And the age of the PDS 70 system, 5 point 4 million years, falls right between the modeled formation times of Callisto and Iapetus — so we may be seeing a system in the middle of building moons.

Subrahmanyan: The paper also models the delivery process itself: the shared gap as a finite reservoir, the angular momentum of the inflow setting the disk size, and the transient nature of any inflow-driven stirring. And it makes a methodological statement by documenting how an eye assistant's inherited assumptions about turbulence needed to be corrected against the observations. Those contributions will outlast the specific detections.

Vera: The open questions remain. There is still no direct measurement of the gas in a circumplanetary disk — no surface density, no gas-to-dust ratio for PDS 70 c. The retained fraction of incoming material, and how much recycles out of the Hill sphere, will require three-dimensional simulations that have not been done yet. This paper is a framework for those next steps rather than a final answer.

Jocelyn: For all that, it is a remarkable convergence: two detections, two objects, and a satellite-formation model from 2001 all lining up into one coherent picture. We will be watching the follow-up observations closely. And we are ready to move on to the next paper.

Vera: That closes our look at "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation." Thank you to our listeners, and to Subrahmanyan for joining the discussion. Stay with us for the next paper.

author1, author2

University1 · Company2

astro-ph.EP, astro-ph.SR

Submitted: 2026-08-11

Updated: 2026-08-25

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 100/100

The gist: I apologize, but you have provided a list of references and citations, but not the actual text of the scientific paper titled "PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and

Key concepts

Circumplanetary disk
A disk of gas and dust orbiting a planet, from which moons can form. The episode focuses on two such disks detected around PDS 70 c and SR 12 c, which are rare examples that provide direct observational constraints on satellite formation.
Hill radius
The region around a planet where its gravity dominates over the star's. The size of a circumplanetary disk is related to the Hill radius, and the paper shows that the disk's observed scale indicates material entered through a gap, not early infall.
Mass-growth timescale
The time it would take for a planet to double its mass at its current accretion rate. For SR 12 c, this is about 1.9 billion years, meaning it has essentially stopped growing, yet still retains a disk—separating planet formation from moon formation timescales.

Terminology

Summary

I apologize, but you have provided a list of references and citations, but not the actual text of the scientific paper titled PDS 70 c and SR 12 c: Observational Constraints on Giant-Planet and Satellite Formation.

As a diligent researcher where mistakes could cost millions of dollars, I must have the source material in front of me to accurately extract, quote, and summarize its contents.

Please provide the full text of the arXiv paper, and I will immediately generate a long, detailed summary using only quoted material from that document.

Improvements for AI systems

The scientific literature provided is deeply rooted in the computational modeling of protoplanetary disks, accretion physics, and orbital dynamics. To improve general-purpose AI systems using this specialized domain knowledge, we must move beyond mere data correlation and enforce physical laws directly into the model architecture.

Here are the specific improvements:

Improvement: Develop constrained neural network layers that use fundamental conservation laws (e.g., conservation of angular momentum, energy dissipation rates, Navier-Stokes equations for fluid flow) as part of the loss function (L total = L data + lambda times L physics).

What the Improved AI System Can Do:

  • Constrained Simulation: The AI will not just predict what happens, but why it happens according to known physics. It can simulate complex phenomena like Type I/II planetary migration or disk gap opening while ensuring the resulting density profiles and velocity fields adhere to established orbital mechanics principles.

  • Parameter Inference: Given incomplete or noisy observational data (e.g., sparse ALMA measurements), the AI can robustly infer missing physical parameters (like the viscosity coefficient alpha or local stellar magnetic field strength) because its predictions are physically constrained, drastically reducing model degeneracy.

Improvement: Create specialized Transformer or Recurrent Neural Network (RNN) architectures designed for modeling multi-scale, non-linear time series data characteristic of astrophysical evolution (e.g., disk structures changing over millions of years). This requires integrating variable time steps and resolution switching mechanisms.

What the Improved AI System Can Do:

  • Predicting Evolutionary States: The system can accurately forecast the evolution of a protoplanetary disk across vastly different timescales—from rapid accretion events (years) to slow structural changes (millions of years). For example, it could predict when and where a gas giant will fully clear a gap in the disk based on current observed accretion rates.

  • Identifying Feedback Loops: It can model complex feedback mechanisms, such as how the tidal interaction with a forming planet influences the disk's thermal structure, which in turn affects the rate of dust settling and planetesimal formation.

Improvement: Build dedicated Convolutional Neural Network (CNN) pipelines optimized for processing spectroscopic data (e.g., emission lines from molecular species like CO or H 2). These modules must be trained to differentiate between physical sources of spectral features: thermal excitation vs. kinematic motion vs. chemical fractionation.

What the Improved AI System Can Do:

  • Automated Chemical Mapping: The AI can analyze complex spectra from multiple wavelengths simultaneously to map the chemical abundance gradients within a disk (e.g., detecting evidence of snow lines or ice condensation fronts), providing detailed insights into planetesimal formation pathways that are often ambiguous in raw data.

  • Decomposition of Sources: It can decompose observed signals into distinct physical components—for instance, separating the spectral signature of the central star's outflow from the emission generated by a ring structure far out in the disk, even when sources overlap spectrally.

Abstract

PDS 70 c and SR 12 c are the only bound planetary-mass objects with secure cold submillimeter disk detections. Together they constrain giant-planet growth and satellite formation. The PDS 70 planets exhibit remarkable parallels to the Jupiter--Saturn pair in our Solar System. Both PDS 70 planets accrete within one shared gap, which links their final masses, the material reaching each Hill sphere, and the properties of the circumplanetary disk. Planetary torques deplete this finite reservoir, causing circumplanetary supply to decline as the protoplanets open a circumstellar gap. SR 12 c separates the planetary-growth and satellite-formation timescales: gas and solids survive even though its current mass-growth timescale is (1.9 plus or minus0.5) times10 9 yr. For PDS 70 c, the 855- mu m flux implies 0.007 -- 0.031, of dust at 26 K in the optically thin limit, while the optically thick limit requires a minimum coplanar radius of 0.58 -- 0.66 au, depending on temperature. This scale is compatible with late-stage gas inflow through a well-formed gap with specific angular momentum, r c about /3. These observations are consistent with our satellite formation model for Jupiter and Saturn (Mosqueira & Estrada 2003a,b, submitted in 2001), in which gas-drag clearing of satellitesimals gives formation timescales of about10 6 yr for Callisto and about10 7 yr for Iapetus. The PDS 70 age of 5.4 plus or minus1.0 Myr lies between these values. These constraints provide strong support for a quiescent, solids-enhanced satellite-forming environment, coupled in the early stages to planetary-gap evolution.

Sources

Related papers