Daily Summary for 2026-10-21

Introduction to the show: ident: Astrophysics Radio. The week's best astrophysics papers, unpacked for curious ears.

Vera: It's the twenty-first of October, twenty twenty-six, and this is the day's research.

Jocelyn: 24 new papers came out today.

Vera: Hi, I'm Vera.

Jocelyn: And I'm Jocelyn, here with Subrahmanyan.

Vera: We'll take the day in one pass, then pull out the papers we're staying with.

Jocelyn: Start us off with Towards the Habitable Worlds Observatory.

The summary: Vera: Today is the twenty-first of October twenty twenty six.

Jocelyn: We mapped tidal tails in open clusters using Gaia DR3 data.

Subrahmanyan: These features show how clusters evolve and interact with their galactic environment.

Vera: The primary focus was detecting these tails in five specific open clusters.

Jocelyn: Which ones were those five clusters?

Subrahmanyan: BH 164, Alessi 2, NGC 2281, NGC 2354 and M67.

Vera: So we focused on those specific objects today.

Vera: We used unsupervised machine learning for detection.

Jocelyn: And statistical techniques like color-magnitude diagram matching?

Subrahmanyan: Plus principal component analysis, PCA.

Vera: We also used density-based clustering to pinpoint tails.

Jocelyn: What did those analyses reveal about the clusters?

Subrahmanyan: Tidal tails exist in every cluster studied.

Vera: These tails stretch forty to one hundred parsecs.

Jocelyn: And how many stars are in those tails?

Subrahmanyan: They contain between one hundred and two hundred stars.

Vera: Further investigation examined kinematics, photometry, and spatial distribution of the detected features.

Jocelyn: The luminosity functions of these tails resembled those of their parent clusters.

Subrahmanyan: They generally lacked massive stars compared to the main bodies and showed higher binary fractions than expected.

Vera: Significant rotation was first detected in M67 and NGC 2281 through this analysis.

Jocelyn: That finding is important for understanding cluster evolution.

Subrahmanyan: Indeed, it suggests a dynamic process at play within those systems.

Vera: We need to look closely at the stellar kinematics next.

Jocelyn: The photometric data provides constraints on the physical extent of these features.

Subrahmanyan: It seems the distribution confirms tidal stripping is active there.

Vera: This study also found that the luminosity functions resembled those of their parent clusters.

Jocelyn: That comparison helps us gauge how much material was lost.

Subrahmanyan: They generally lacked massive stars compared to the main bodies in these tails.

Vera: Higher binary fractions were also exhibited than we anticipated for these structures.

Jocelyn: We should focus on the rotational signatures detected in M67 and NGC 2281.

Subrahmanyan: That significant rotation was a key discovery from this analysis.

Vera: It opens new avenues for modeling cluster dynamics based on this data.

Jocelyn: The results provide concrete evidence for these tidal interactions in action.

Subrahmanyan: Agreed, the kinematic and photometric data are very illuminating here.

Vera: Today's lucky papers are Tidal Tails and Their Dynamics in Open Clusters Using Gaia DR3 because it uses machine learning to find and study tidal tails in five open clusters using Gaia data.

Jocelyn: Significant rotation was also detected in M67 and NGC 2281 for the first time.

Subrahmanyan: This paper is worth hearing because it detects significant rotation in M67 and NGC 2281 for the first time.

Vera: Agreed, that detection of rotation is a major finding.

Jocelyn: That research also analyzes stellar properties like kinematics and binaries to find these tails.

Subrahmanyan: Indeed, analyzing those specific properties reveals the underlying dynamics clearly.

Lucky paper: 2508.00076: Vera: Alright, we're looking at the paper titled Towards the Habitable Worlds Observatory: Retrieval of Reflection Spectra from Evolving Earth Analogues using 1D CNNs.

Jocelyn: So, this work uses a one-dimensional convolutional neural network to retrieve key planetary parameters and molecular abundances from reflection spectra.

Subrahmanyan: It was trained on over one million synthetic, noise-injected spectra simulating Archean, Proterozoic, and Modern Earth analogs using NASA’s PSG2 model.

Vera: That training set had a specific filtering process where they rejected any spectrum with noise exceeding the 3sigma threshold at any wavelength.

Jocelyn: Out of over three point seven million computed samples, only one million eight hundred eighty six thousand exoplanets met all the physical and observational requirements for the final dataset.

Subrahmanyan: The model architecture uses residual convolutional blocks with two convolutions per block, each followed by batch normalization, the SiLU activation function, and a spatial dropout mask.

Vera: Then they followed that up with a multi-head self-attention layer to capture broader dependencies before branching into three heads for Physical Features, Main Chemical Features, and Other Chemical Features.

Jocelyn: And the retrieval performance showed that the model performs best where spectral signatures are prominent, like for CH4 and CO2 in Archean atmospheres.

Subrahmanyan: It also reported that planetary radius, gravity, surface pressure are recovered with low bias and dispersion-limited scatter.

Vera: What about uncertainty estimation when testing unseen data?

Jocelyn: They use Monte Carlo Dropout to transform the deterministic model into a stochastic ensemble by keeping those dropout layers active during inference.

Subrahmanyan: The analysis of the Monte Carlo Dropout posteriors showed that while test point fraction inside the predicted interval increases monotonically with sigma, coverage remains below the ideal Gaussian level, especially for temperature.

Vera: That’s a bit worrying regarding temperature uncertainty.

Jocelyn: But look at the credibility curves; for O3, every slice maintains a probability greater than zero point eight five across the full explored range of log-abundances.

Subrahmanyan: The paper also mentions that integrated gradients show the model bases its prediction on physically meaningful features, like the Fraunhofer A band for O2 and the Hartley–Huggins band for O3.

Vera: So, depending on whether it’s Archean or Modern Earth, the network correctly weights those specific bands differently.

Jocelyn: Under Proterozoic conditions, that LUVOIR-B retrieval path effectively nullifies the importance of O2 and O3, while HabEx/SS still assigns some weight to O2 in the UV.

Subrahmanyan: The study’s main limitation stems from simplifying assumptions about vertical atmospheric composition and a complete absence of clouds or hazes in their training set.

Vera: So, for future work, they need to incorporate vertical atmospheric gradients and the effects of multiple scattering by clouds and hazes.

Jocelyn: This CNN architecture establishes itself as a viable alternative to traditional Bayesian methods because it achieves inference in mere seconds for five thousand Monte Carlo Dropout samples on a single RTX two thousand sixty GPU.

Subrahmanyan: It’s capable of processing direct-imaging spectra with the Habitable Worlds Observatory on an operational cadence.

Lucky paper: 2509.03802: Vera: We're looking at Structure of Turbulent Radiative Mixing Layers today.

Jocelyn: This paper tackles those turbulent radiative mixing layers that are common in the interstellar and circumgalactic media.

Subrahmanyan: The authors show that a positive vertical compressive stress, Rzz, driven by radiative cooling, causes the TRML pressure to dip.

Vera: That positive Rzz stress is what establishes this universal steady-state structure for the TRMLs.

Jocelyn: So instead of just random turbulence, there's a specific physical mechanism sustaining this layer.

Subrahmanyan: Yes, it’s absent in incompressible turbulence, which is key to understanding the dynamics here.

Vera: They find that radiative losses are balanced by the divergences of enthalpy flux and turbulent heat flux Qz.

Jocelyn: What does that mean for how we think about energy transport through these layers?

Subrahmanyan: It means cooling is balanced by enthalpy flux divergence and turbulent heat flux, Qt, which is subdominant compared to the enthalpy term.

Vera: The volume-averaged temperature follows a tanh profile, resulting in universal emissivity distributions that are consistent across simulations.

Jocelyn: That universal structure sounds like something that should be common everywhere in these environments.

Subrahmanyan: It is, and they find the cooling-rate surface density saturates and becomes independent of the box size in the strong-cooling limit.

Vera: That independence from box size is a defining hallmark for these TRMLs, which is very important for modeling them.

Jocelyn: How does that translate into what this means for how we model multiphase gas dynamics in galactic environments?

Subrahmanyan: It enables numerical calibration of the entrainment rate as a function of key parameters of the setup.

Vera: They also define a fiducial cooling time t0 equal to tcool at T0 equal to pTcTh equals p0 times gamma minus one times n squared over twenty Lambda(T0).

Jocelyn: That definition sets a specific timescale for when the system settles into this steady state.

Subrahmanyan: After roughly one cooling time, the system ceases to grow and settles into a statistically steady turbulent radiative mixing layer.

Vera: In the TRML frame, the mean density profile is stationary because they enforce ∂t⟨ρ⟩ equals zero in that specific frame.

Jocelyn: So within that layer, what is happening with the vertical velocity?

Subrahmanyan: The net vertical mass flux, ⟨ρuz⟩, becomes uniform across the layer once steady state is established.

Vera: Within the TRML itself, rising dense plumes and sinking hot clumps generate a strong positive turbulent mass flux of delta rho times uz.

Jocelyn: That turbulence in mass flux seems to be sustained by the compressive Rzz stress we discussed earlier.

Subrahmanyan: Exactly, the vertical momentum equation reaches steady state when you look at ddz

⟨ρuz⟩⟨uz⟩ plus Rzz plus p: .

Vera: Locally, cooling is balanced by the divergence of enthalpy flux and that turbulent heat flux Qt.

Jocelyn: So we have a complete picture of how energy moves and cools across these interfaces now.

Subrahmanyan: The mean TRML temperature profile is well described by T(z) equal to Th minus Tc squared tanh z over z0 plus Th plus Tc squared, with z equals zero defined at the average of Th and Tc.

Vera: They also found that in quasi-isobaric TRMLs, the mass- and emissivityweighted PDFs follow PM(T) proportional to PV times T.

Jocelyn: That relationship between temperature and the probability distribution function is very specific.

Subrahmanyan: And PE(T) is proportional to PV times Lambda of T times T squared, which agrees with simulations in excellent agreement.

Vera: The implications for modeling are that shear-driven turbulence fundamentally changes the radiative cooling layers between dense and diffuse CGM phases.

Jocelyn: The pressure dip caused by the positive Rzz stress allows a steady, subsonic cooling flow where zero stress would not allow it.

Subrahmanyan: For sufficiently large shear-to-cooling time ratio, ξ equal to L perpendicular over delta ut0, the transport coefficients and the entrainment rate density sigma dot cool become independent of the box size.

Vera: That independence is what makes TRMLs universal building blocks for multiphase gas dynamics in galactic environments.

Jocelyn: It means we can use these results to calibrate how much material is being entrained into hotter regions based on setup parameters.

Subrahmanyan: The paper establishes this scaling relation between the dimensionless tidal deformability of neutron stars and a fundamental central equation of state parameter without needing a specific EOS model.

Vera: That connection to gravitational wave observations like GW170817 is what makes that scaling relation so important for probing neutron star core microphysics.

Lucky paper: 2604.26393: Vera: We're moving on to number three of five and this is Tidal Heating of Stellar Clusters in Fuzzy Dark Matter Halos.

Jocelyn: This paper looks at how ultra-light dark matter behaves in dwarf galaxies.

Subrahmanyan: It shows that when the de Broglie wavelength is much larger than the stellar cluster size, tidal heating becomes the main mechanism.

Vera: That transition from diffusive to tidal heating is really interesting because it changes everything about how we calculate heating efficiency at low particle masses.

Jocelyn: The paper explains that this happens when you have a reduced soliton mass and a tidally stripped halo.

Subrahmanyan: Exactly, the growth rate kR is proportional to the particle mass ma, specifically kR ∼ one / ma.

Vera: So if we want to constrain the dark matter particle mass from cluster heating, we have to carefully consider the structure and environment of the FDM halo.

Jocelyn: The paper uses a wave-Schwarzschild method instead of expensive Schrödinger-Poisson simulations to build these halo structures.

Subrahmanyan: They construct halos as linear combinations of stationary eigenmodes of the Schrödinger equation, which creates a density profile by superposing an NFW profile with a central soliton core.

Vera: The total density is split into two parts: ρ = ρbase plus ρper, where the oscillatory interference terms are denoted by δρ/ρ ∼ O(one).

Jocelyn: That means the heating depends on those oscillatory terms, which come from the wave interference.

Subrahmanyan: And they found that tidal stripping of the FDM halo strongly suppresses this heating mechanism.

Vera: They quantified this suppression by looking at principal quantum number nprincipal equal to one where heating is completely suppressed and the cluster size remains nearly constant.

Jocelyn: This result is important because it shows that the net heating of a UFD therefore depends sensitively on its orbital history and the degree of tidal stripping.

Subrahmanyan: The paper concludes that deriving a universal, one-parameter lower bound on ma might be too simplistic because of this dependence on orbital history.

Vera: So, to summarize Tidal Heating of Stellar Clusters in Fuzzy Dark Matter Halos, it's about showing that the heating mechanism shifts based on scale comparison and environment.

Jocelyn: It challenges previous constraints derived from simpler diffusion approximations by highlighting the tidal stripping effect.

Subrahmanyan: The key implication is that we need to model individual UFDs with their orbits and tidal histories to get accurate predictions, rather than just a single mass bound.

Lucky paper: 2601.12315: Vera: So we’re looking at a paper called The MAGPI Survey: co-evolution of baryons and dark matter in star-forming disk-like galaxies at zero point one z zero point eight five today.

Jocelyn: It uses spatially resolved kinematic data to study how dark matter content relates to the baryon content in these disk-like systems.

Subrahmanyan: The authors use three deeBarolo to model gas kinematics, getting moment maps, surface brightness profiles, rotation velocity profiles and velocity dispersion profiles.

Vera: They correct the observed rotation velocity profiles for inclination and gas pressure support to get intrinsic rotation curves.

Jocelyn: Then they decompose the total dynamical mass profile into stellar disk MDisk(R), bulge Mbulge(R), atomic gas MHI(R), molecular gas MH2(R) and dark matter MDM(R).

Subrahmanyan: They estimate the dark matter fraction fDM within both the effective radius Re and the optical radius Ropt.

Vera: The selection process was pretty strict, requiring confirmed spectroscopic redshift and detection of Halpha or OIII.

Jocelyn: And inclination angles had to be between twenty-five degrees and seventy-five degrees with a signal-to-noise ratio greater than three.

Subrahmanyan: The final kinematic sample consists of three hundred ninety-seven galaxies, which they call UIDs.

Vera: They compare the circular velocities within Re and Ropt to derive characteristic circular velocity Vc for each galaxy.

Jocelyn: They then calculate fDM(R) by comparing the total dynamical mass to the baryonic mass budget within a given radius.

Subrahmanyan: The key finding is that low-mass galaxies with Mstar less than one hundred nine point five Msun are dark matter dominated with an average fDM of about zero point eight five in both Re and Ropt.

Vera: That means they are almost entirely dark matter in their inner and outer regions compared to what we might expect.

Jocelyn: In high-mass galaxies, the DM fraction is low within the inner regions when compared to the outskirts of those same systems.

Subrahmanyan: Specifically, at fixed circular velocity, low-mass galaxies exhibit about forty percent higher DM fraction than their high-mass counterparts within Re.

Vera: That points toward an inside-out disk growth scenario for these galaxies as they evolve.

Jocelyn: And the analysis also highlights an inverse correlation between fDM and baryon surface density when looking at both regions.

Subrahmanyan: Galaxies with a central baryon surface density sigma bar greater than one hundred eight point zero Msun kpc-two cannot have a lower total observed baryonic mass in that region.

Vera: In contrast, systems with low baryon surface densities, sigma bar less than one hundred eight point zero Msun kpc-two show different behavior entirely.

Jocelyn: This work also touches on testing the isothermal Jeans model for self-interacting dark matter halos during their deep collapse phase.

Subrahmanyan: That model provides a computationally efficient description of how dark matter halos evolve during collapse when applied to gravitational lensing observables like projected mass and logarithmic density slope.

Vera: They show this model tracks simulated evolution more closely than parametric approaches in the deep collapse regime for those observables.

Jocelyn: It’s important because it links structure formation directly to observable gravitational lensing data.

Subrahmanyan: The MAGPI Survey analysis also provides empirical constraints on how dark matter content relates to galaxy structure at intermediate redshifts.

Vera: This is supported by the paper's conclusion that the structural correlations in galaxies were established by redshift zero point, as seen in a tight inverse correlation between dark matter fraction and baryon mass surface density.

Jocelyn: That inverse correlation indicates an inside-out assembly scenario for these disk-like galaxies at zero point one z to zero point eight five z.

Subrahmanyan: The multi-tracer technique, combining photometric surveys with HI intensity mapping, improves constraints on magnification biases by factors ranging from eighteen to fifty depending on which surveys are used.

Vera: That improvement allows for high-precision measurements of both the magnification bias parameters and the amplitude of the Weyl potential.

Jocelyn: This opens a new pathway to test general relativity on cosmological scales when you combine those survey types.

Subrahmanyan: The MAGPI Survey is crucial because it shows that for star-forming disk galaxies, dark matter content is strongly mass-dependent, with low-mass systems being the most dark matter dominated.

Lucky paper: 2603.28840: Vera: Let's talk about From photometric surveys to HI intensity mapping improving constraints on magnification biases while testing gravity because it tackles how we can test general relativity on cosmological scales.

Jocelyn: The core idea is combining photometric galaxy surveys with HI intensity mapping surveys for high-precision measurements of the magnification bias sG(z) and the gravity probe β.

Subrahmanyan: This combination dramatically improves constraints on β by factors of twenty-five to fifty depending on which surveys you use.

Vera: And for sG(z), the improvements can be marginal, maybe by a factor of two to eight.

Jocelyn: The paper uses statistical estimators like the Angular Power Spectrum Cl and the Two-point Angular Correlation Function w(θ) to quantify projected matter distribution.

Subrahmanyan: The connection between them is given by Peebles one thousand nine hundred eighty where w(θ) equals Xlmaxl≥zero 2l + one 4π Pl(cos θ)C WXWX′l i j!, which is a Legendre polynomial relation.

Vera: They test general relativity by constraining the sum of gravitational potentials Φ + Ψ through Equation three which is a key observable for testing gravity on cosmological scales.

Jocelyn: The most dramatic improvement is seen for the parameter β, which probes modifications to general relativity.

Subrahmanyan: The best performance is achieved with Euclid-like combined with SKAO reaching a relative uncertainty of ∼ zero point zero five percent.

Vera: They use three photometric galaxy surveys, a DES-like, an LSST-like, and an Euclid-like survey.

Jocelyn: And they consider two HI intensity mapping surveys called MeerKLASS and SKAO for the H I intensity mapping component.

Subrahmanyan: The Hi IM differs from galaxy surveys because it measures the collective 21cm emission from unresolved neutral hydrogen instead of individual sources.

Vera: They use the Fisher matrix formalism to estimate future constraints on these magnification biases and β.

Jocelyn: A key caveat is that their analysis assumes Gaussian covariance and doesn't account for non-linear structure formation at small scales.

Subrahmanyan: They also neglected systematic effects like photometric redshift uncertainties beyond a simple scatter model, intrinsic alignment of galaxies, and foreground contamination in the Hi intensity mapping.

Vera: The effectiveness of the multi-tracer method really depends critically on overlapping sky area; as noted with DES-like and SKAO, smaller overlapping fsky can partially offset gains from superior instrument specifications.

Jocelyn: A natural extension would be to apply Markov Chain Monte Carlo methods to actual or mock data, directly measuring sG(z) and β.

Subrahmanyan: Looking further ahead, the combination of photometric surveys like Euclid with next-generation radio arrays like SKAO offers an unprecedented opportunity to test general relativity on cosmological scales with percent-level precision.