The StarXiv ✨ podcast

This plot shows the spectral energy density of a real little red dot, compared with that predicted from a globular cluster that harbours a super massive star in its core. This is a key observable for little red dots, and the forming globular cluster hypothesis is broadly consistent with it

Michelle's final paper discussed yet another theory about how Little Red Dots found by JWST are formed. Could they be globular clusters in formation? This paper thinks so, and suggests observational tests! arxiv.org/abs/2602.15935 🔭 ☄️

Payel's final paper examines the possibility of habitable zones around massive stars. They vanish on the Main Sequence above about 12 solar masses but make a brief come back post main sequence even in more massive stars. Tune in to find out more! 🔭 ☄️ arxiv.org/pdf/2602.13875

A plot showing likely orbits generated by three different asteroid families, Flora, Nysa and Eunomia. Apophis' orbit is most compatible with being ejected from the Flora family.

Michelle's second paper discussed the Apophis asteroid which will make a close pass with the Earth on Friday 13th April 2029. But where did it come from? And how often do we expect encounters with Apophis like objects? This paper discusses all! 🔭 ☄️ arxiv.org/abs/2602.19849

Figure 3: M_UV vs M_stars

Payel's first paper introduces a very simple model for bursty star formation that is able to replicate recent observations of where galaxies lie in the plane of UV magnitude and stellar mass. ☄️ 🔭 arxiv.org/pdf/2602.16706

An image of the 7th known triple-double radio galaxy detected with MIGHTEE. Its three lobes tell us about different epochs of AGN activity, and are shown as contours. The oldest are the outer most and the youngest are the inner most.

Michelle's first paper discussed an object she'd never heard about before: triple-double radio galaxies.While very rare, they have the potential to teach us about the duty cycle of active galactic nuclei ☄️ 🔭 arxiv.org/abs/2602.19729

Our latest episode is live! Tune in to hear about where the Apophis asteroid came from, bursty star formation at high redshift, truple-double radio galaxies and more! Available below and wherever you get your podcasts 🔭 ☄️

Episode 31 – Near Earth Objects, Little Red Dots, bursty star formation and life around massive stars In this podcast episode, Michelle and Payel discuss theories on little red dot formation, bursty star formation, and the origins of asteroids like Apophis. They also discuss the potential for life around massive stars and unusual triple-double radio galaxies.

Episode 31 – Near Earth Objects, Little Red Dots, bursty star formation and life around massive stars

In this podcast episode, Michelle and Payel discuss theories on little red dot formation, bursty star formation, and the origins of asteroids like Apophis. They also discuss the potential for life…

Me and @readdark.bsky.social are advertising aPhD position in our lovely group here at Surrey university. We're looking to understand unusual globular clusters and faint dwarf galaxies. Check out the advert and get in touch if you have questions 🔭🧪 www.surrey.ac.uk/fees-and-fun...

Payel's last paper is another JWST discovery paper! The earliest ever observed nuclear stellar disc. ☄️ 🔭 arxiv.org/pdf/2601.18871

Left: [Si/Mn] vs [Al/Fe] abundance plane for the two clusters associated with GSE and the in-situ sample. The three highest-density levels are shown with filled contours (spanning from 60% to 100% of the peak KDE density), while stars outside these regions are plotted as scatter points to highlight lowerdensity members. Black lines denote limits commonly adopted in the literature. Right: Same for [Si/Mn] vs [Na/Fe], with a gray line at solar.

Nicole’s second paper uses a graph-attention network to mine GALAH DR4 chemistry and finds that Gaia-Sausage-Enceladus splits into two chemically distinct groups with similar dynamics. arxiv.org/abs/2602.02226 ☄️ 🔭

Payel's second paper presents hydrodynamic simulations of direct collapse black holes and associated synthetic spectra that explain pretty much every observational detail of the enigmatic Little Red Dots. ☄️ 🔭 arxiv.org/pdf/2601.14368

The top and bottom rows display example galaxies with and without tidal features, respectively. Each column represents different image processing methods: (a) original images with no mask (NM), (b) images after masking faint features (MF), (c) images after masking bright features (MB), and (d) images after masking bright features and inverting unmasked, star-particle pixels (MBI).

Nicole’s first paper looks at whether faint tidal debris in LSST-depth images can actually help CNNs spot galaxy mergers - and finds they do! Models perform best when bright cores are masked out, letting faint tidal tails dominate the classification signal. arxiv.org/abs/2602.03312 ☄️ 🔭

Payel's first paper reveals the highest resolution map of dark matter to date! JWST enabled of course. ☄️ 🔭 arxiv.org/pdf/2601.17239

Episode 30 – Direct collapse black holes, nuclear stellar discs, and machine learning merger histories In this episode, Payel and Nicole delve into more JWST discoveries and further pushing the frontier of machine learning in astronomy - ultra-deep view of the cosmic web, machine-learning deep images to look for mergers, a direct collapse black hole explanation to Little Red Dots, machine-learning the Milky Way to reveal complex star formation histories of accreted systems, and the earliest nuclear stellar disc observed to date.

Episode 30 – Direct collapse black holes, nuclear stellar discs, and machine learning merger histories

In this episode, Payel and Nicole delve into more JWST discoveries and further pushing the frontier of machine learning in astronomy - ultra-deep view of the cosmic web, machine-learning deep…

Photo of the slide advertising Payel's podcast! There is also light casting in from a window because we didn't shut the blinds oops

Just had a great talk here in Vienna about Galactic archeology by @payeldas.bsky.social! Go listen to her podcast @starxiv.bsky.social!

JWST imaging of Pseudo-LRD-NOM across all filters. The red compact source is clearly observed in all images.

Michelle's final paper focused on the observations of a pseudo-little red dot! A red, compact source at the heart of a lensed star bursting dwarf galaxy. It shares some properties with Little red dots, but not all. Could it be an LRD in waiting? arxiv.org/abs/2601.11466 🔭☄️

The stunning WEAVE large IFU dataset. The ring nebula is clearly seen in the three pointings with WEAVE. Each pixel is also a spectrum, allowing them to probe the kinematics and chemistry of the ring.

JWST imaging of the Ring nebula! The iron bar feature is highlighted as contours in the centre of each image.

Michelle's second paper showed the discovery from the fabulous new WEAVE instrument in the Canary island. This IFU found a surprising bar of iron in the centre of the ring nebula. It doesn't seem to be a jet feature, so what is it?!? 🔭 ☄️ arxiv.org/abs/2601.10635

Carbon abundances ([C/Fe]) as a function of [Fe/H] for the five CEMP stars compared to literature samples. Specifically, we compare to results from Chiti et al. (2024) which reports abundances for some of the most metal poor stars known in the LMC, along with results from APOGEE spectra in SDSS-IV DR17. The commonly-used definition for CEMP ([C/Fe]>0.7) is shown as a black dashed line, with the five stars from this work being the only stars above said line.

Nicole’s second paper reports the first CEMP stars ever found in the LMC, 5 extremely metal-poor giants with strong carbon enhancement, revealing that early carbon production operated even in this massive dwarf. arxiv.org/abs/2601.10514 ☄️ 🔭

The MDF of the Thamnos sample. The individual histograms show the Rg8 (orange) and Rg9 (green) subgroups, which sum to the total distribution (black solid line). The distribution is f it by a double Gaussian model (red solid line). This fit reveals two components: Component 1 (green dashed line, with a peak at [Fe/H] = −2.17), representing the dominant progenitor, and Component 2 (blue dashed line, with a peak at [Fe/H] = −1.49).

Nicole’s first paper confirms that the retrograde groups Rg8 and Rg9 are a single, very metal-poor halo substructure - Thamnos. High-resolution abundances show coherent chemistry and no α-knee, pointing to a progenitor quenched early after falling into the Milky Way. arxiv.org/abs/2601.09796 ☄️🔭

Figure 1 fromWiseman et al which shows that the Son et al inferred bias of SN with host mass (n=middle panel) is something that isn't seen once host corrections on galaxy stellar mass (left and right panels)

Michelle started us off with a response to last years claim that Supernova Ia aren't as reliable as we thought as cosmic probes. This work by Phil Wiseman et al. steps through the claims and shows that these ARE the dark energy tracers we are looking for. ☄️ 🔭 arxiv.org/abs/2601.13785

Episode 29 – Exploding stars, carbon stars, and starbursting pseudo little red dots In this episode, Michelle and Nicole explore recent findings on Thamnos, carbon stars in the Large Magellanic Cloud, Supernova Type Ia as cosmic probes, and an intriguing object identified as a 'pseudo little red dot'.

Episode 29 – Exploding stars, carbon stars, and starbursting pseudo little red dots

In this episode, Michelle and Nicole explore recent findings on Thamnos, carbon stars in the Large Magellanic Cloud, Supernova Type Ia as cosmic probes, and an intriguing object identified as a 'pseudo little red…

Sound the klaxon! The latest StarXiv has dropped. Tune in to hear about Supernovae, carbon stars, Thamnos, the Ring nebula and more as we delve into the last 2 weeks on astro-ph. 🔭 ☄️

It will. But maybe not for about 100,000 years 😉

Autoencoders are only trained on 'normal' datasets, meaning the reconstruction of 'anomalous' data will be much worse, as you can see in this comparison. The left panel shows the original normal spectrum and its (very good) reconstruction. On the right is the spectrum and reconstruction for the anomalous spectrum. It is worse, and measuing how much can let you automatically detect anomalies.

Michelle's final paper studies the use of autoencoders and dimension reduction for detecting anomalous features in exoplanet atmospheres. It can be quick and accurate, great for future exoplanet missions like ESA Ariel. 🔭 ☄️ arxiv.org/abs/2601.02324

Finally, Payel gave a quick shout out to what seems like an incredible piece of work recommending best practises for improving the situation for minorities in Astronomy. 🔭 ☄️
arxiv.org/abs/2512.24465

From the figure caption: Comparison with cold dark matter (CDM) predictions. The contours represent the number density of CDM subhaloes with Msub > 105h−1M⊙. These were obtained using the SASHIMI-C semi-analytical subhalo model [50, 51] assuming a host redshift and mass of zh = 0.881 and Mh = 1012M⊙. The coloured crosses show the Vmax and rmax mean values and 1-σ uncertainties inferred for different models of the detected object: an NFW profile with free concentration (NFW, ∆ ln E=−44), an NFW halo with concentration drawn from the mass-concentration relation in [28] (NFWCDM,field, ∆ ln E=−75) and an NFW subhalo with concentration drawn from [29] (NFWCDM, ∆ ln E=−147). In any case, it would be extreme!

Michelle's second paper focused on an unusal, thin gravitational arc around the near-Einstein ring around JVAS-1933+666. When inferring the mass profile of the perturber, it is much more dense than expected from cold or warm dark matter if its a subhalo. 🔭 ☄️ arxiv.org/abs/2601.02466

Payel's second paper talked about the incredible prospect of chemical signatures in host stars resulting from engulfing planets! Cool that this is even possible. 🔭☄️
arxiv.org/abs/2601.00949

Payel's first paper looked into another case of a star seemingly about to explode as a supernova, but a little further afield than Betelgeuse - a companion star seems again to blame! 🔭☄️
arxiv.org/abs/2601.02057

From figure caption: schematic drawing from above of Betelgeuse and the observed wake caused by the companion star. The orbit of the companion, at∼2.3 R⋆ is shown by a broken blue line. It is well within the total extent of the Mg II emission which reaches 6.4R⋆ - marked by diffuse orange color. When observed at different phases, along the direction of the arrows, the expanding wake can be detected.

This figure shows how properties of the Manganese I line vary with the period of the proposed companion stars orbital period. The changes in radial velocity and equivalent width are particularly clear, as the values oscillate, increasing and decreasing at transit and eclipse.

Michelle's first paper focuses on her all time favourite star, Betelgeuse! Recent work has shown that it likely has a little companion orbiting it, and this paper searches for (and finds) evidence of its wake! 🔭☄️ arxiv.org/abs/2601.00470

Our latest episode is live! Tune in on our website or wherever you get your podcasts! Today, Michelle and Payel are back together, discussing unusual variable stars and their compansions, signatures of exoplanets on their host stars and problems for dark matter. 🔭☄️ starxiv.com/2026/01/12/e...