Aims. We model the atmosphere of the ultra-hot Jupiter (UHJ) WASP-178b while accounting for non-local thermodynamical equilibrium (NLTE) effects and compare synthetic transmission spectra with near-ultraviolet (NUV) and optical observations. Methods. We used the HELIOS code (LTE) in the lower atmosphere and the ClOUDY code (LTE or NLTE) in the middle and upper atmosphere to compute the temperature-pressure (TP) and abundance profiles. We further used CLOUDY to compute the theoretical planetary transmission spectrum both in LTE and NLTE for comparison with observations. Results. We find an isothermal TP profile at pressures higher than 10 mbar and lower than 10⁻⁸ bar, with an almost linear increase from ∼2200 K to ∼8100 K in between. The temperature structure is driven by NLTE effects, particularly in the form of increased heating resulting from the overpopulation of long-lived FeII levels with strong transitions in the NUV band, where the stellar emission is strong, and of decreased cooling due to the underpopulation of MgI and MgII levels that dominate the cooling. The planetary atmosphere is hydrostatic up to pressures of ∼1 nbar, and thus accurately modelling spectral lines forming at pressures lower than ∼1 nbar requires accounting for both hydrodynamics and NLTE effects. The NLTE synthetic transmission spectrum overestimates the observed Hα and Hβ absorption, while the LTE model is in good agreement, which is surprising because the opposite has been found for the other UHJs for which NLTE modelling has been performed. In the NUV, we find an excellent match between the NLTE transmission spectrum and the HST/UVIS data, contrary to the LTE model. This contrasts with previous LTE results requiring SiO absorption to fit the observations. Conclusions. The accurate characterisation of the atmosphere of UHJs is possible only when accounting for NLTE effects and particularly for the level population of Fe and Mg, which respectively dominate heating and cooling.

Context. Accretion from protoplanetary or debris discs can contaminate the stellar photosphere, especially in stars that have radiative envelopes. Due to the relatively slower photospheric mixing, these stars can exhibit clear contamination signatures. The contaminated photosphere reflects ongoing disc processes, which are detectable through stellar spectroscopy. Aims. We investigate the composition of six gas-rich debris disc-hosting A-type stars to understand possible links with their debris disc or earlier accretion stages. Methods. We used archival spectra and the ZEEMAN spectral synthesis code to estimate the stellar parameters and abundances of six debris disc-hosting A-type stars. We also estimated the stellar photospheric accretion contamination parameter, f_ph, using CAMSTARS, which indicates the fraction of accreting material on the stellar photosphere. Results. The oxygen abundance in intermediate-mass stars decreases with age until the debris disc stage (<20 Myr), after which it could end up rising. The downward trend could result from H₂O ice accumulating in dust traps or the formation of hydrated asteroids in the protoplanetary disc, locking oxygen in solids and reducing its accretion onto the star. All the stars share similar volatile abundances (C, O), but HD 110058 and HD 32297 exhibit refractory depleted abundances. The near-zero f_ph values in the six stars suggest that any currently accreted gas would not overwhelm mixing in the photosphere and would not impact the observed composition. The refractory depleted abundances in HD 110058 and HD 32297 suggest residual or even chronic, accretion contamination from their earlier protoplanetary stages when the accretion rates were about five orders of magnitude higher. For HD 110058, with the highest refractory depletion, we estimated a lower limit on its earlier protoplanetary accretion rate of 9 × 10⁻⁸ M_⊙/yr, similar to other Herbig stars and equal to the Herbig star HD 100546. This supports our hypothesis that refractory depletion in HD 110058 originates from a prior phase of higher accretion of dust-poor material. We further develop this hypothesis by comparing HD 110058 with the HD 100546 protoplanetary disc system, which is of a similar age.

Context. The hypergiant RW Cep is one of the largest stars in the Galaxy. The evolution and mass loss of such stars have profound effects on their surrounding regions and their galaxy as a whole. Between 2020 and 2024, RW Cep experienced a historic mass-loss event known as the Great Dimming. Aims. This study provides a spectroscopic analysis of RW Cep during the Great Dimming. We examine its atmospheric dynamics and place it in the context of the star’s variability behaviour since the early 2000s. Methods. We conducted high-cadence spectroscopic observations of RW Cep during the dimming event using the Tartu Observatory 1.5-meter telescope and the Nordic Optical Telescope. We analysed the atmospheric dynamics by measuring the radial velocities and line depths of Fe I and other spectral lines. Results. The radial velocities of the Fe I lines reveal a vertical velocity gradient of 10–20 km s⁻¹ in the atmosphere, correlating with the strength of the spectral lines. Stronger lines, formed in higher atmospheric layers, have higher radial velocities. We measured the systemic velocity at −50.3 km s⁻¹. During the dimming, radial velocities were affected by additional emission from the ejected gas, which was blue-shifted relative to the absorption lines. Post-dimming, we observed large-scale atmospheric motions with an amplitude of ~25 km s⁻¹. Strong resonance lines of Ba II, K I, Na I, and Ca I showed stable central emission components at −56 km s⁻¹, which are likely of circumstellar origin.

Context . The characterisation of exoplanetary systems depends on the accurate determination of host star parameters. The Ariel mission will probe the atmospheres of a statistically significant sample of exoplanets, and so requires a precise characterisation of the stellar properties well before its launch in 2031. The homogeneous determination of stellar parameters for Ariel will enable both the optimisation of the final target list and set roots for a reliable interpretation of the formation and evolution of planetary systems. Such a homogeneous characterisation has thus far only been carried out for the cool ( T _eff ≲ 7000 K) host stars among the Ariel target candidates. Aims . We present a uniform determination of fundamental stellar parameters for 18 hot stars ( T _eff ≳ 7000 K) in the Tier 1 candidate list of the Ariel mission candidate sample. Methods . We adopted an iterative spectro-trigonometric approach optimised for high-temperature stars ( T _eff ≳ 7000 K). Highresolution spectra were analysed using the ZEEMAN code with χ ² minimisation, combining model fits to metal and Balmer lines. Surface gravity was refined using photometry-based radii and masses from stellar evolutionary tracks. Results . We derived effective temperatures, surface gravities, projected rotational velocities, microturbulent velocities, overall metal-licities, iron abundances, stellar masses, and radii for our sample of 18 hot stars. Our results were validated against a set of benchmark stars previously presented in the literature, confirming that our methods yield results consistent with previous studies. This ensures an internally consistent parameter scale and maintains continuity across the transition from cool to hot stars within the Ariel sample. Conclusions . The derived parameters provide an internally consistent basis for studying the link between stellar properties and planetary characteristics in intermediate-mass stars (1.5 < M < 2.32 M _⊙ ). Building on our previous work on FGK host stars, we show that correlations between stellar mass, metallicity, and planetary radii also extend to early-type stars, and stellar properties influence the architecture of multi-planet systems.

Context. Phosphorus (P) is an essential element for life on Earth and a potential tracer of planet formation history. However, there has been no detection of P-bearing molecules in protoplanetary discs so far. Herbig Ae/Be stars constantly accrete matter from their protoplanetary disc, which alters the composition of the stellar photosphere due to their shallow convective or fully radiative envelope. The altered surface composition reflects the composition of the accreting matter, and thus, that of the inner protoplanetary disc. This accretion contamination of stellar photosphere can persist after accretion has ended in young A- and B-type stars (age ≲ 50 Myr). Aims. We quantify the fraction of P that is locked in dust (the refractory fraction of P) compared to gas in the inner protoplanetary disc around Herbig Ae/Be stars. Methods. We measured the stellar parameters and abundance of five Herbig Ae/Be stars using optical and UV spectra to compare their P and Fe abundances. We also used a ∼20 Myr old main-sequence B-type star, whose P and Fe abundance was estimated from an optical spectrum. Fe is assumed to be completely locked in refractory reservoirs in the inner disc. A parametrised relation between the stellar P and Fe abundance gives the fraction of P that is locked in refractory reservoirs. Results. The refractory fraction of P in the inner protoplanetary disc is ≥96% within the 95th percentile of the posterior distribution. The reference P abundance for young stars in the solar neighbourhood is −6.66 ± 0.05, which is similar to solar values. Conclusions. Consistent with a previous finding in the HD 100546 system, we conclude that most of the P in the inner protoplanetary disc is locked in dust, likely in refractory minerals such as schreibersite or apatite. Our result, combined with the low cosmic abundance of P, is also consistent with the lack of infrared and millimeter-wavelength observations of P-bearing molecules in protoplanetary discs to date.

We introduce the data reduction pipeline for the Tartu Observatory Fiber-fed Echelle Spectrograph (TOFES). TOFES is installed in the Coudé room and will be connected to the 1.5 m Tartu Observatory AZT-12 telescope through a four-channel instrument adapter to be mounted at the Cassegrain focus of the telescope. The spectrograph has an average spectral resolution of ∼30,000 and covers the 390–900 nm wavelength band in a single exposure. The data reduction pipeline, based on the PyReduce package, was tested on spectra of the Sun. We also present the Spectroscopy-Toolbox package, which was developed to provide additional tools for diagnostics and spectral line identification for radial velocity measurements. The spectrograph will address a range of scientific questions, including the stellar characterisation of Herbig Ae/Be stars to measure accretion contamination from their protoplanetary disks, the stellar characterisation of exoplanet host-stars including the Ariel space mission targets, and radial velocity monitoring of large-scale atmospheric variability in massive stars.