Explore the vast range of titles available.

We still do not understand how planets form or why extrasolar planetary systems are so different from our own Solar System. However, the past few years have dramatically changed our view of the disks of gas and dust around young stars. Observations with the Atacama Large Millimeter/submillimeter Array and extreme adaptive-optics systems have revealed that most—if not all—disks contain substructure, including rings and gaps1–3, spirals4–6, azimuthal dust concentrations7 and shadows cast by misaligned inner disks5,8. These features have been interpreted as signatures of newborn protoplanets, but the exact origin is unknown. Here we report the kinematic detection of a few-Jupiter-mass planet located in a gas and dust gap at 130 au in the disk surrounding the young star HD 97048. An embedded planet can explain both the disturbed Keplerian flow of the gas, detected in CO lines, and the gap detected in the dust disk at the same radius. While gaps appear to be a common feature in protoplanetary disks2,3, we present a direct correspondence between a planet and a dust gap, indicating that at least some gaps are the result of planet–disk interactions.Pinte et al. report the kinematic detection of a few-Jupiter-mass planet orbiting at 130 au from the young star HD 97048. The radial position of the planet coincides with a gap in both the gas and dust components of the protoplanetary disk, showing that at least some gaps can be linked to the presence of planets.

The main challenge of exoplanet high-contrast imaging (HCI) is to separate the signal of exoplanets from their host stars, which are many orders of magnitude brighter. HCI for ground-based observations is further exacerbated by speckle noise originating from perturbations in Earth's atmosphere and imperfections in the telescope optics. Various data post-processing techniques are used to remove this speckle noise and reveal the faint planet signal. Often, however, a significant part of the planet signal is accidentally subtracted together with the noise. In the present work, we use explainable machine learning to investigate the reason for the loss of the planet signal for one of the most used post-processing methods: principal component analysis (PCA). We find that PCA learns the shape of the telescope point spread function for high numbers of PCA components. This representation of the noise captures not only the speckle noise but also the characteristic shape of the planet signal. Building on these insights, we develop a new post-processing method (4S) that constrains the noise model to minimize this signal loss. We apply our model to 11 archival HCI datasets from the VLT-NACO instrument in the L'-band and find that our model consistently outperforms PCA. The improvement is largest at close separations to the star (\\(\\leq 4 \\lambda /D\\)) providing up to 1.5 magnitudes deeper contrast. This enhancement enables us to detect the exoplanet AF Lep b in data from 2011, 11 years before its subsequent discovery. We present updated orbital parameters for this object.

In order to solve the puzzle of the origin of the exoplanets and the Solar System, it is necessary to observe on-going planet formation. The young gas- and dust-rich circumstellar disks, also called protoplanetary disks, are the expected birthplace of planets. A fraction of these disks, referred to as transition disks,were identified to harbor inner clearings in their dust distribution, with some of these gaps extending over several dozens au, possibly due to dynamical carving by nascent giant planets. This thesis takes advantage of the synergy between the Atacama Large Millimeter Array (ALMA) and high-contrast imaging (HCI) instruments to study planet formation and the planet-disk mutual feedback in these transition disks with large gaps.The first part of my thesis is dedicated to the imaging of transition disks with large gaps. New high-quality images obtained with ALMA and new HCI instruments have revealed several signposts for the presence of companions in these disks, including large gaps, asymmetric dust distributions, shadows, warps and spiral arms. In this context, I performed a detailed analysis of the spirals found in the transition disks of MWC 758 and HD 142527. The spiral arms of MWC 758, along with other features of the disk, appear most compatible with the presence of two embedded companions, with one of them being tentatively detected inside the disk cavity based on our data. In the case of HD 142527, we conclude that the spirals stemming from the edge of the cavity are, along with other features of the disk, by-products of the disk-binary dynamical interaction. The very cold large-scale CO spirals seen with ALMA have a more uncertain origin, which could be related either to gravitational instability or the shadows cast by the inclined inner disk.To test the hypothesis that large gaps in transition disks are due to embedded companions, I carried out a HCI survey of those disks using VLT/NACO in thermal IR, presented in the second part of this thesis. I implemented a data reduction pipeline, and used it to search for faint companions in all the disks already observed in this (still on-going) survey. So far, four companion candidates have been identified (out of 15 sources observed), although follow-up observations are required to confirm they are genuine companions. I also present my contribution to the detection of a young extremely red substellar companion in the debris disk of HD 206893 using thermal-IR NACO+AGPM data.The potential of integral field spectrographs (IFS) to detect and characterize faint companions is investigated in the third part of this thesis. A sample of five transition disks with large gaps was observed with VLT/SINFONI in near-IR. The combination of angular and spectral differential imaging (ASDI) enabled to efficiently suppress speckles and achieve high contrasts in a range of radial separations. We (re)detected three companions (one new detection to be confirmed) and two spiral patterns (one tentative). In particular, for HD 142527 the low-mass companion was redetected in most spectral channels, which allowed me to carry out a detailed spectral characterization and infer its physical parameters.

Circumstellar disk images have highlighted a wide variety of morphological features. Recovering disk images from high-contrast angular differential imaging (ADI) sequences are however generally affected by geometrical biases, leading to unreliable inference of the morphology of extended disk features. Recently, two types of approaches have been proposed to recover more robust disk images from ADI sequences: iterative principal component analysis, and inverse problem approaches. We introduce MUSTARD, a new IP-based algorithm designed to address the problem of the flux invariant to the rotation in ADI sequences; a limitation inherent to the ADI observing strategy, and discuss the advantages of IP approaches with respect to PCA-based algorithms. The MUSTARD model relies on the addition of morphological priors on the disk and speckle field to a standard IP approach to tackle rotation-invariant signal in circumstellar disk images. We compare the performance of MUSTARD, I-PCA, and standard PCA on a sample of high-contrast imaging data sets acquired in different observing conditions, after injecting a variety of synthetic disk models at different contrast levels. MUSTARD significantly improves the recovery of rotation-invariant signal in disk images, especially for data sets obtained in good observing conditions. However, the MUSTARD model is shown to inadequately handle unstable ADI data sets, and to provide shallower detection limits than PCA-based approaches. MUSTARD has the potential to deliver more robust disk images by introducing a prior to address the inherent ambiguity of ADI observations. However, the effectiveness of the prior is partly hindered by our limited knowledge of the morphological and temporal properties of the stellar speckle halo. In light of this limitation, we suggest that the algorithm could be improved by enforcing a prior based on a library of reference stars

High-contrast imaging (HCI) is a technique designed to observe faint signals near bright sources, such as exoplanets and circumstellar disks. The primary challenge in revealing the faint circumstellar signal near a star is the presence of quasi-static speckles, which can produce patterns on the science images that are as bright, or even brighter, than the signal of interest. Strategies such as angular differential imaging (ADI) or reference-star differential imaging (RDI) aim to provide a means of removing the quasi-static speckles in post-processing. In this paper, we present and discuss the adaptation of state-of-the-art algorithms, initially designed for ADI, to jointly leverage angular and reference-star differential imaging (ARDI) for direct high-contrast imaging of circumstellar disks. Using a collection of high-contrast imaging data sets, we assess the performance of ARDI in comparison to ADI and RDI based on iterative principal component analysis (IPCA). These diverse data sets are acquired under various observing conditions and include the injection of synthetic disk models at various contrast levels. Our results demonstrate that ARDI with IPCA improves the quality of recovered disk images and the sensitivity to planets embedded in disks, compared to ADI or RDI individually. This enhancement is particularly pronounced when dealing with extended sources exhibiting highly ambiguous structures that cannot be accurately retrieved using ADI alone, and when the quality of the reference frames is suboptimal, leading to an underperformance of RDI. We finally apply our method to a sample of real observations of protoplanetary disks taken in star-hopping mode, and propose to revisit the protoplanetary claims associated with these disks.

The advent of high-contrast imaging instruments combined with medium-resolution spectrographs allows spectral and temporal dimensions to be combined with spatial dimensions to detect and potentially characterize exoplanets with higher sensitivity. We develop a new method to effectively leverage the spectral and spatial dimensions in integral-field spectroscopy (IFS) datasets using a supervised deep-learning algorithm to improve the detection sensitivity to high-contrast exoplanets. We begin by applying a data transform whereby the IFS datasets are replaced by cross-correlation coefficient tensors obtained by cross-correlating our data with young gas giant spectral template spectra. This transformed data is then used to train machine learning (ML) algorithms. We train a 2D CNN and 3D LSTM with our data. We compare the ML models with a non-ML algorithm, based on the STIM map of arXiv:1810.06895. We test our algorithms on simulated young gas giants in a dataset that contains no known exoplanet, and explore the sensitivity of algorithms to detect these exoplanets at contrasts ranging from 1e-3 to 1e-4 at different radial separations. We quantify the sensitivity using modified receiver operating characteristic curves (mROC). We discover that the ML algorithms produce fewer false positives and have a higher true positive rate than the STIM-based algorithm, and the true positive rate of ML algorithms is less impacted by changing radial separation. We discover that the velocity dimension is an important differentiating factor. Through this paper, we demonstrate that ML techniques have the potential to improve the detection limits and reduce false positives for directly imaged planets in IFS datasets, after transforming the spectral dimension into a radial velocity dimension through a cross-correlation operation.

Observations of WASP-107b suggest a metal-rich and carbon-deprived atmosphere with an extremely hot interior based on detections of SO\\(_2\\), H\\(_2\\)O, CO\\(_2\\), CO, NH\\(_3\\), and CH\\(_4\\). In this paper, we aim to determine the reliability of a 1D radiative-convective photochemical-equilibrium (1D-RCPE) retrieval method in inferring atmospheric properties of WASP-107b. Our grid of radiative-convective balanced pressure-temperature profiles and 1D photochemical equilibrated models covers a range of metallicities (Z), carbon-to-oxygen ratios (C/O), intrinsic temperatures (T\\(_{int}\\)), and eddy diffusion coefficients (K\\(_{zz}\\)). We obtain good fits with our 1D-RCPE retrievals based on a few molecular features of H\\(_2\\)O, CO\\(_2\\), SO\\(_2\\), and CH\\(_4\\), but find no substantial contribution of NH\\(_3\\). We find that the degeneracy between metallicity, cloud pressure, and a model offset is broken by the presence of strong SO\\(_2\\) features, confirming that SO\\(_2\\) is a robust metallicity indicator. We systematically retrieve sub-solar C/O based on the relative amplitude of a strong CO\\(_2\\) feature with respect to the broad band of H\\(_2\\)O, which is sensitive to a wavelength-dependent scattering slope. We find that high-altitude clouds obscure the CH\\(_4\\)-rich layers, preventing the retrievals from constraining T\\(_{int}\\), but that higher values of K\\(_{zz}\\) can transport material above the cloud deck, allowing a fit of the CH\\(_4\\) feature. However, T\\(_{int}\\) and K\\(_{zz}\\) can vary substantially between retrievals depending on the adopted cloud parametrization. We conclude that the 1D-RCPE retrieval method can provide useful insights if the underlying grid of forward models is well understood. We find that WASP-107b's atmosphere is enriched in metals (3 to 5 times solar) and carbon-deprived (C/O <= 0.20). However, we lack robust constraints on the intrinsic temperature and vertical mixing strength.

Effective image post-processing algorithms are vital for the successful direct imaging of exoplanets. Standard PSF subtraction methods use techniques based on a low-rank approximation to separate the rotating planet signal from the quasi-static speckles, and rely on signal-to-noise ratio maps to detect the planet. These steps do not interact or feed each other, leading to potential limitations in the accuracy and efficiency of exoplanet detection. We aim to develop a novel approach that iteratively finds the flux of the planet and the low-rank approximation of quasi-static signals, in an attempt to improve upon current PSF subtraction techniques. In this study, we extend the standard L2 norm minimization paradigm to an L1 norm minimization framework to better account for noise statistics in the high contrast images. Then, we propose a new method, referred to as Alternating Minimization Algorithm with Trajectory, that makes a more advanced use of estimating the low-rank approximation of the speckle field and the planet flux by alternating between them and utilizing both L1 and L2 norms. For the L1 norm minimization, we propose using L1 norm low-rank approximation, a low-rank approximation computed using an exact block-cyclic coordinate descent method, while we use randomized singular value decomposition for the L2 norm minimization. Additionally, we enhance the visibility of the planet signal using a likelihood ratio as a postprocessing step. Numerical experiments performed on a VLT/SPHERE-IRDIS dataset show the potential of AMAT to improve upon the existing approaches in terms of higher S/N, sensitivity limits, and ROC curves. Moreover, for a systematic comparison, we used datasets from the exoplanet data challenge to compare our algorithm to other algorithms in the challenge, and AMAT with likelihood ratio map performs better than most algorithms tested on the exoplanet data challenge.

We present the re-detection of a compact source in the face-on protoplanetary disc surrounding HD 169142, using VLT/SPHERE data in YJH bands. The source is found at a separation of 0.''319 (\\(\\sim\\)37 au) from the star. Three lines of evidence argue in favour of the signal tracing a protoplanet: (i) it is found in the annular gap separating the two bright rings of the disc, as predicted by theory; (ii) it is moving at the expected Keplerian velocity for an object at \\(\\sim\\)37 au in the 2015, 2017 and 2019 datasets; (iii) we also detect a spiral-shaped signal whose morphology is consistent with the expected outer spiral wake triggered by a planet in the gap, based on dedicated hydrodynamical simulations of the system. The YJH colours we extracted for the object are consistent with tracing scattered starlight, suggesting that the protoplanet is enshrouded in a significant amount of dust, as expected for a circumplanetary disc or envelope surrounding a gap-clearing Jovian-mass protoplanet.

Disks around very low-mass stars (VLMS) provide environments for the formation of Earth-like planets. Mid-infrared observations have revealed that these disks exhibit weak silicate features and strong hydrocarbon emissions. This study characterizes the dust properties and geometrical structures of VLMS and brown dwarf (BD) disks, observed by the James Webb Space Telescope (JWST)/Mid-Infrared Instrument (MIRI), and connects these to gas column density and potential evolutionary stages. We analyze mid-infrared spectra of ten VLMS and BD disks as a part of the MIRI mid-Infrared Disk Survey (MINDS) program. Spectral slopes and silicate band strengths are compared with hydrocarbon emission line ratios, which probe the gas column density. Moreover, the Dust Continuum Kit with Line emission from Gas is used to quantify grain sizes, dust compositions, and crystallinity in the disk surface. The disks are classified into less, more, and fully settled geometries based on their mid-infrared spectral slopes and silicate band strengths. Less-settled disks show a relatively strong silicate band, high spectral slopes, and low crystallinity, and are dominated by 5 \\(\\mu\\)m-sized grains. More-settled disks have weaker silicate band, low spectral slope, enhanced crystallinity, and higher mass fractions of smaller grains. Fully-settled disks exhibit little or no silicate emission and negative spectral slopes. An overall trend of increasing gas column density with decreasing spectral slope suggests that more molecular gas is exposed when the dust opacity decreases due to dust settling. Our findings may reflect possible evolutionary pathways with dust settling and thermal processing or may point to inner-disk clearing or a collisional cascade. These results highlight the need for broader samples to understand the link between dust and gas appearance in regions where Earth-like planets form.