Identification and characterization of low mass stars and brown dwarfs using Virtual Observatory tools

Context

Two thirds of the stars in our galactic neighborhood (d < 10 pc) are M-dwarfs which also constitute the most common stellar objects in the Milky Way. This property, combined with their small stellar masses and radii, increases the likelihood of detecting terrestrial planets through radial velocity and transit techniques, making them very adequate targets for the exoplanet hunting projects. Nevertheless, M dwarfs have associated different observational difficulties. They are cool objects whose emission radiation peaks at infrared wavelengths and, thus, with a low surface brightness in the optical range. Also, the photometric variability as well as the significant chromospheric activity hinder the radial velocity and transit determinations. It is necessary, therefore, to carry out a detailed characterization of M-dwarfs before building a shortlist with the best possible candidates for exoplanet searches.

Brown dwarfs (BDs) are self-gravitating objects that do not get enough mass to maintain a sufficiently high temperature in their core for stable hydrogen fusion. They represent the link between low-mass stars and giant planets. Due to their low temperatures, BDs emit significant flux at mid-infrared wavelength which makes this range very adequate to look for this type of objects.

The Virtual Observatory (VO) is an international initiative designed to help the astronomical community in the exploitation of the multi-wavelength information that resides in data archives. In the last years the Spanish Virtual Observatory is conducting a number of projects focused on the study of substellar objects taking advantage of Virtual Observatory tools for an easy data access and analysis of large area surveys. This is the framework where this thesis has been carried out.

This dissertation addresses three problems in the framework of low-mass stars and brown dwarfs, namely, the search for brown dwarf candidates crossmatching catalogues (Chapter 4), the search for nearby bright M dwarfs and the subsequent spectroscopic characterization (Chapter 5), and a study of binarity in mid to late-T brown dwarfs (Chapter 6); the first two topics use Virtual Observatory tools.

Aims and methodology

In the first paper we carried out a search of brown dwarfs in the sky area in common to the WISE, 2MASS Point Source and SDSS catalogues. A VO-workflow with the criteria that must accomplish our candidates was built using STILTS. The workflow returned 138 sources that were visually inspected. For the six new candidates that passed the inspection, proper motions were calculated using the positions and the different observing epochs of the catalogues previously quoted. Effective temperatures were estimated using VOSA and spectral types and distances using appropriate photometric calibrations.

In the second publication we conducted an all-sky photometric search by cross correlating the Carlsberg Meridian Catalogue (CMC14) and the 2MASS Point Source Catalogue with the aim of increasing the number of known, nearby M dwarfs that could be used as targets for exoplanet searches in general and CARMENES in particular. This VO search was combined with low-resolution spectroscopic followup of 27 objects using the IDS spectrograph at the Isaac Newton telescope at La Palma, as well as with an astrometric and photometric study.

In the third paper we attempted to refine the multiplicity properties of T dwarfs studying the largest sample so far observed with high angular resolution imaging. We undertook two parallel programs using the Wide Field Camera 3 (WFC3) installed on the Hubble Space Telescope (HST). We used a PSF-fitting subtraction technique to reveal the presence of any close companion to the sources in our sample. Monte Carlo simulations were carried out to estimate the capability of WFC3 to detect close binaries in terms of angular separation and magnitude difference. Simulations were also used to determine the fraction of binaries that would have been detected around each source based on assumed separations, mass ratio distributions and orientations of the systems.

Results

The main conclusion from this dissertation is that the Virtual Observatory has proved to be an excellent research methodology in the field of low mass stars and brown dwarfs. In particular, it allowed an efficient management of the queries to different catalogues and archives as well as the estimation of physical parameters through VO-tools.

In the first publication we present the identification of 31 brown dwarf (25 known and 6 strong candidates not previously reported in the literature) identified in the sky area in common toWISE, 2MASS and SDSS. This is a remarkable number considering that 2MASS has been extensively searched for ultracool dwarfs and clearly show how new surveys and the use of VO tools can help to mine older surveys. The robustness of our methodology was confirmed with the spectroscopic confirmation of our candidate targets making it an ideal technique to identify brown dwarfs and, by extension, other rare objects.

In the second paper, we show the potential of the VO and a purely photometric approach for finding new bright, nearby M dwarfs that escaped previous surveys mostly based on proper motions. We discover 24 new potential targets for exoplanet hunting (7 at less than 20 pc), 12 of which have been included in the CARMENES input catalogue of M dwarfs. We also identify three young very low-mass stars (M4-M5 spectral types) in the Taurus-Auriga region and a wide (110 AU) binary system. 

In the third paper we infer an upper limit for the binary fraction of >T5 dwarfs of <16 − < 25% depending of the underlying mass ratio distribution. This binary fraction is consistent with previous estimations. From this work we also conclude that theWFC3 is more sensitive to cool companions than otherHST instruments like NICMOS or WFPC2 but its lower angular resolution makes it unsuitable to detect tight brown dwarf binary systems.