Search and characterization of exo-earths participation in the development of the espresso spectrograph for the VLT
The search for exo-Earths using Doppler spectroscopy faces nowadays two main difficulties. The first is a technical limitation. The precision of the instruments limits the measurable amplitudes to signals produced by super-Earths orbiting close to their host stars. Terrestrial planets orbiting solar...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Fecha de publicación: | 2016 |
| País: | España |
| Institución: | Universidad de La Laguna (ULL) |
| Repositorio: | RIULL. Repositorio Institucional de la Universidad de La Laguna |
| OAI Identifier: | oai:riull.ull.es:915/26025 |
| Acceso en línea: | http://riull.ull.es/xmlui/handle/915/26025 |
| Access Level: | acceso abierto |
| Palabra clave: | Astronomía óptica Planetas |
| Sumario: | The search for exo-Earths using Doppler spectroscopy faces nowadays two main difficulties. The first is a technical limitation. The precision of the instruments limits the measurable amplitudes to signals produced by super-Earths orbiting close to their host stars. Terrestrial planets orbiting solar-type stars induce variations smaller than 1 m s−1 except when orbiting very close to the star. The arrival of ESPRESSO and the Laser Frequency Comb as calibration unit will palliate this limitation and move the measurement precision to the range of 5 cm s−1 . So far the search in solar-Type stars is limited to super-Earths. The alternative is to find terrestrial planets in the habitable zone is the search around M-type stars. Their lower mass makes the signals of terrestrial planets detectable by today’s instruments. The second issue is the presence of radial velocity variations induced by the homogeneities of the stellar surface. The changes in the stellar surface and the stellar rotation induce periodic signals that go from less than 1 m s−1 in the quietest stars to dozens of m s−1 in active stars and might be easily mistaken for signals induced by rocky planets. In the case of M-dwarfs the need of performing the analysis of the stellar induced signals is even more important. Signals produced by early M-dwarfs have periods compatible with the habitable zone of those stars and amplitudes similar to those of super-Earths. The rise in precision expected for the next years will make this kind of signals visible even in the quietest stars making even more important to understand and correctly model them. Using chromospheric indicators from high resolution spectroscopic and photometric time series we study both long term and short term variability (solarlike cycles and rotation) of a sample of 176 low magnetic activity stars with spectral types going from late F-type to mid M-type stars. We provide new measurements of magnetic cycles for 105 stars, and rotation periods for 123 stars. We study the distribution of rotation periods, finding that there is an evolution in the rotation period lengths, getting longer for later type stars, that saturates when reaching K-type stars and begins to increase again at types later than M2. We study the distribution of magnetic cycles, finding no clear difference among different spectral types. We study the relationships between the measured periods and the chromospheric activity level of the stars, measured as the log10(R0 HK). To do so we extend the log10(R0 HK) calibration to cover stars up to B-V ∼ 2.0. We find a strong correlation between the measured rotation period and the mean level of chromospheric activity that allows to predict the rotation periods of main sequence stars from G-type to mid M-type with an accuracy of the ∼ 20% of the period. We analyse the radial velocity time series of 133 stars searching both for activity induce signals and planetary signals. We detect rotation induced signals in 46 stars, with amplitudes going from 0.3 to 16.4 ms−1 . We study the relationship between the amplitude of the induced signals and the activity indicators, finding a correlation between the mean level of chromospheric activity log10(R0 HK) and the measured amplitude of the signal. This relationship shows that FGK stars show a similar behaviour, which is different for M-type stars. We find also a relationship between the amplitude of the modulation in the Mount Wilson S index and the amplitude of the induced radial velocity signal. This relationship behaves differently for two populations of stars, with a breaking point around mid-K stars. After cleaning the radial velocity series form stellar activity induced signals we performed a search for dynamical signals in the same sample of stars. We recovered most of the previously published planet candidates, except for a few cases where we attribute a stellar origin to the previously assumed planetary companion. We present the discovery of six new planets around the stars HD 1581, HD 161098, HD 176986, GJ 536 and GJ 3998, with minimum masses in M⊕ units of 5.87 ± 0.99, 26.60 ± 4.39, 4.86 ± 0.66 and 5.47 ± 0.84, 2.35 ± 0.42 and 6.12 ± 1.11 respectively located at 0.321, 2.503, 0.061, 0.066, 0.029 and 0.089 au from their parent stars. The latter two have been reported by Affer et al. (2016). Five of them are presumably rocky planets. We suggest the presence of a seventh one in the star HD 161098 with a minimum mass of 5.30 ± 1.22 M⊕ located at 0.317 au of the star, which could be habitable given the appropriate atmospheric conditions. We suggest also the presence of five additional candidates that still require further analysis and observations. |
|---|