Project Topics
Get Complete Project Material File(s) Now! »
SOPHIE
SOPHIE (Spectrographe pour l’Observation des PHénomènes des Intérieurs stellaires et des Ex-oplanètes) is a high resolution échelle spectrograph at the 1.93 m telescope of Observatoire de Haute-Provence (OHP), France, that has been in operation since 2006 (Bouchy & Sophie Team, 2006; Perruchot et al., 2008). It is the successor of the ELODIE spectrograph which was in opera-tion from 1994 to 2006 (Baranne et al., 1996). Perruchot et al. (2008) provides a detailed description of the instrument’s optical design and technical key points for obtaining high accuracy radial ve-locity measurements. In 2011, SOPHIE was upgraded to SOPHIE+, by replacing the circular fibre with an octagonal-section fibre in the fibre link to improve the scrambling (Bouchy et al., 2013; Perruchot et al., 2011).
Specifications of the SOPHIE
SOPHIE is a fiber-fed cross-dispersed echelle spectrograph dedicated to high-precision radial ve-locity measurements. It is enclosed in an environmentally stabilized chamber to avoid drift in the spectrum due to temperature and pressure variations. The spectrograph has two spectral resolu-tions (R) corresponding to its two modes – the high resolution (HR) mode with R = 75000, and the high-efficiency (HE) mode with R = 40000. The light from the telescope is fed into the spectrograph through optical fibers. One of the two fibers is illuminated by the target while the other is illumi-nated either by the sky spectrum for estimating background moon pollution or the simultaneous calibration lamp exposure (see next paragraph) for tracking spectrograph drift. The spectrum is then projected onto an e2V 44-82 CCD (Charge Coupled Device) detector (4096×2048, 15-micron pixels), which yields 41 spectral orders, out of which 39 orders are recorded covering a wavelength range from 3872Å to 6943 Å.
SOPHIE has a calibration unit with five slots for calibration lamps: two Thorium-Argon lamps, Tungsten, LDLS, and a Fabry-Perot etalon that are stored in a temperature-controlled room. The LDLS and the tungsten lamps are used for order localization and flat-field calibrations. The Thorium-Argon (Th-Ar) lamps are used for wavelength calibration. A Fabry-Perot etalon (FP) is used when observations with very accurate radial velocities are needed. The dense grid of lines and homo-geneous amplitude of the FP spectrum have definite merit over the irregular distribution of the Th-Ar emission lines.
HARPS-N
HARPS-N (High Accuracy Radial velocity Planet Searcher – North) is an echelle spectrograph lo-cated at the 3.6 m Telescopio Nazionale Galileo (TNG), La Palma. It is designed to obtain high-precision radial velocity measurements. The designs of SOPHIE and HARPS-N are quite similar. HARPS-N is fiber-fed and has two fibers – one for calibration or sky exposures and another for stel-lar light. HARPS-N is enclosed in an environmentally stabilized vacuum chamber to avoid drifts due to variations in temperature and pressure. The spectral wavelength range covered is 383 nm- 690 nm. It has a higher spectral resolution as compared to SOPHIE with R = 115000. The spectrum is projected on an e2V CCD 231 detector, which allows 69 spectral orders. For wavelength calibra-tion, HARPS-N has two calibration lamps similar to SOPHIE: Th-AR and Fabry-Perot (Cosentino et al., 2012).
HARPS-N has a better short-term radial velocity precision than SOPHIE with » 0.3 ms¡1, thanks to the larger telescope and an environmentally stabilized vacuum chamber. Therefore, HARPS-N is a much more suitable instrument to obtain radial velocity data for detecting small exoplanet and for obliquity measurement of planets with small radii.
SPIRou
SPIRou (SPectropolarimètre InfraROUge) is a near-infrared spectropolarimeter and a high-precision velocimeter installed at the 3.6 m Canada-France-Hawaii Telescope (CFHT), Hawaii. It is enclosed in a vacuum cryogenic vessel cooled down to a temperature of 73 K, and stabilized at a sub-mK level (Reshetov et al., 2012). The spectrograph provides spectra in various bands, i.e., Y, J, H, and K (in the wavelength range 0.95-2.35 „m). It has a spectral resolution of » 70,000, and it is fed with three fluoride fibers – two science fibers collecting light out of the polarimeter and a calibration fiber. The spectrum is recorded with a 15-micron science grade H4RG detector. SPIRou has two modes: spectropolarimetric and spectroscopic. The radial velocities of the star can be obtained from both modes.
SPIRou is designed to be one of the most precise infrared velocimeters worldwide on the sky since 2019. It will unveil many new exoplanets around different stellar populations, mainly late M-dwarfs. The polarimeter on SPIRou is an additional tool for stellar characterization that allows the study of magnetic fields and stellar activity signals in the RV data. Along with the additional information offered by the polarimeter, SPIRou enables detailed characterization of the planetary system.
Data and Data Reduction
An automatic Data Reduction Software (DRS) is used to extract spectra from the CCD images. The process of spectra extraction includes localization of the orders on the images, optimal order ex-traction, cosmic-ray rejection, wavelength calibration, and one-dimensional flat-field correction. SOPHIE, HARPS-N, and SPIRou have their own DRS (Bouchy et al., 2009b; Cosentino et al., 2014, , Cook et al.; 2020, in prep.)).
The resulting spectra are then cross-correlated with numerical masks (F0, G2, K0, K5, M4, and M5) corresponding to the spectral type of the star (see Figure 2.1). The obtained cross-correlation functions (CCF) is then fitted with Gaussians in order to derive the radial velocities (Baranne et al. (1996); Pepe et al. (2002)). The mean of the fitted Gaussian gives the radial velocity of the star. The thus obtained radial velocities (RVs) are then corrected for the charge transfer inefficiency (CTI) effect following Bouchy et al. (2009a). These RVs also include the barycentric correction corresponding to the observatory’s motion relative to the Solar System barycenter. Other stellar parameters such as metallicity, the projected rotational velocity, and activity index can also be obtained directly from the pipeline.
These radial velocity measurements are then fitted with a Keplerian model derived from the physics behind the Doppler motion of the star, as described in Section 2.3. These models reveal the information about the exoplanet around that star. However, it should be taken into account that this radial velocity signal of a few ms¡1 can be mimicked by stellar activity (Queloz et al., 2001) and face-on binaries (e.g. Díaz et al., 2012; Wright et al., 2013). To tackle such false positive scenarios, some indicators can be used to establish that the RV signal is, in fact, due to the Doppler of the star and not because of other stellar effects (see Section 2.5 for a more detailed explanation).
Rossiter–McLaughlin Effect and Obliquity Measurement
The Rossiter–McLaughlin Effect is a spectroscopic phenomenon that is observed when an exo-planet or a star transits across the disk of the observed star. In addition to the photometric transit signal, there is a small spectroscopic signal along with the basic orbital Doppler shift. It was orig-inally predicted as early as by Holt (1893). Later in 1924, Rossiter and McLaughlin simultaneously observed and described this effect, thus designating it as the ’Rossiter-McLaughlin effect.’ Since the RM effect is a phenomenon related to the transit of the planet, I explain the basics of the transit method, which will be useful in modeling the RM effect.
Reloaded Rossiter–McLaughlin
The Reloaded Rossiter-McLaughlin (RRM) technique is similar to DS, where the light blocked be-hind the planet is isolated. However, RRM does not assume any particular shape for the intrinsic line profiles and allows us to directly analyze the local CCF occulted by the planet during transit. Apart from that, both differential rotation and center-to-limb net convective variations can also be taken into account while using the RRM technique.
Cegla et al. (2016a) first presented the RRM technique and applied it to HD 189733. Like DS, the Doppler-reflex motion-induced due to the presence of the planetary companion(s) is removed. Unlike DS, a master is built CCF instead of having a model for the stellar line profile. The master CCF is created by adding all the CCFs taken outside of the transit. This master CCF is then used to align all the CCFs in the stellar rest-frame. The continuum fluxes for in-transit and out-transit CCFs are also scaled considering a quadratic limb darkening model. This allows the direct sub-traction of all the CCFs from the master CCF. The top panel of Figure 2.11 shows the residual time series map of CCFs for WASP-8 b (Bourrier et al., 2017). The center of all the CCFs occurs at » 0 kms¡1 as the orbital and systemic radial velocities are removed. The residual CCF during the tran-sit shows the starlight behind the planet. A Gaussian profile is fitted to the residual CCFs to find the local RVs. The middle panel of Figure 2.11 shows the local RVs against the orbital phase. These local RVs are then fitted with a model to compute ‚. These local RVs are modeled by taking into account the differential stellar rotation. The differential rotation law derived from the Sun (Rein-hold et al., 2013, , Eq. 1) is assumed. In this case, the equatorial regions of the star rotate faster than its poles. Therefore, the stellar rotation velocity vrot? is defined as
vrot? ˘ up ›S sin IS (1 ¡fiz0 ) (2.24).
Challenges in the Radial Velocity Method
The radial velocity method is a powerful method for detecting exoplanets, but it has its limita-tions. It gives no information about the size of the planet. A large low-density planet and a small high-density planet having the same mass will produce the same radial velocity signal. Another limitation is due to the system’s geometry, i.e., the orbital inclination of the planet is unknown. The m sin Ip can, therefore, either correspond to a low-mass planet with high inclination or a high-mass planet or a brown dwarf with low inclination. It is, therefore, important to constrain the orbital inclination to confirm the nature of the planet.
There are other limitations of the RV method that arise due to the atmosphere of the star or due to the presence of a binary companion. The radial velocity of the star is determined by measuring the centroid of the spectral-line profile. The shape of the stellar line profile can be altered due to activity in the stellar atmosphere, which can give rise to variations in radial velocity that can either mask or mimic the planetary signal. It is, therefore, important to secure the planetary nature of the detected signal. It was first reported by Queloz et al. (2001) where they found that the RV variation for HD 166435 was due to starspots on the surface and not due to the gravitational interaction between the star and an orbiting planet. There have been many examples of such effects in the literature, such as HD 219542 (Desidera et al., 2003) and BD +201790 (Figueira et al., 2010).
To circumvent these challenges, various indicators are frequently used to detect signals that are induced by stellar activity or stellar systems. Some of the indicators are discussed below.
Table of contents :
1 Introduction
1.1 History
1.2 Searching for Exoplanets
1.2.1 Transit
1.2.2 Radial Velocity
1.2.3 Direct imaging
1.2.4 Microlensing
1.3 Solar System and Extra-Solar Systems
1.3.1 Solar System
1.3.2 Extra Solar Systems
1.4 Exoplanets
1.4.1 Diversity in Exoplanets
1.4.2 Classification of Exoplanets
1.5 Thesis
1.5.1 Keynote
1.5.2 Outline
1.6 References
2 Detection instruments and techniques
2.1 Detection Instruments
2.1.1 SOPHIE
2.1.2 HARPS-N
2.1.3 SPIRou
2.2 Data and Data Reduction
2.3 DetectionMethod – Radial Velocity
2.3.1 Keplerian Orbit
2.3.2 Model
2.3.3 Keplerian Fitting
2.4 Rossiter–McLaughlin Effect and ObliquityMeasurement
2.4.1 Rossiter–McLaughlin Effect
2.4.2 Obliquity
2.4.3 Classical Rossiter–McLaughlin
2.4.4 Doppler Shadow
2.4.5 Reloaded Rossiter–McLaughlin
2.5 Challenges in the Radial VelocityMethod
2.5.1 Activity Indicators
2.6 References
3 A stellar catalog for giant planet detection
3.1 Catalog definition
3.1.1 Criterion 1 – Volumetric Constraint
3.1.2 Criterion 2- Selection ofMain Sequence stars
3.1.3 Criterion 3 – Removing SB9 binaries
3.1.4 Criterion 4 – Removing CORAVEL binaries and fast rotators
3.1.5 Criterion 5 – Removing targets from KECK survey
3.1.6 Criterion 6 – Removing targets from ELODIE survey
3.1.7 Criterion 7 – Removing known planets
3.2 Comparsion with old Catalog
3.3 Conclusion
3.4 References
4 New Detections
4.1 Motivation
4.1.1 Giant Planets
4.1.2 Brown Dwarfs
4.2 Observations, Data Reduction and Keplerian fit
4.3 New Detections
4.3.1 Giant planets without drifts
4.3.2 Giant planets with drifts
4.3.3 Brown Dwarfs
4.3.4 Stellar Companions
4.4 False Positive Indicator Analysis
4.5 Summary and Conclusion
4.6 References
5 Obliquity Measurements
5.1 Why measure star-planets alignment/misalignment?
5.2 Obliquitymeasurement of a sub-Neptune HD 3167 c
5.2.1 Stellar and planetary parameters
5.2.2 Spin-Orbit (mis)alignment measurement
5.2.3 Results and Conclusion
5.2.4 Additional Information – True Obliquity
5.3 Revisiting obliquity of HD 189733 with SPIRou
5.3.1 Stellar and planetary parameters
5.3.2 Keplerian Orbit of HD 189733 in near infrared
5.3.3 Observation and Analysis
5.3.4 Obliquity measurement using classical RMFit
5.3.5 Obliquity measurement using Doppler Shadow
5.3.6 Results and Conclusion
5.4 Summary
5.5 References
6 Conclusion & Prospects
6.1 References
A Annexes I
