-
Radial Velocity Techniques for Exoplanets
- University of Arizona Press
- Chapter
- Additional Information
1. INTRODUCTION Since the end of the nineteenth century, radial velocities have been at the heart of many developments and advances in astrophysics. In 1888, Vogel at Potsdam used photography to demonstrate Christian Doppler’s 1842 theory that stars in motion along our line of site would exhibit a change in color. This color change, or wavelength shift, is commonly known as a Doppler shift, and has been a powerful tool over the past century, used to measure stellar kinematics, determine orbital parameters for stellar binary systems, and identify stellar pulsations. By 1953, radial velocities had been compiled for more than 15,000 stars in the General Catalogue of Stellar Radial Velocities (Wilson, 1953) with a typical precision of 750 m s−1, not the precision that is typically associated with planet-hunting. However, at that time, O. Struve proposed that high-precision stellar radial velocity work could be used to search for planets orbiting nearby stars. He made the remarkable assertion that Jupiterlike planets could reside as close as 0.02 AU from their host stars. Furthermore, he noted that if such close-in planets were 10 times the mass of Jupiter, the reflex stellar velocity for an edge-on orbit would be about 2 km s−1 and detectable with 1950s Doppler precision (Struve, 1952). Two decades later, Griffin and Griffin (1973) identified a key weakness in radial velocity techniques of the day; the stellar spectrum was measured with respect to an emission spectrum. However , the calibrating lamps (typically thorium argon) did not illuminate the slit and spectrometer collimator in the same way as the star. Griffin and Griffin outlined a strategy for improving Doppler precision to a remarkable 10 m s−1 by differentially measuring stellar line shifts with respect to telluric lines. Assuming that telluric lines are at rest relative to the spectrometer, these absorption lines would trace the stellar light path and illuminate the optics in the same way and at the same time as the star. Although Griffin and Griffin did not obtain this high precision, they had highlighted some of the key challenges that current techniques have overcome. By 1979, G. Walker and B. Campbell had a version of telluric lines in a bottle: a glass cell containing hydrogen fluoride (HF) that was inserted in the light path at the Canada France Hawaii Telescope (CFHT) (Campbell and Walker, 1979). Like telluric lines, the HF absorption lines were imprinted in the stellar spectrum and provided a precise wavelength solution spanning about 50 Å. The spectrum was recorded with a photon-counting Reticon photodiode array. Working from 1980 to 1992, they monitored 17 main-sequence stars and 4 subgiant stars and achieved the unprecedented precision of 15 m s−1. Unfortunately, because of the small sample size, no planets were found. However, upper limits were set on M sin i for orbital periods out to 15 years for the 21 stars that they observed (Walker et al., 1995). Cross-correlation speedometers were also used to measure radial velocities relative to a stellar template. In 1989, an object with M sin i of 11 MJup was discovered in an 84-day orbit around HD 114762 (Latham et al., 1989). The velocity amplitude of the star was 600 m s−1 and although the single measurement radial velocity precision was only about 400 m s−1, hundreds of observations effectively beat down the noise to permit this first detection of a substellar object. In 1993, the ELODIE spectrometer was commissioned on the 1.93-m telescope at Observatoire de Haute-Provence 27 Radial Velocity Techniques for Exoplanets Christophe Lovis Université de Genève Debra A. Fischer Yale University The radial velocity technique was utilized to make the first exoplanet discoveries around Sun-like stars and continues to play a major role in the discovery and characterization of exoplanetary systems. In this chapter we describe how the technique works, and the current precision and limitations. We then review its major successes in the field of exoplanets. With more than 250 planet detections, it is the most prolific technique to date and has led to many milestone discoveries, such as hot Jupiters, multiplanet systems, transiting planets around bright stars, the planet-metallicity correlation, planets around M dwarfs and intermediate-mass stars, and recently, the emergence of a population of low-mass planets: Neptune-mass planets and super Earths. In the near future radial velocities are expected to systematically explore the domain of rocky and icy planets down to a few...