Methods to Find Small Mass Companions

Several of the search methods for extrasolar planets are similar to those employed to search for eclipsing and other periodic variable stars, but need to be more exacting because of the low amplitudes involved. At present there are five direct and a few indirect methods available in the search for extrasolar planets:

• astrometric variations;

• (direct) imaging/spectroscopy of planets;

• gravitational lensing;

• radial velocity variations;

• indirect effects of planets on (O-C) diagrams of EBs, and on stellar disks such as warps, gaps, and clumps.

5.4.2.1 Astrometry Variations

Periodic and nonlinear proper motions indicate binarity. Astrometric binaries involve low-mass companions detected through proper motion variation. Their usefulness in determining the properties of unseen companions has already been discussed on page 11. The process for extracting the mass, Minv, and barycentric distance of the non-visible, low-mass object is described in Milone & Wilson (2008). If the astro-metric precision is high enough, the method can work for brown dwarfs (defined roughly as having masses between 13 and 75 Jupiter masses (Mj) if they have solar composition and up to 90 if extremely metal-poor, or even for planets (objects with Mmv < 13 Mj).

On the basis of astrometry, Gatewood (1996; 2000) suggested the presence of planets in the systems Lalande 21185 and e Eri, but the former, at least, still requires confirmation. Benedict et al. (2002) have made astrometric measurements of a planet previously detected from radial velocities of stellar reflective motions, Gliese 876b, planet of an M4 dwarf, also known as Ross 780. The measurements were made using the Fine Guidance Sensor on the HST. With radial velocity data, the mass is not in doubt. Together, the data yield an unprojected mass of 1.89 ± 0.34MJ for the planet.

Advanced astrometry space missions may be capable of finding variations due to precise and frequent astrometric measurements. These missions include NASA's SIM (Space Interferometry Mission) and ESA's GAIA, both of which are expected to achieve several micro-arc-seconds of positional precision. SIM is a pointed mission, while GAIA will be a survey instrument. GAIA will be equipped with a radial velocity spectrometer (resolution ~11,200) and from which photometric fluxes may be integrated across any number of photometric passbands. The spectrometer resolution may be insufficient to detect variations due to planets, but the astrometric resolution will be and transits may be detected in the integrated flux.

5.4.2.2 Direct Imaging and Spectroscopy

In both optical and infrared spectral regions, one can look for faint companions to nearby stars, but true planets are rarely likely to be luminous enough to be seen directly, if they are as close to their parent stars as are the "hot Jupiter," typified by 51 Peg b or HD 209458b. It is even more difficult to obtain high spectral resolution to discern identifying features in the spectrum of any such candidate objects. The difficulty is that the overwhelming light of the parent star makes it difficult to separate the flux of the planet from its star. Coronagraphic (Lyot & Marshall 1933) and diffraction techniques are beginning to yield results as new generation instruments come into play, as have high-resolution techniques on existing telescopes. Such techniques include median averaging of rotating fields to produce clean flats for background subtraction. IR surveys are turning up very red and faint objects, and a number of these have been confirmed to be brown dwarfs through subsequent spectroscopy. Some very red objects in clusters are also turning out to be substellar objects (see Basri 2000 for a still useful summary).

In 2004, an apparent companion to the brown-dwarf 2MASS J12073346-3932539 (or "2M1207", for short) was observed in the infrared on the Very Large Telescope (VLT) in Chile. If it is at the same distance as the primary, the companion is 55 AU from the star (Chauvin et al. 2004). Observations made 4 months later with the Near Infrared Spectrometer Camera and Multi-Object Spectrometer (NICMOS) on the HST showed no relative proper motion, and from color indices confirmed its temperature at 1250K. The brown-dwarf candidate is in the TW Hydrae cluster, thought to be only 8 million years old. If it is indeed gravitationally associated with the brown-dwarf, the fainter and cooler object is modeled to have a mass equal to 5Mj. In 2005, a faint companion to the variable star GQ Lupi was observed in visible light; the objects display common proper motion, but is very far from the parent star, and its mass is uncertain within a factor of 10 (Neuhauser et al. 2005), so it could, in fact, be a brown-dwarf star. Most recently, Kalas et al. (2008) has demonstrated orbital motion in HST images of Fomalhaut (a Piscis Austrini). Finally, a system of substellar objects has been imaged by Marois et al. (2008) with the VLT around HR 8799; the estimated masses of these objects are between 7 and 10 MJ, and so may prove to be brown dwarfs. In any case, this is an interesting system.

5.4.2.3 Radial Velocity Variations of the Visible Component

Periodic variations in the Doppler shift of the star as seen in its spectrum are a dead giveaway for something pulling the star around. Because masses of planets are much smaller than those of stars, the orbital motion of the star around the common centre of mass is small also. Therefore high accuracy and precision are required: tens of meters per second or better.

Detection of planets through stellar radial velocity variations has been the major method of detection thus far. The method is illustrated in Fig. 16.3 in Milone & Wilson (2008). Technical improvements in spectrograph stability, in spectral comparison techniques, and in analysis methods have now reduced the uncertainties to ~3 m/s. In addition to detection improvement, long-term averaging of data is beginning to yield long-period, low-amplitude effects in the radial velocity signatures of the parent stars — the effects of planets several AU or more away from the star, in other words, the searches have begun to probe the region occupied by giant planets in our own solar system.

Further improvements to ~1 m/s means investigators must enter a realm dominated by noise effects in the atmospheres of solar analogue stars. Solar-like activity may generate localized velocity variations that will be modulated by both stellar rotation and magnetic activity cycle intervals. The separation of these effects from the effects of multiple low-amplitude planetary periodicities will become a major problem.

5.4.2.4 Gravitational Lensing

The gravitational field of a star causes light from more distant objects lying in nearly the same direction to be bent. Thus the star acts as a lens. The passage of a single star (lens) in front of a more distant one causes varying brightness resulting in two peaks. If the stars are in syzygy with the observer, so that there is an exact match in direction on the sky, an "Einstein ring" is seen instead. From the first detection in 1993, hundreds of events have been seen. Usually there is no consensus of the distance of the star that is acting as the lens, but when the lens turns out to be a binary star, the additional lensing action of the second star and an assumption about the motion of the lensing system in the plane of the sky permit a distance estimate to be made. See Fig. 16.5 in Milone & Wilson (2008), where the crossing by the "caustic" (a surface of maximum brightness created by spherical aberration in a spherical lens/mirror) lasted 8 1/2 h; for comparison, a corresponding event in the galactic halo, 15 Kpc away, would have taken only ~1/2h. If there is a planet in the system, a sharp spike will be seen, in addition to the star's effects.

The ground-based OGLE (for Optical Gravitational Lensing Experiment) survey of the galactic bulge region of the galaxies has now detected some of these. The OGLE program also has detected apparent planetary transits, as we note below.

5.4.2.5 Transit Eclipses

The presence of a planet can be established through an eclipse of the star's light by a planetary transit of the star's disk. This is now a proven technique, but it requires highly precise photometry, relatively small stars and/or large planets, or very long monitoring intervals; selection effects favor planets close to their parent stars with occultations on timescales of hours.

In 2000, radial velocity variation detected with the 1.5-m telescope at the Harvard College Observatory revealed a planetary candidate around the field star HD 209458. D. Charbonneau monitored the star for photometric evidence and, with the help of other observers at Texas and Hawaii, succeeded in observing it (Char-bonneau et al. 2000, Henry et al. 2000). It has subsequently been observed with the HST and limitations on perturbations due to moons, and rings have been established (none have been seen). Subsequent investigation shows the detection of such transits in Hipparcos satellite data. Finally, through a careful analysis of the HST data set, the spectral signature of sodium has been detected from the absorption of the star's light as it passed through the planet's atmosphere (Charbonneau et al. 2002). This is the first such identification!

The OGLE lensing survey has revealed tens of potential transit-like events of very low depth, many of these are repeating, suggesting planetary transits. Several of these have been followed-up with radial velocities studies on large telescopes. Three cases that have proven to be planetary transits are TR-56b, TR-113b, and TR-132b (Konacki et al. 2004). The planets in these systems are even closer to their parent stars than the previously found "hot Jupiters."

More recently, large-field surveys of brighter stars have been revealing transits. The first such detection, of TrES-1, was announced by Alonso et al. (2004). It has an orbit similar to that of HD 209458b, and similar mass, but smaller radius (~ 1.08 Rj).

5.4.2.6 Indirect Effects: O-C Variation

A team headed by E. F. Guinan (Villanova Univ.) claimed detection of one or more planets in the CM Draconis system, an eclipsing M-dwarf binary. The Villanova group claims only one photometric event (which another group disputes) but it has studied the timing of the mutual eclipses and has compiled an O-C (for Observed-Computed instants of mid-eclipse) curve of the eclipsing system, which, Guinan feels, furnishes evidence of the gravitational effect on the orbits of the two stars. At present, a planet in this system remains unconfirmed.

5.4.2.7 Effects on Disks

Protoplanetary disks have been seen around several stars, including p Pictoris and Vega, and remnants of disks have been seen around older stars. Gaps and warping have been attributed to the presence of planets or protoplanets in some of these systems. Gorkavyi et al. (2004) summarize the case for a 10ME planet orbiting p Pictoris, the first star discovered to have a disk around it. Subsequent direct imaging at two epochs has revealed the existence of a planet orbiting in a gap in the disk (see the discussion under Direct Imaging, above).

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment