The multiple planet systems include two planets detected around the star mu Ara (in the constellation of Ara "The Altar"). The inner planet has an orbital period of 645 days and a minimum mass of 1.7 Jupiter masses. The outer planet has an orbital period of 8.2 years and a minimum mass of 3.1 Jupiter masses. Both planets have quite eccentric (ie non-circular) orbits. These two planets were recently "joined" by a third inner (but very much smaller) planet in a 9 day orbit announced by Santos et al.

The three low-mass planets have all been detected with orbital periods of between 26 and 129 days, and minimum masses of between 0.16 and 0.4 times that of Jupiter. These low-mass planets are exciting to the Anglo-Australian Planet Search team because they all have small velocity amplitudes -- that is the represent the detection of quite small "wobbles" in the parent stars due to these planets. Indeed at just 12 to 18m/s these results obtained from data streams stretching back to 1998, represent exactly the levels of precision that our search needs to attain to detect Solar Systems like our own around other stars via the orbital motion of a Jupiter-like planet in a Jupiter-like 12 year orbit. They give us confidence that in the next 6 years, if there are any "Solar System-like" systems amongst our 240 targets stars, we will find them.

How it works

A Jupiter-like planet exerts a small gravitational pull on its parent star, causing the star to wobble. The velocity of this wobble depends on the distance at which the planet orbits, and the mass of the planet. For typical giant planets the velocity variation is in the range 1 to 100 m/s.

This motion can be detected via the Doppler Effect. When the unseen planet is moving away from the Earth, the star will move slightly towards the Earth. The light emitted by a star when it is doing this is Doppler shifted to shorter (bluer) wavelengths. The reverse happens when the unseen planet moves towards the Earth -- the star moves away, and the light it emits is shifted to longer (redder) wavelengths. Because of the small velocities involved, the effect is subtle -- it doesn't effect the apparent colour of the star, for example. But it can be detected by very high precision astronomical instruments like the University College London Echelle Spectrograph (or UCLES) on the AAT.

A star with a Jupiter-mass planet will be revealed by the periodic Doppler shift of its light. After one or two orbital periods the information from the Doppler measurements allows us to calculate the orbit and mass of the unseen planet. Our current measurement precision is 3 meters per second (a brisk walk). For comparison, Jupiter causes the Sun to wobble with a velocity of 12.5 meters per second over a 12 year period. Saturn induces a 2.7 meter per second wobble on the Sun with a 30 year period. The other planets (in particular the terrestrial planets like the Earth, Mars and Venus) are far too small to produce a measureable effect on the Sun.

The project

AAPS has been operating since January 1998, and is expected to run until 2010, at which point we will have observed for long enough to detect Jupiter-like planets in Jupiter-like orbits around other stars. We are currently monitoring the 200 nearest and brightest Sun-like stars visible from the AAT's Southern Hemisphere location on 20 nights per year. We perform these observations using the University College London Echelle Spectrograph (UCLES). UCLES enables us to observe almost the entire visible spectrum in a single observation. Doppler shifts in the stellar spectra are measured with reference to a precision calibrated iodine vapour absorption cell (ike that shown to the left). The absorption that produces the faint purple colour of the iodine gas in this cell, imprints a dense network of narrow lines on our spectra, telling us everything we need to know about UCLES's performance. The only example of a star with planets that know in much detail is our own Solar System. Ultimately we need to know what fraction of Sun-like stars have Jupiter- and Saturn-mass planets in Jupiter- and Saturn-like orbits. In other words, what fraction of extra-solar planetary systems are similar to our own?

The Anglo-Australian Planet Search Team

• R. Paul Butler (Carnegie Institution of Washington)
• Chris Tinney (Anglo-Australian Observatory)
• Hugh Jones (Liverpool John Moores University)
• Geoff Marcy (University of California Berkeley, San Francisco State University)
• Chris McCarthy (San Francisco State University)
• Alan Penny (Rutherford Appleton Laboratory)
• Brad Carter (University of Southern Queensland)

[NOTE: Images supporting this release are available at http://www.aao.gov.au/local/www/cgt/planet/index.html ]

From: Chris McCarthy [view email]
Date: Tue, 14 Sep 2004 13:11:05 GMT   (71kb)

Precise Doppler measurements from the AAT/UCLES spectrometer reveal two companions to both HD 154857 and HD 160691. The inner companion to HD 154857 has a period of 398 d, an eccentricity of 0.51, and a minimum mass of 1.8 M_Jupiter An outer companion has a period much longer than 2 years and is currently detected only as a linear trend of 14 m/s per year. The inner companion to HD 160691, previously announced from AAT data, has a period of 645 d, an eccentricity of 0.20, and a minimum mass of 1.7 M_Jupiter. For the outer planet, whose orbit is less well constrained, a two Keplerian fit yields a period of 8.2 yr, an eccentricity of 0.57, and a minimum mass of 3.1 M_Jupiter. With these orbital parameters, its maximum separation from the star of 0.4 arcsec makes it a viable target for direct imaging.

References and citations for this submission:
SLAC-SPIRES HEP (refers to , cited by, arXiv reformatted);
CiteBase (autonomous citation navigation and analysis)

Which authors of this paper are endorsers?