Previous
abstract 
Next
abstract
We describe stellar Doppler measurements with a precision of 3$~$m$~$s$^{\rm -1}$, cabable of detecting extra-solar planets having masses less than one Jupiter-mass. No planetary systems around main sequence stars have yet been detected, nor does there exist a compelling detection of a brown dwarf, as a stellar companion or otherwise. Several groups are engaged in detecting planetary systems by using precise radial velocities to detect the wobble of the host star (cf. Campbell and Walker, 1988, ApJ,331 ,902 ; Cochran and Hatzes (1994, in ``Planetary Systems: Formation, Evolution, and Detection'', Kluwer Acad.).
Among current search efforts, the Lick Observatory Planetary Search is extensive in both sample size and span of spectra types (Marcy and Butler, 1992, PASP,104 ,270). The project consists of a Doppler survey of 100 main sequence stars of spectral type, F, G, K, and M. The Lick Obs. full--format echelle gathers all Doppler information from 5000 to 5800 Ang.
The wavelength calibration is accomplished with a gaseous iodine absorption cell placed at the slit of the Hamilton spectrometer which superimposes sharp $I_2$ lines on the stellar spectrum. The $I_2$ lines also provide the PSF of the spectrometer. A great breakthrough has occurred in the Doppler precision as a result of the improved Schmidt camera on the Hamilton. The Lick echelle now boasts a PSF with FWHM = 1.25 pixels for a narrow (0.3 arcsec) slit.
The Doppler precision achieved is 3$~$m$~$s$^{\rm -1}$, based on velocity scatter during the past 5 months in our standard star, $\tau$ Ceti (G8V) . The velocities exhibit a standard deviation of 4 $~$m$~$s$^{\rm -1}$ per exposure. However, we routinely obtain 4 quick exposures of the brightest target stars on our survey and average the 4 velocities. Such averaging for $\tau$ Ceti velocities yields a scatter of 3 $~$m$~$s$^{\rm -1}$ due to errors in our Doppler measurements. For comparison, our Jupiter perturbs the Sun by 12.5 $~$m$~$s$^{\rm -1}$ rendering Jupiter--like planets detectable at the 4--$\sigma$ level, even for typical orbital inclinations. We also report no detections of Jupiter-like planets for 30 main sequence stars, based on 7 years of past Doppler measurements at lower precision of 10 $~$m$~$s$^{\rm -1}$. We expect to achieve precision of $\sim$ 2 $~$m$~$s$^{\rm -1}$ by improving the Doppler Analysis.
