A Deep Search for Planets with the Keck Planet Finder

California Institute Of Technology, Pasadena CA

Investigators

Abstract

Starting with the Nobel-Prize winning detection of a “hot Jupiter” in 1995, the Doppler technique based on stellar spectra has been a ground-breaking method for finding planets around stars other than the Sun. This and other techniques have indicated that most stars have planets with masses of just a few Earth masses – small enough to have rocky surfaces – with orbital periods of ~1-300 days. Direct imaging space missions may one day record the light of these planets, but techniques to measure such small masses still need maturation. In this project, researchers in the California Planet Search collaboration will survey nearby stars for these low-mass planets, with the Keck Planet Finder (KPF) spectrometer at Keck Observatory. These data will almost surely detect new planets, and they will also allow researchers to learn how to reach Doppler precisions of 10 centimeters-per-second using the spectral features, which is necessary to detect Earth analogues. The project will support student researchers, as well as run an Introduction to Astronomy Research program for less privileged students, allowing greater access to the research skills students need for astronomy or other STEM fields. KPF is a high-resolution, fiber-fed spectrometer on the Keck I telescope, with high thermomechanical and optical stability for precise radial velocity measurements. Its advanced technology includes a laser frequency comb for calibration, exotic materials to reduce thermal sensitivity, and a separate UV spectrometer for stellar activity tracking. Early results demonstrate KPF's high throughput, efficient operation, and impressive Doppler stability, achieving the precision needed for this survey of ten nearby G and K-type stars. Each star will be observed approximately 100 times over three years with KPF, with an expected limiting detection amplitude on month-timescale orbital periods of 50 centimeters per second. To achieve this goal, the team will develop new algorithms to analyze the stellar spectra in a line-by-line approach. Discoveries of new systems of small planets will teach us about the formation and evolution of planets like our own and will be the best targets nearly all future studies of nearby stars. The survey spectra will have SNR = 600-1600 per reduced pixel, a new regime of precision. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.