ence of planets is inferred⁸ through the tiny deflections that they impose on passing light rays from background stars. A survey for such events is one of the two main tasks of the proposed WFIRST satellite. Because microlensing is sensitive to planets of all masses having orbits larger than about half of Earth’s, WFIRST would be able to complement and complete the statistical task underway with Kepler, resulting in an unbiased survey of the properties of distant planetary systems. The results from this survey will constrain theoretical models of the formation of planetary systems, enabling extrapolation of current understanding to systems that will still remain below the threshold of detectability.

However, in addition to determining just the planetary statistics, a critical element of the committee’s exoplanet strategy is to continue to build the inventory of planetary systems around specific nearby stars. Therefore, this survey strongly supports a vigorous program of exoplanet science that takes advantage of the observational capabilities that can be achieved from the ground and in space.

The first task on the ground is to improve the precision radial velocity method by which the majority of the close to 500 known exoplanets have been discovered. The measured velocity amplitude of a star depends on the ratio of the planetary to the stellar mass, and on the distance from the star, with a Jupiter-mass body at 5 times the Earth-Sun distance from a Sun-like star producing a 12-meter-per-second signal and an Earth at the Earth-Sun location just a 6-centimeter-per-second signal. Improving the velocity precision will allow researchers to measure the masses of smaller planets orbiting nearby stars. Using existing large ground-based or new dedicated mid-size ground-based telescopes equipped with a new generation of high-resolution spectrometers in the optical and near-infrared, a velocity goal of 10 to 20 centimeters per second is realistic. This could allow detection of bodies twice or three times the mass of Earth around stars the mass of the Sun, and truly Earth-mass planets around stars a factor of two or three less massive than the Sun. The radial velocity technique is also of high value when paired with complementary techniques. For example, transits can determine planet sizes and, in combination with the mass found from another technique, yield clues regarding the bulk planetary compositions—just as we know that Earth is mostly rock and iron from its mass and size and a calculation of the average density. Improved precision astrometry and interferometric techniques that are sensitive to planets at larger separations could not only detect new Jupiter-class planets but also study known planetary systems in combination with radial velocity methods so as to resolve the ambiguity regarding true mass as distinct from the inferred minimum mass.

Success with endeavors to determine the solar neighborhood planetary census will be very important because knowing that Earth-mass planets exist around nearby stars will give much higher confidence that a future space mission to