Publications

The Regolith x-ray Imaging Spectrometer (REXIS) is a coded-aperture soft x-ray imaging instrument on the OSIRIS-REx spacecraft to be launched in 2016. The spacecraft will fly to and orbit the near-Earth asteroid Bennu, while REXIS maps the elemental distribution on the asteroid using x-ray fluorescence. The detector consists of a 2x2 array of back-illuminated 1kX1k frame transfer CCDs with a flight heritage to Suzaku and Chandra. The back surface has a thing p+-doped layer deposited by molecular-beam epitaxy (MBE) for maximum quantum efficiency and energy resolution at low x-ray energies. The CCDs also feature an integrated optical-blocking filter (OBF) to suppress visible and near-infrared light. The OBF is an aluminum film deposited directly on the CCD back surface and is mechanically more robust and less absorptive of x-rays than the conventional free-standing aluminum-coated polymer films. The CCDs have charge transfer inefficiencies of less than 10^-6, and dark current of le-/pixel/second at the REXIS operating temperature of -60 degrees C. The resulting spectral resolution is 115 eV at 2 KeV. The extinction ratio of the filter is ~10^12 at 625 nm.

Utilizing the largest available data sets for the observed taxonomic and albedo distributions of the near-Earth object population, we model the bias-corrected population. Diameter-limited fractional abundances of the taxonomic complexes are A-0.2%; C-10%, D-17%, O-0.5%, Q-14%, R-0.1%, S-22%, U-0.4%, V-1%, X-34%. In a diameter-limited sample, ~30% of the NEO population has jovian Tisserand parameter less than 3, where the D-types and X-types dominate. The large contribution from the X-types is surprising and highlights the need to better understand this group with more albedo measurements. Combining the C, D, and X complexes into a "dark" group and the others into a "bright" group yields a debiased darkto- bright ratio of ~1.6. Overall, the bias-corrected mean albedo for the NEO population is 0.14 +/-0.02, for which an H magnitude of 17.8 +/-0.1 translates to a diameter of 1 km, in close agreement with Morbidelli et al. Coupling this bias corrected taxonomic and albedo model with the H magnitude dependent size distribution of yields a diameter distribution with 1090 +/-180 NEOs with diameters larger than 1 km. As of 2004 June, the Spaceguard Survey has discovered 56% of the NEOs larger than 1 km. Using our size distribution model, and orbital distribution of we calculate the frequency of impacts into the Earth and the Moon. Globally destructive collisions (~10 ^21 J) of asteroids 1 km or larger strike the Earth once every 0.60 +/-0.1 Myr on average. Regionally destructive collisions with impact energy greater than 4 x 10 ^18 J (~200 m diameter) strike the Earth every 56,000 +/-6000 yr. Collisions in the range of the Tunguska event (4-8 x 10^16 J) occur every 2000-3000 yr. These values represent the average time between randomly spaced impacts; actual impacts could occur more or less closely spaced solely by chance. As a verification of these impact rates, the crater production function of Shoemaker et al. has been updated by combining this new population model with a crater formation model to find that the observed crater production function on both the Earth and Moon agrees with the rate of crater production expected from the current population of NEOs.