Analysis of Projection Effects in OSIRIS-REx Spectral Mapping Methods
Description
We search for an optimized mapping protocol and explore the characteristics of mapping remote sensing observations from a point spectrometer onto a shape model comprised of triangular facets - in the context of NASA's Asteroid Sample Return Mission: OSIRIS-REx. The shape model used was created from ground based radar observations by \citet{nolan2013shape}. Our study is conducted before the spacecraft arrives at Bennu and we use observational sequence plans of the OSIRIS-REx Visible and Infrared Spectrometer (OVIRS). We explore 6 methods of mapping data to the shape model facets, and using 3 different shape model spatial resolutions. We also attempt to increase map fidelity by increasing the observational coverage of the surface. We find that: (A) increasing shape model resolution increases mapping quality. However, once the shape model's mean facet edge length is smaller than about two fifths of the diameter of the instrument's field of view (FOV - or ``spot" size), the increase in quality tapers off and may not be worth the additional computation time. (B) The 6 mapping methods we have conceived can be broken into two categories: 1) selecting or 2) averaging observations when assigning values to each facet of the shape model. The quality differences between similar methods that average values from different spots together are insignificant, i.e. averaging, taking the median, or using the weighted average. Selecting the nearest observation to a facet best preserves an enclosed outcrop shape and signal, but averaging spots is more conservative against errors in spot positioning or signal. (C) A completely enclosed outcrop border expands into the surrounding region by 0.8-1.5 radii of the instrument's FOV after mapping. (D) Regions smaller than the instrument's FOV are present (detectable) in resulting maps, however their signal strength is reduced as a function of their size relative to the instrument FOV. We expected the sub-FOV detectability to depend on whether the contrast between the signal of the sub-FOV region and the surrounding area is greater than the instrument uncertainty. Lastly, (E) we have demonstrated that mapping quality can be improved by increasing the number of observations that cover the surface. For OSIRIS-REx, increasing the observation count requires adding together observations from the 3 dayside surveys that were obtained at different phase angles. Uncertainties due to photometric correction of these observations to a standard viewing and illumination geometry can result in systematic differences that result in map artifacts. We suggest that if combining data sets from different phase angles while mapping, averaging the spots together can help to de-emphasize any small systematic differences.