Assessment of Systematic Differences Between the SPC and SPG Vesta Shape Models Derived from the Dawn Mission

Abstract:

Data from the Dawn spacecraft were used to construct shape models of asteroid 4 Vesta using stereophotogrammetry (SPG) and stereophotoclinometry (SPC) techniques (Jaumann et al., 2012; Gaskell 2012). The purpose of this paper is to identify systematic differences between the models and assess their spatial distribution. Understanding errors in shape models is crucial for confident interpretations in geophysical and geological studies of Vesta.

The two shape models were constructed using the same coordinate system. However, we find a difference between the prime meridians of the two models. Using least squares, we find longitudinal offset as a function of latitude. The offset is less than 0.5 arcmin near the equator and increases poleward. It reaches 1.5 arcmin at 50°N, which is equivalent to a displacement of ≈80 meters on the surface, which represents approximately one pixel in the images that constitute the bulk of the data used in constructing the shape models.

The most distinctive effect is the difference of elevations that resembles the J₂ signal. The SPG model is flatter than the SPC model. The longitude-averaged difference of elevations is minimal at the high southern latitudes (≈25 meters) and reaches ≈175 meters at 15°S. The shape of the longitude-averaged elevation difference resembles the shape of the longitude-averaged radius. This observation alludes to a possible cause of the effect; it may be at least in part a consequence of the difference of image altitudes and therefore image scale. In addition, the mean radius of the two models is different: the SPG’s model mean radius is 118 meters greater than that of the SPC model.

We have also analyzed local scale differences. The largest differences in depths are in high northern latitudes where the surface illumination conditions were poorer. The difference reaches several kilometers in deep craters due to different methods of gap filling.

While this analysis is useful in pointing out inconsistencies between models, it does not and cannot distinguish which model is closer to truth, because image-derived topography lacks an absolute reference. In models where image-derived topography has been compared to topographic observations from laser altimetry, long wavelengths in image-derived models are the most poorly determined.