Mapping Vesta using a hybrid method for incorporating spectroscopic and morphologic data

Defining criteria for mapping material units on airless, rocky bodies is challenging. Where the primary geologic process for most of a body’s history is impact cratering, traditional morphology-based mapping approaches may fail, because differences in morphologic characteristics among the various cratered surfaces can be hard to discern, and surface morphology is muted by the regolith’s physical and mechanical properties. In constructing a global geologic map of Vesta at 1:300,000-scale using the Dawn Framing Camera (FC), DTM-derived slope and contour, and multispectral data, we have countered this problem by utilizing a hybrid method of mapping that first requires creating two maps independently. The first map depends on morphology and topography to define map units, while the second uses spectral data to define units. The unique results of each map are then combined into the hybrid map units. 

 

Multispectral data provide unique insight into stratigraphy (material brought up through cratering processes) that is easily lost when using an albedo mosaic as the basemap. However, solely using a “color” ratio mosaic as a basemap easily magnifies potentially misleading data, because spectroscopy in the shorter wavelengths (UV-VIS-near IR) can only sample the upper few µm of the surface, and very little unique material is required to affect the signal of a regolith. Contacts defined by multispectral data may not coincide with clear morphologic boundaries as a result, so caution must be used in how the two maps are merged and clear criteria should be established to define hybrid map units.

 

We found that the crucial exercise in ensuring unique data were retained when combining these two maps was to create a decision tree for determining which data would be primary in choosing where to draw unit boundaries. We divided the decision tree into the following if-then statements:

• If saturated colors (meaning the color signal in color-ratio spectral data was strong and the color itself was easy to describe) matched unit boundaries derived from morphology, there was no conflict. For example, saturated colors on Vesta tend to be associated with fresher expressions or exposures of regolith, which are more likely found at the youngest, freshest craters/ejecta, easily demarcated morphologically.
• If muted colors exist, where the morphology is relatively clear, the morphology is the primary guide for unit definition, as it retains the least altered record of geologic processes and the most reliable record of the nature of the rock bodies. Colors provide additional characteristics of such units, allowing for some interpretation of composition.
• If saturated colors are not associated with morphologic boundaries, the color boundaries are interpreted to record the most recent (even if very thin) impact evidence. In such cases we have mapped the saturated color data as impact material. This preserves the underlying morphology/topography information while supporting stratigraphic interpretations based on excavated subsurface layers revealed by crater ejecta.
• In the case of muted colors where the morphology is unclear, decisions must be made case-by-case, using all available data to make a reasonable determination of where to mark unit boundaries.