The Effects of Misregistered Topography on Lunar Thermal Modeling
Abstract
There has been significant debate over the existence of OH/H2O on the surface of the Moon since the discovery of a 3 μm absorption feature. Many independent studies have been conducted to understand the distribution of the 3 μm feature and its variability. Currently the Moon Mineralogy Mapper (M3) has the most complete spatial coverage of the lunar surface with spectral data out to 3 μm. Interpreting reflectance spectra longward of 2 μm is challenging because of thermal emission from the surface. We model the temperature contribution to M3 data following Bandfield et al. (2018), but this requires precise viewing geometry which in turn requires accurate registration to topography. The M3 data set is not precisely aligned to the LROC global control network, introducing bias to our thermal correction near steep topographic slopes. To account for this offset, we select ground control points (GCPs) tying the M3 data to the LROC base map. The tie-points are then used to warp the image to reduce alignment errors and more accurately register the spectral information from M3 to the lunar topography. We conduct an empirical study to quantify any residual offsets.
We compare three different algorithms (1st order polynomial, 2nd order polynomial, thin plate spline) that warp the M3 images. We sample sharp features from the base map and measure offsets from these features to their new locations in the warped images. We compute the mean, quartiles, and standard deviation of the offsets and test whether they correlate with latitude, longitude, and control points. Initial results show that thin plate spline produces the smallest residuals compared to the other algorithms. We observe a slight correlation between offsets and high latitudes (poleward of 60°), indicating that this technique may be more accurate at low latitudes. Since the selection of GCPs contains inherent observational bias, we also constrain the average offsets between multiple mappers. We compare the offsets in images warped using the GCPs of co-authors by making offsets between them. We present a robust analysis of how hand selected GCPs can accurately register planetary data sets. We also present initial results constraining the variability in a lunar thermal model due these offsets. We interpret the 3 μm feature in the companion abstract by Ruiz et al. (this conference).- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMP079.0015G
- Keywords:
-
- 1060 Planetary geochemistry;
- GEOCHEMISTRY;
- 1094 Instruments and techniques;
- GEOCHEMISTRY;
- 3934 Optical;
- infrared;
- and Raman spectroscopy;
- MINERAL PHYSICS;
- 5410 Composition;
- PLANETARY SCIENCES: SOLID SURFACE PLANETS