Dynamic aperture factor analysis/target transformation (DAFA/TT) for serpentine and mg-carbonate mapping on Mars with CRISM near-infrared data

1,2Honglei Lin,3J.D.Tarnas,3J.F.Mustard,1Xia Zhang,2Yong Wei,
2Weixing Wan,4F.Klein,5J.R.Kellner
Icarus (in Press) Link to Article [https://doi.org/10.1016/j.icarus.2020.114168]
1Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China
2Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing. 100029, China
3Dept. of Earth, Environmental, and Planetary Sciences, Brown University, RI 02912, The United States of America
4Dept. of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, MA 02543, The United States of America
5Institute at Brown for Environment and Society, Brown University, RI 02912, The United States of America
Copyright Elsevier

Serpentine and carbonate are products of serpentinization and carbonation processes on Earth, Mars, and other celestial bodies. Their presence implies that localized habitable environments may have existed on ancient Mars. Factor Analysis and Target Transformation (FATT) techniques have been applied to hyperspectral data from the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) to identify possible serpentine and Mg-carbonate-bearing outcrops. FATT techniques are capable of suggesting the presence of individual spectral signals in complex spectral mixtures. Applications of FATT techniques to CRISM data thus far only evaluate whether an entire analyzed image (≈3 × 105 pixels) may contain spectral information consistent with a specific mineral of interest. The spatial distribution of spectral signal from the possible mineral is not determined, making it difficult to validate a reported detection and also to understand the geologic context of any purported detections. We developed a method called Dynamic Aperture Factor Analysis/Target Transformation (DAFA/TT) to highlight the locations in a CRISM observation (or any similar laboratory or remotely acquired data set) most likely to contain spectra of specific minerals of interest. DAFA/TT determines the locations of possible target mineral spectral signals within hyperspectral images by performing FATT in small moving windows with different geometries, and only accepting pixels with positive detections in all cluster geometries as possible detections. DAFA/TT was applied to a hyperspectral image of a serpentinite from Oman for validation testing in a simplified laboratory setting. The mineral distribution determined by DAFA/TT application to the laboratory hyperspectral image was consistent with Raman analysis of the serpentinite sample. DAFA/TT also successfully mapped the spatial distribution of serpentine and Mg-carbonate previously detected in CRISM data using band parameter mapping and extraction of ratioed spectra. We applied DAFA/TT to CRISM images in some olivine-rich regions of Mars to characterize the spatial distribution of serpentine and magnesite outcrops.


Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s