Mathematics – Logic
Scientific paper
Dec 2006
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2006agufm.p23c0074h&link_type=abstract
American Geophysical Union, Fall Meeting 2006, abstract #P23C-0074
Mathematics
Logic
5464 Remote Sensing, 5470 Surface Materials And Properties, 5494 Instruments And Techniques
Scientific paper
Remotely sensed mineralogy will likely play a large role in the selection of future Mars landed science locations, and with the success of the OMEGA experiment on Mars Express (identification of sulfate, phyllosilicate minerals), the data from CRISM on MRO will also be important. One of the crucial challenges in this activity is how to identify promising targets from spatially coarse data for in situ investigation. We present an approach for systematic integration of data from remote sensing to rover scales. We incorporate data from the sample, to in situ, to regional remotely sensed data. We test the applicability of this approach to Mars using VNIR field- and remotely sensed spectra from Rio Tinto, Spain. The Rio Tinto has a diverse iron oxide and hydrous sulfate mineralogy in a very acidic environment and is considered a mineralogic and biologic analog for Mars, especially Terra Meridiani. We use remotely sensed Hymap data (8 m per pixel) to predict small areas with high mineral diversity, and then verify the identified locales through field work at the sample and in situ scales. We apply the Spectral Variance Index (SVI), a method which blends spatial and spectral components, to locate regions with increased mineral diversity. Remotely acquired data from the Hymap spectrometer (8 m per pixel) are binned into 25x25 pixel (or 200m x 200m) cells. We compute their mean and variance by wavelength and their expected spectral variance due to their albedo. Cells with a spectral variance significantly higher than their expected variance have more spectra variability than average, and thus are likely to have more mineralogic diversity. While this technique identifies areas of mineral diversity, it is less able to identify unique mineralogy. To catalog the mineralogy present at the Rio Tinto, we perform a Minimum Noise Function (MNF) followed by a Pixel Purity Index (PPI) function on the Hymap data. Pixels that are flagged by this process are unique in the scene and tend to represent pure mineralogy. Minerals thus identified include goethite, hematite, gypsum, hydrated Fe3+ sulfates including copiapite, hydrated Fe2+ sulfates including rozenite and melanterite, and various clay minerals. The SVI approach identifies areas of high mineralogic diversity in the Rio Tinto, and could also be applied to Mars orbital spectral datasets, such as OMEGA and CRISM. This exploration strategy for iron oxide and sulfate-rich environments over a range of spatial scales could allow us to pinpoint Rio Tinto-like deposits on Mars, a useful ability in the search for future landing sites of mineralogic and astrobiologic interest.
Amils Ricardo
Fernández-Remolar David
Hutchison L. E.
Mustard John F.
No associations
LandOfFree
Quantifying mineralogic diversity from remotely sensed VNIR spectra to identify mineralogic hotspots does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Quantifying mineralogic diversity from remotely sensed VNIR spectra to identify mineralogic hotspots, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Quantifying mineralogic diversity from remotely sensed VNIR spectra to identify mineralogic hotspots will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-961530