HR: 0800h
AN: P21A-0222    [Abstracts]
TI: Comparisons of Thermal Emission Spectroscopy and Laser-Ablation Inductively Coupled Plasma Mass Spectrometry Derived Bulk Chemistries for Natural Surfaces of Volcanic Rocks
AU: * Neal, K
EM: kevin_neal@brown.edu
AF: Department of Geological Sciences, Brown University, Providence, RI 02912, United States
AU: Wyatt, M
EM: michael_wyatt@brown.edu
AF: Department of Geological Sciences, Brown University, Providence, RI 02912, United States
AU: Kelley, K
EM: kelley@gso.uri.edu
AF: Graduate School of Oceanography, University of Rhode Island, Narrgansett, RI 02882, United States
AB: The GRS and TES datasets provide unique and complementary insights into the bulk compositions of martian surface materials. GRS measures the composition of the upper few tens of centimeters of the surface while TES measures the composition of the upper hundred microns. Recent GRS studies have reported global distributions of bulk chemical abundances for Si, K, Fe, Th, and K/Th (Karunatillake et al., 2007; Taylor et al., 2007). A major result from these studies is the near uniform distribution of relative Si concentrations across martian low-albedo regions. This is significant because a primary difference between the TES Surface Type 1 and Surface Type 2 global spectral end-members is the modeled abundance of high-silica mineral-phases (e.g. Bandfield et al., 2000; Wyatt et al., 2002; McSween et al., 2003). A way of reconciling this apparent discrepancy between the two datasets is to account for the different penetration depths of the two techniques and attribute compositional variations to near-surface coatings of high-silica phases. This, however, leads to another question involving the accuracy of linear deconvolution results for materials with surface coatings. Laboratory studies by Rampe et al. (2007) reveal that secondary alteration products on a surface can cause non- linear mixing between spectral end-members and thus affect the modeled abundances of primary minerals in a mixture. This is significant because small amounts of weathering products on the martian surface may thus result in modeled primary mineral abundances that differ from the actual composition, resulting in a possible misclassification of volcanic compositions. In this study, we further the work by Rampe et al. (2007) and examine the accuracy of bulk chemical oxide abundances derived from thermal infrared laboratory measurements of chemically weathered natural surfaces and fresh cut surfaces of basalt. TES has mainly been utilized as a mineralogical tool, but thermal emission spectroscopy also provides a means for deriving chemical oxide abundances. Chemical compositions can be calculated from deconvolved modal mineralogies (vol. %) by combining the compositions of the spectral endmembers (wt. % oxides) in proportion to their relative modeled abundances. Wyatt et al. (2001) quantified the uncertainties in derived chemical compositions for fresh cut surfaces of volcanic rocks and demonstrated their use in correctly classifying compositions based on total silica contents. This study will specifically examine natural surfaces of volcanic rocks and whether non-linear mixing of mineral-phase spectral endmembers, which can adversely affect modeled mineral abundances, also affects derived chemical oxide abundances or if substitution of chemically similar phases results in consistent values. We analyze the bulk chemistries of basaltic natural surfaces using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) and electro microprobe (EMP) and compare bulk chemistries to values derived from thermal emission data.
DE: 5410 Composition (1060, 3672)
DE: 5415 Erosion and weathering
DE: 5455 Origin and evolution
DE: 5464 Remote sensing
SC: Planetary Sciences [P]
MN: 2007 Fall Meeting