Remote laser-induced breakdown spectroscopy analysis of East African Rift sedimentary samples under Mars conditions

• Published in: Chemical geology Vol. 294-295; pp. 135 - 151
• Main Authors: Dyar, M.D., Carmosino, M.L., Tucker, J.M., Brown, E.A., Clegg, S.M., Wiens, R.C., Barefield, J.E., Delaney, J.S., Ashley, G.M., Driese, S.G.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 10-02-2012
• Subjects: Bulk analysis; chemical analysis; data collection; East African Rift; Elemental analysis; hydrolysis; igneous rocks; LIBS; Mars; model validation; oxidation; oxides; paleosolic soil types; prediction; Quantitative analysis; sedimentary rocks; spectroscopy; weathering; Mars; East African Rift; LIBS; Bulk analysis; Elemental analysis; Quantitative analysis
• ISSN: 0009-2541; 1872-6836
• DOI: 10.1016/j.chemgeo.2011.11.019

Laser-induced breakdown spectroscopy will be used by the ChemCam instrument on the Curiosity rover to obtain chemical analyses of the martian surface. Surficial and deeper hydrothermal processes on Mars have produced a diverse family of chemical and clastic sedimentary lithologies from primary igneous rocks through physical and chemical transport, deposition, and diagenesis. This study uses 16 samples from the East African Rift (EAR) as martian analogues to assess use of LIBS to evaluate weathering reactions in sedimentary rocks. Data were acquired at 9m distance with samples under conditions designed to simulate Mars. Use of external validation, in which a few sedimentary samples are added to the validation set, coupled with choice of the first local minimum in the root mean square value in all the components of the validation model, provides optimal results in this data set. Accuracy is measured using root mean square error predictions for major elements in the sedimentary rocks, as expressed in wt.% oxides. Even lower errors can be obtained by using a more focused training set. These results are sufficiently accurate to usefully characterize the four competing kinds of chemical weathering reactions in paleosols (hydrolysis, oxidation, hydration, and salinization).
►East African Rift samples were analyzed with LIBS to evaluate weathering in sedimentary rocks. ►Partial least square techniques were used to predict errors on rock analyses. ►External validation and the first local minimum in RMSEP provide optimal results. ►Results accurately characterize four chemical weathering reactions in paleosols.