2015 GSA Annual Meeting in Baltimore, Maryland, USA (1-4 November 2015)

Paper No. 206-22
Presentation Time: 9:00 AM-6:30 PM


SHAVERS, Ethan, Center for Sustainability & Department of Earth and Atmospheric Sciences, Saint Louis University, St.Louis, MO 63108, GHULAM, Abduwasit, Center for Sustainability, Saint Louis University, St. Louis, MO 63108 and ENCARNACION, John, Department of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO 63108, eshaver1@slu.edu

Enrichment of some trace elements is common in melilitite-carbonatite complexes, sometimes reaching economically-viable levels. VNIR-SWIR spectroscopy is an important means of mineral exploration and lithologic mapping. Here we employ chemometric modeling using laboratory spectral measurements and geochemical analyses of samples from the Avon Volcanic District (Missouri, USA) to determine whether trace element enrichment can be remotely detected in melilitite-carbonatite intrusive complexes. The Avon Volcanic District is a late Devonian intrusive complex with more than 80 dikes and diatremes grading from calciocarbonatite to olivine melilitite. The majority of reflectance features characteristic of these rocks are due to vibration of the CO3 radical and its lattice along with ferrous and ferric iron electron transitions. In addition, many trace elements are susceptible to electron transitions that cause sharp absorption features in the Vis-NIR region with detection limits below 500 ppm. The elements analyzed in this study are: Cu, Pb, Zn, Ag, Ni, Co, Sr, V, La, Cr, Ba, Zr, Ce, Nd, Rb and Nb. Initial modeling using 101 spectra from 12 different samples suggests that Ni (39-630 ppm) and Cr (67-1290 ppm) have the highest correlation with spectral features while Nd (18-110 ppm) and Ba (57-1367 ppm) also show potential. The R2 values of 0.77 for Ni and 0.76 for Cr were obtained using principle components regression analysis with the spectral range limited to 400-1100 nm. Further analysis is needed to understand the role these elements play in shaping the spectral response and model adjustments can likely improve trace element quantitative detection.