--> Can One Use Trace Element Concentrations, Determined by XRF, to Identify Santa Barbara Source Formations of Rock Samples?

AAPG Pacific Section Convention, 2020 Vision: Producing the Future:

Datapages, Inc.Print this page

Can One Use Trace Element Concentrations, Determined by XRF, to Identify Santa Barbara Source Formations of Rock Samples?

Abstract

Seasonal storms and floods erode sandstone boulders and cobbles from the mountains behind Santa Barbara, transport them, and eventually deposit them in stream beds, on flood plains and along the beaches. Can one use trace elements, analyzed using X-Ray Fluorescence, XRF, to identify the source formation of rock samples? The main formations supplying the boulders are the Coldwater, Matilija, and the Vaqueros sandstones. The author used a geologic map of the Santa Barbara Quadrangle to drive to road cuts having outcrops of the above listed formations to collect hand samples at a score of locations. These samples were for establishing the compositional baseline and variability of known formations. The author pried these samples from the decaying bedrock to be sure of their provenance. To determine the compositions of the samples, the author used a Bruker hand held X-ray fluorescence spectrometer belonging to the Earth Science Department of the Santa Barbara City College. Repeat measurements without moving the samples allowed estimation of the measurement variability for each element detected. Repeat measurements of various faces of the same sample allowed estimation of the variability within hand samples. Pooling the results from each formation allowed estimation of the variation between formations and established the expected range of compositions of samples from each formation. To analyze the data, the author used MATLAB to make plots of the concentrations of elements with the highest signal to noise ratios and made x-y plots of the first four principle components. The first four principle components were able to explain over 95% of the variability in the measurements. The results for the Coldwater and Matilija sand stones largely overlap, so one could only assign unknown samples to the correct source formation in a minority of cases. Principle components of samples from the Vaqueros Formation hardly overlapped with the principle components of the Coldwater and Matilija Formation samples, so one could confidently separate samples from the Vaqueros Formation from the others.