Corascope: Software assisted XRF scan merging

Abstract

Due to the physical constraints of X-ray Fluorescence (XRF) scanners, all but the shortest sedimentary cores are typically scanned in a piecewise fashion. Pieces are either extracted a section at a time (e.g. Russian peat corer) or are cut and sectioned after extraction (e.g. piston and gravity cores). The XRF core scanner generates an optical line-scan image, an X-ray radiograph, and a table of downcore variations in elemental composition. The subsequent processes of digitally joining the individual section scans so that they correspond to the original core is typically done in an ad-hoc manner with little visual feedback. Manual manipulation of composition tables while maintaining proper alignment with the optical scans is error prone and slow. As a result, image data are typically either discarded or used as a rough reference. This means a potentially valuable data source is normally lost. Corascope is a first-pass data reconstruction tool for the Itrax XRF scanner that aims to streamline data merging before doing any analysis. Immediate visual feedback is provided during the cropping and merging of core sections; it automatically preserves optical and elemental data alignment. It further provides a statistically-guided set of tools for adjusting overlapping core sections. These allow users to go from several overlapped cores to one merged virtual core. This virtual core then provides a reconstruction of the sedimentary record as one continuous piece. In line with the goal of being a first-pass tool, Corascope leaves all data values unperturbed and merged core data are returned in the same format as generated by the Itrax XRF scanner. This ensures that preexisting analysis software and laboratory procedures can continue to work on preprocessed data.