Pre-flight calibration and initial data processing for the ChemCam laser-induced breakdown spectroscopy instrument on the Mars Science Laboratory rover

The ChemCam instrument package on the Mars Science Laboratory rover, Curiosity, is the first planetary science instrument to employ laser-induced breakdown spectroscopy (LIBS) to determine the compositions of geological samples on another planet. Pre-processing of the spectra involves subtracting the ambient light background, removing noise, removing the electron continuum, calibrating for the wavelength, correcting for the variable distance to the target, and applying a wavelength-dependent correction for the instrument response. Further processing of the data uses multivariate and univariate comparisons with a LIBS spectral library developed prior to launch as well as comparisons with several on-board standards post-landing. The level-2 data products include semi-quantitative abundances derived from partial least squares regression. A LIBS spectral library was developed using 69 rock standards in the form of pressed powder disks, glasses, and ceramics to minimize heterogeneity on the scale of the observation (350–550 μm dia.). The standards covered typical compositional ranges of igneous materials and also included sulfates, carbonates, and phyllosilicates. The provenance and elemental and mineralogical compositions of these standards are described. Spectral characteristics of this data set are presented, including the size distribution and integrated irradiances of the plasmas, and a proxy for plasma temperature as a function of distance from the instrument. Two laboratory-based clones of ChemCam reside in Los Alamos and Toulouse for the purpose of adding new spectra to the database as the need arises. Sensitivity to differences in wavelength correlation to spectral channels and spectral resolution has been investigated, indicating that spectral registration needs to be within half a pixel and resolution needs to match within 1.5 to 2.6 pixels. Absolute errors are tabulated for derived compositions of each major element in each standard using PLS regression. Sources of errors are investigated and discussed, and methods for improving the analytical accuracy of compositions derived from ChemCam spectra are discussed.