APPLICATION OF VIS/NIR SPECTRAL REFLECTANCE IN SOURCING AND RECOGNITION OF HEAT-TREATMENT IN CHERTS

Traditionally, chert sourcing techniques include visual identification, petrography, x-ray diffraction, and trace element analysis. Visual methods can be subjective and interpretations are often contentious amongst experts. XRD, petrographic, and trace element analysis are destructive to the sample. By contrast, diffusive spectral reflectance (DSR) has been successfully used as a geologic tool to identify mineral phases in rocks and sediments. The method is non-destructive, fast, and relatively inexpensive: all attributes which make it an attractive technique for provenance studies of archaeological cherts. We present preliminary results from DSR analysis on a small chert reference collection. Data was collected with an ASD Labspec Pro FR spectrometer in the visible (VIS) and near-infrared (nIR) spectral range (400 to 2500 nm). Flaked surfaces were examined with a 3mm fiber-optic probe at 0.5mm from the probe surface. Much of our work is concerned with the mechanics of collecting meaningful data in a non-destructive, but repeatable way. Also under investigation is the degree to which sample surface texture and the incident angle of the probe effect the results. Preliminary data suggests that polished surfaces have higher intensities across the spectra due to enhanced specular reflectance, but show little change in the spectral shape, which is controlled by the diffuse components of reflectance. Samples from different sources that are virtually indistinguishable visibly yield distinct spectra with signal-to-noise ratio as high as 70 times. In addition to lithic sourcing, our results suggest that DSR analysis will also be of utility in determining thermally altered cherts. Intentional heat-treatment of raw lithic material was a widespread prehistoric technique used to increase workability or aesthetics. Traditionally archaeologists use color changes and various morphological features (e.g. pot-lid fractures, luster, crazing) to identify heat-treatment. Quantitative analysis outside the visual spectrum may allow identification of heat-treatment in the absence or in support of the usual diagnostic characteristics. Further work will include heat-treatment under experimental conditions and their spectral characteristics.