Modelling the transmission component in TIR reflectance spectra of sandstones to understand the effect of surface roughness and clinging fines

Surface roughness of rocks influences the spectral shape and amplitude of thermal infrared spectra, but the relationship is poorly understood. This research aims to understand the physical processes that cause the observed spectral variation by modeling the reflectance spectra with a nonlinear transmission model. We designed a model that combines rock surface reflection with transmission through clinging fines that coat the surface. The novelty of this approach is that it does not model individual particles, but compiles different combinations of thin particle transmission and reflectance of the solid surface until it has an optimal fit with actual measured rock spectra. We use directional-hemispherical reflectance spectra of two quartz sandstones with a varying surface roughness, a transmission spectrum of a powder from one of these sandstones, and a pure mineral (kaolinite) transmission spectrum as input for the model. The model reproduces spectral amplitude and shape of the principal quartz reststrahlen doublet of three out of four measured sandstone spectra. It also correctly displays the strong bands of the quartz reststrahlen doublet. With the model we demonstrate that reflectance and transmission based modeling is a promising technique for identification and correction of spectral characteristics that result from a different surface roughness.