THE NON-LINEAR SPECTRAL EFFECT OF VARNISH COATINGS ON THE THERMAL EMISSION OF ROCKS

Thermal infrared (TIR) spectroscopy is widely implemented in attempts to determine the composition of rocks and other inorganic materials on both the Earth and on extra-terrestrial bodies. Coatings of varnish and dust will affect the emission spectra from the rock substrate making its identification from spectra alone equivocal. Dust coatings appear to have a linear effect on the spectra [1], as can varnishes [2]. However, sufficiently thick varnish coatings can have a non-linear effect. This non-linear effect can be used as a test for the presence of varnish or other coating or rind. Varnishes and coatings on rocks are common in terrestrial desert environments [3,4] and may occur on other solid surface bodies with atmospheres, such as Mars or Venus. The ability to detect the presence of a coating on rocks from only remotely sensed data would help in the understanding of the composition of the surface being sensed and can be indirectly diagnostic of the terrain morphology as well. For varnished granite, we have found that the low-emissivity 8.5 micron reststrahlan band remains present while the longer wavelength reststrahlen band becomes more emissive – reversing the relative strength of the two bands. The doublet at approximately 12.5 microns disappears. These effects are consistent with our model of the non-linear effects that a clay-rich varnish should have on the emissivity of a quartz-rich rock substrate. If the effect were linear, as expected for dusty surfaces or discrete patches of rock and clay, the emissivity of the 9.5 micron band would increase with greater dust or clay fraction, and the strengths of the two reststrahlen bands would not invert. From these and other results we are exploring the possibility that a judicious choice of only a few broad band wavelengths can be diagnostic of the presence of varnish on rocks. [1] Johnon et al., (2002), JGR, 107, E6. [2] Christensen, P. R. and S. T. Harrison, (1993) JGR, 98, 19819–19834. [3] Potter, R. M. and G. R. Ross-man, Science, 196, 1446-1448, 1977. [4] Potter, R. M. and G. R. Rossman, Chem. Geol., 25, 79-94, 1979.