Redeconvolve

We resampled our laboratory rock spectra to the 10 cm⁻¹ sampling of the TES and redeconvolved them to ascertain whether or not adequate results for rock classification could still be obtained with data acquired at a spectral resolution comparable to that of the lowest resolution TES data.

Next, the deconvolved trace in Figure 2(a) is subjected to the redeconvolution process and the result is shown in Figure 2(b). Comparison of this redeconvolved trace with the reflectivity sequence in Figure 1(b) shows that the agreement is approximately the same as before.

To account for this shortcoming, we introduced two Gaussian peaks in these positions (highlighted in orange), and redeconvolved these spectra, to obtain an optimum fit.