Infrared spectroscopy tracing of sediment sources in a small rural watershed (French Alps).

The present article describes a first attempt to use infrared spectroscopy to trace the origin of suspended river sediments. Fifty samples of the main potential sediment sources within a small catchment area (990 ha) in the French Alps were collected and compared with samples of suspended sediment from the river, collected on various dates during 2006 and 2007 using sediment traps. Two major categories of sediment source were identified: topsoils and river channel sediments. For the qualitative part of the study, each of these two main categories was divided into two sub-categories, that is to say, cultivated and pastureland topsoils, and riverbed and riverbank sediments. Discriminant analysis on the source samples showed that Diffuse Reflectance Infrared Fourier Transform (DRIFT) spectroscopy can be used to differentiate between the four potential source materials. To determine whether or not immersion in the river altered the infrared spectra of these source materials, we measured the infrared spectra of samples that had been immersed in the river, in litter bags, for periods of up to 24 days. Immersion did not cause any major changes in the infrared spectra. The contribution of each type of source material to the suspended sediment in the river was quantified using partial least squares (PLS) analyses of DRIFT spectra to compare actual river sediment samples with an experimental model. This model was produced from the DRIFT spectra of a range of calibration samples produced by mixing source material samples in different ratios. The predictions of the model were valid and fell within the confidence interval calculated for the calibration set. Comparisons between suspended sediment samples and the model indicate that the predominant source of the sediment is riverbank erosion, which, in this case, is probably due to trampling by cattle.