Development of a fast and inexpensive method for detecting the main sediment sources in a river basin

Abstract

The sediments are a key link to understand the dynamics of a water basin, since they are the main transport and retention agent of nutrients and contaminants. Therefore, a method to identify the main sediment sources and study their distribution through the basin over time without the need to use reference standards was developed. Water samples with suspended sediments were collected for one year from different sites of the Ludueña stream basin (Argentina). The sediments were deposited in fiberglass filters and their near infrared spectrum was measured. The data obtained were processed by bilinear and trilinear methods. Principal component analysis (PCA), multivariate curves resolution with alternating least squares (MCR-ALS) with and without trilinearity; and parallel factor analysis (PARAFAC) were applied. These algorithms allow identifying the three main sediment sources. The spectral loadings that characterize each source and the scores that reflect their distribution along the different sampling sites were obtained. It was concluded that the main contribution of sediments is given by the plant tissue developed by photosynthetic organisms that have a seasonal behavior. The remaining sources are characterized by being soil particles of the region under study (incorporated into the system by wind and rain) and the development of organisms with anoxygenic photosynthesis (their growth is favored due to the contribution of fluid discharges from anthropogenic activities).