Spectral fingerprinting: characterizing suspended sediment sources by the use of VNIR-SWIR spectral information

摘要: Knowledge of sediment sources is a prerequisite for sustainable management practices and may furthermore improve our understanding water fluxes. Investigations have shown that number characteristic soil properties can be used as “fingerprints” to trace back the river sediments. Spectral recently been successfully such characteristics in fingerprinting studies. Despite being less labour-intensive than geochemical analyses, example, spectroscopy allows measurements small amounts material (>60 mg), thus enabling inexpensive analyses even intra-event variability. The focus this study on examination spectral fluvial samples detect changes source contributions, both between within individual flood events. Sediment from following three different origins were collected Isabena catchment (445 km2) central Spanish Pyrenees: (1) main potential areas, (2) stored fine channel bed once each season 2011 (3) suspended during four events autumn spring 2012 at outlet well several subcatchment outlets. All dried measured laboratory using an ASD spectroradiometer. Colour parameters physically based features (e.g. organic carbon, iron oxide clay content) calculated spectra. Principal component (PCA) applied all types determine natural clustering samples, mixing model was contributions. We found seems mainly influenced by grain size seasonal variability, while sampling location—and effect tributaries or subcatchments—seem minor importance. Suspended vary between, within, events; although badlands always major source. Forests grasslands contributed little (<10 %), other (not further determinable) up 40 %. suggested differ among subcatchments comprising relatively large proportions most analyzed. fingerprints provide rapid cost-efficient alternative conventional fingerprint properties. However, combination could potentially permit discrimination larger types.