The sedimentary provenance investigations contribute to improve the understanding of fluvial palaeo-dynamics at different spatial and temporal scales. Conventional approaches for source to sink studies are usually based on the physical, mineralogical, geochemical, magnetic or isotopic properties of sediments. Efficient, accurate and relatively less time-consuming that the conventional fingerprinting techniques, Mid- Infrared Spectroscopy (MIRS) is an alternative method for tracing the origin of fine sediments. The research objectives of this study are (i) to use the MIRS method to discriminate the origin of fine alluvial sediments, by using multivariate statistics on a large set of reference samples from three different sedimentary basins (Rhine, Ill, Vosges tributaries) in the Upper Rhine area, (ii) to verify whether the IR spectra signatures of sediments are influenced by the geomorphological deposition context (bars versus banks), and, to some extent, by high organic matter content in the samples and thereafter (iii) to apply the MIRS method to study the fluvial palaeo-dynamics and to determine the source of palaeo-channel infillings in the Upper Rhine alluvial plain. The three sedimentary basins have been characterized by 196 reference samples of sub-recent alluvial deposits from 78 collecting sites. Discriminant analysis revealed a strong discriminating capacity by classifying correctly the samples by their origin without any inter-groups overlap, independently from the geomorphological context (bars versus banks), slight changes in organic matter contents or sediment texture. Combining MIRS technique and multivariate statistics for palaeo-channel infilling determination, complemented by radiocarbon 14C dating and mineralogical data, allowed to understand local morpho-sedimentary dynamics over the last ca.12 millennia.