Study of Mineralogical Composition of Sediment in Brahmaputra River in Urban Stretch of Guwahati City, Assam, India

Abstract

High sediment load is an integral component of the Brahmaputra River System. Due to its sheer amount and the complex behavior of the sediment transport, its control has remained a challenge. In this study, an attempt has been made to study the mineralogical composition of suspended sediment in unban stretch of River Brahmaputra. In order to understand the properties of the sediment, Scanning electron microscope, X-ray diffractometer, Atomic Absorption Spectrometer, and Particle size distribution analysis are carried out. In the meanwhile, the composition of suspended sediment is also evaluated so as to know about the present scenario of the river system that can be used for water supply in Guwahati city. Artificial neural network (ANN) and nonlinear regression (using dataFit software) were developed, to predict both sediment concentration and concentration of chemical parameters of sediment. The nonlinear nature of suspended sediment load time series necessitates the utilization of nonlinear methods for simulating the suspended sediment load. Si, Al, Fe, Mg, Ca, Na, and K were analyzed in AAS for determining chemical composition of the sediments. Here, ANNs model is developed using the toolbox of the MATLAB software and also compared with results of dataFit software models.