Sikder, Mohammad Abdus Salam
(2012)
Probabilistic analysis and prediction of bend migration in meandering alluvial streams.
PhD thesis, University of Nottingham.
Abstract
A meandering river is one which exhibits a serpentine course, formed by a series of alternating bends connected by short, relatively straight reaches. Meandering behaviour has been found to be ubiquitous in alluvial streams and rivers. The tendency of the bends in meandering rivers to evolve and shift is a natural outcome of interactions between the flowing water, sediment in transport and the resistance to erosion of the boundary sediments. The resulting formation, evolution and shifting of bends is intrinsic to the morphological behaviour of meandering, alluvial streams. The need to understand and predict this behaviour stems not only from the desire to explain the morphodynamics of meandering rivers, but is also required to manage down the risks to property and infrastructure located along the course of meandering rivers, as well as the people who live, work or travel in the vicinity of alluvial streams with meandering planforms. In this context, the aim of this research is to develop an original, scientifically-based, and practical approach to the risk-based prediction of the hazard associated with bend migration that accounts for the uncertainty inherent to the morphological behaviour of rivers. A database compiled for the National Cooperative Highway Research Program (NCHRP) was used as the primary source of data. This includes data for 1,505 bends taken from 125 locations on 89 rivers spread across 24 States.in the USA The rivers were classified into nine categories of meandering, based on a modified Brice Typology. Bend geometry data in the database were measured from aerial photographs and maps representing the bends in the 1930/40s, 1950/60s and 1990s. It has long been recognised that the dimensionless rate of bend migration (M/W) is related to the stage of bend evolution (initiation, growth, translation, termination) and that this can be represented by the ratio of the bend radius of curvature to the channel width (R/W). It was decided to analyse bend migration based on these parameters and use the dimensionless form of the bend radius as the independent variable in the probabilistic analysis. Using normal conditional density functions fitted to the data, rates of meander migration were estimated corresponding to: 50%, 60%, 75%, 80%, 90%, 95%, 97%, 98%, and 99% probabilities of not being exceeded. The risk associated with a hazard depends not only on the probability that it will occur but also the consequences should it do so. In selecting the acceptable level of probability that the predicted rate of migration will not be exceeded, care should therefore be exercised by the user in considering all the possible consequences should this occur.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
