Bayesian methods in hydrology
Abstract
The main objective of this study is to apply the Bayesian methods to solve problems in hydrology. These methods have not been used in the past. Five problems are investigated in this thesis by using the Bayesian methods. The first of these is the estimation of Instantaneous Unit Hydrographs which is a link between rainfall and runoff. The second problem deals with the detection of changes in characteristics in the mean, variance and other statistical properties of a time series. The third problem is to estimate the variable coefficients of a linear regression model by using a multicriteria optimization method. The fourth problem deals with recursive forecasting in multivariate models and tracking the time evolution of the state vector in a Kalman filter representation of a multivariate model. In the last problem, the Bayesian theory is employed to estimate the periodicity in time series. Based on the results of this study, the Bayesian methods have been shown to be flexible and better for investigation of problems in hydrology.
Degree
Ph.D.
Advisors
Rao, Purdue University.
Subject Area
Civil engineering|Mechanical engineering|Hydrology
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.
