Empirical orthogonal function analysis of hydrologic data
Abstract
In recent years, the empirical orthogonal functions (EOFs) have been widely used in meteorology for analysis of spatial data. The EOF analysis has not been used much in hydrological data analysis. In fact, the problems associated with their use and the limitation of the use of EOFs in the analysis of hydrological data are not clear. The purpose of this thesis is to investigate the suitability of the application of empirical orthogonal functions to hydrologic problems, such as estimating data at ungaged sites, the accuracy of forecasts obtained by using regional data and others. The following hydrological problems are investigated. The spatial and temporal variabilities of hydrologic processes are first analyzed and characterized. The estimation of hydrologic data at ungaged locations is investigated next. The accuracy of the basinwide forecasts of hydrologic data is estimated. Finally, the physical mechanisms underlying the predictability of runoff process are investigated. The data used in this thesis are maximum temperature, rainfall, and runoff in Illinois, Indiana, Kentucky, and Ohio. Different time scales of these data, such as daily, weekly, and monthly, are also used. Based on the results shown in this thesis, the empirical orthogonal functions are found to be quite useful in the analysis of hydrologic problems.
Degree
Ph.D.
Advisors
Rao, Purdue University.
Subject Area
Hydrology
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.