Graduation Semester and Year

2013

Language

English

Document Type

Thesis

Degree Name

Master of Engineering in Civil Engineering

Department

Civil Engineering

First Advisor

Dong-Jun Seo

Abstract

Precipitation estimation is a very important topic from the societal perspective as heavy precipitation can cause flooding from which loss of lives and damage to properties can occur. In current practice, a number of spatial interpolation techniques are used for precipitation estimation using rain gauge data. Most of them are based on minimizing error variance but none of them consider Type-II conditional bias. As such, the existing techniques work well in the mid ranges of the precipitation distribution but tend to under- and overestimate large and small precipitation amounts, respectively. Conditional bias- penalized kriging (CBPK) adds a penalty term for Type-II conditional bias in addition to the error variance to improve estimation of large precipitation amounts. CBPK, however, tends to produce negative estimates in areas of very small or no precipitation. This problem is addressed in this work by an extension of CBPK, referred to as extended conditional bias-penalized kriging (ECPBK). For comparative evaluation, several real world experiments have been carried out using hourly rain gauge data. Also, synthetic experiments have been carried out for MAP analysis using the Stage IV data as truth and by creating synthetic gauge networks within the Stage IV precipitation field. The cross validation results of ECBPK are compared with those of the Single Optimal Estimation technique used in the NWS's Multisensor Precipitation Estimator.

Disciplines

Civil and Environmental Engineering | Civil Engineering | Engineering

Comments

Degree granted by The University of Texas at Arlington

Download

Share

COinS