J. Korean Soc. Hazard Mitig Search

CLOSE


Journal of the Korean Society of Hazard Mitigation 2015;15(4):215-220.
Published online August 31, 2015.
Evaluation of Flood Damages Using Principal Component Analysis
Evaluation of Flood Damages Using Principal Component Analysis
Marc Sinan B Azotea, Imee V Necesito, Tae Sung Cheong, Insang Yu, Sangman Jeong
Abstract
Despite the growing economy of Korea, natural disasters such as floods are still knocking the economic ladder. The counteractions made by flood disasters that were also induced by typhoons have caused significant damages to properties and human life. In this regard, the searches for the primary parameters that are affecting the flood damage amount were focused in this study. The data used in this study is flood damage from the 485 facilities in Gunsan City which were further divided based on the facility type. It identifies what type of parameter variable namely: flood damage, flood depth, flood duration, inundation area, family income and land price greatly affects or are relevant to each facilities being classified using Principal Component Analysis (PCA). Results from the Prin-cipal Component Analysis (PCA) show the relationship that exists among the variables and that it had narrowed down to which parameter was correlated to flood damage. This paper used the eigenvector and eigenvalue of the data matrix in order to compute the principal component. The result of the total variance in the flood damage of residential, commercial and agricultural facilities using the eigenvector and eigenvalue were calculated to be 0.691, 0.626 and 0.700, respectively. Significant level of Principal Component Analysis greater than 0.5 and less than -0.5 must be attained for relativity variables to exist and as for this study it proves that flood depth was the most significant among the rest.
Key Words: Flood damage; Principal Component Analysis; Water depth; Significant level
요지
한국의 경제성장에도 불구하고 홍수와 같은 자연재해는 여전히 경제적 지위에 타격을 가하고 있으며 태풍에 의해 발생한 홍수피해는 막대한 재산과 인명피해의 원인이 되어왔다. 이와 관련하여 본 연구에서는 홍수피해에 영향을 주는 주요 매개변수를 산정하였으며 시 설의 종류에 따라 구분된 군산시의 485개 시설에 대한 홍수피해액 자료를 활용하였다. 주성분분석을 통하여 홍수피해액, 침수심, 침수 시간, 침수범람면적, 수입, 지가 중 어떤 종류의 매개변수가 각각의 시설에 가장 크게 영향을 미치는지 분석하였다. 주성분분석 결과는 변수들 간의 상관관계와 어떠한 매개변수가 홍수피해액과 연관성이 있는지 나타낸다. 본 연구에서는 주성분을 산정하기 위해 데이터 매트릭스의 고유벡터와 고유치를 활용하였으며 고유벡터와 고유치를 사용한 주거, 상업, 농업시설의 홍수피해 변량 산정 결과, 각각 0.6918, 0.6265, 0.7001으로 산정되었다. 주성분분석의 유의수준이 0.5보다 크거나 -0.5보다 작으면 변수의 상관성이 인정되는데 본 연 구에서는 여러 변수 중 침수심이 가장 상관성이 높은 것으로 나타났다.
핵심용어: 홍수피해; 주성분분석; 침수심; 유의수준


ABOUT
ARTICLE CATEGORY

Browse all articles >

BROWSE ARTICLES
AUTHOR INFORMATION
Editorial Office
1010 New Bldg., The Korea Science Technology Center, 22 Teheran-ro 7-gil(635-4 Yeoksam-dong), Gangnam-gu, Seoul 06130, Korea
Tel: +82-2-567-6311    Fax: +82-2-567-6313    E-mail: master@kosham.or.kr                

Copyright © 2024 by The Korean Society of Hazard Mitigation.

Developed in M2PI

Close layer
prev next