Journal of Natural Disaster Science

PREDICTION METHOD FOR RISK RATE OF SLOPE FAILURE CAUSED BY HEAVY RAINFALL*

Masanori B. OKAMOTO
Professor, Faculty of Economics, Hiroshima University, Hiroshima

(Received 14 November, 1979 and in revised form 25 January, 1980)

Abstract

An estimation of the spatial and temporal change in the risk rate of slope failure approaching the centre of a heavy rainfall area is obtained by the use of Hayashi's quantification method I. The data relevant to slope failure were obtained from topographical and geotechnical characteristics using aerial photographs and existing information. Most data is categorical except for the risk rate of slope failure and rainfall amount. Two-factor interactions among five items are checked by 5-way with one covariate (rainfall amount). The interaction between them is not found to be statistically significant.

A prediction of the risk rate of slope failure for any other given rainfall amount is obtainable with this method by the use of the scores determined from data specific to that area. The prediction results of the risk rate are automatically plotted by the computer. Prediction errors are also estimated.

Key words

disaster, slope failure, heavy rainfall, statistical analysis

* Part of this work was published in the Journal of Analytical Researches on Data of Natural Disaster, Vol.6, 1977 (in Japanese).