LANDSLIDE SUSCEPTIBILITY MAPPING USING FREQUENCY RATIO, LOGISTIC REGRESSION, ARTIFICIAL NEURAL NETWORKS AND THEIR COMPARISON: A CASE STUDY FROM KAT LANDSLIDES (TOKAT-TURKEY)

7th International Scientific Conference - SGEM2007, www.sgem.org, SGEM2007 Conference Proceedings/ ISBN: 954-918181-2, June 11-15, 2007

ABSTRACT/Full article not available/

This case study presented herein compares the landslide susceptibility mapping
methods of frequency ratio (FR), logistic regression and artificial neural networks
(ANN) applied in the Kat county (Tokat-Turkey). Digital Elevation Model
(DEM) was first constructed using a GIS software. Parameter maps affecting the
slope stability such as; geology, faults, drainage system, topographical elevation,
slope angle, slope aspect, topographic wetness index (TWI) and stream power
index (SPI) were then produced from DEM of the study area. In the last stage of
the analyses, landslide susceptibility maps were produced using the frequency
ratio, logistic regression and neural networks, and they were then compared
by means of their validations. As a result of this study, higher accuracies of
susceptibility maps for all three models were obtained. However respective
coefficient of correlations 84.6%, 86.2%, 87.7% for frequency ratio, logistic
regression and artificial neural networks showed that the map obtained from
ANN model looks like more accurate than the other models, accuracies of all
models can be evaluated relatively similar. The results obtained in this study also
showed that the frequency ratio model can be used as a simple tool in assessment
of the landslide susceptibility when a sufficient number of data was obtained.
Because input process, calculations and output process are very simple and can
be readily understood in the frequency ratio model, however logistic regression
and neural networks require a conversion of data used in analyses into ASCII or
other formats. Moreover, it is also very hard to process the large amount of data
in the statistical package.

Keywords: Landslide, susceptibility map, GIS, frequency ratio, logistic regression,
artificial neural networks, Kat (Tokat-Turkey).