You are here: Home » Past Issues » Volume 9, 2014 - Number 3 » LANDSLIDE HAZARD ASSESSMENT IN HASHTCHIN AREA, NW-IRAN
Agriculture and Natural Research center of Ardabil Province, P.O. Box 56135-545, Ardebil, Iran.
email@example.com and firstname.lastname@example.org
LANDSLIDE HAZARD ASSESSMENT IN HASHTCHIN AREA, NW-IRAN
Landslide hazard assessment is a key component of landslide studies. The most important and difficult task in landslide hazard analysis is estimating the probability of landslide occurrence in a study area. This task can be addressed with a spatial model that estimates the landslide probability in two stages. The proposed method has been used to perform a comprehensive landslide hazard assessment for Hashtchin area which is situated in the western Alborz Mountains, northwest of Iran. To this end, all the landslide data of the study area, which have occurred or have been reactivated at least once with moderate to high intensities in the last 50 years, were randomly divided into an estimation group containing 75% of landslide pixels and a validation group containing the remaining 25% of landslide pixels. In the first stage, the study area was divided into several prediction classes according to their relative likelihood of landslide occurrence, based on layers of effective factors. In the second step, the accuracy of the probability of landslide occurrence was examined using cross validation technique based on validation group, and by Receiver Operating Characteristic curve (ROC). The first stage of the research was repeated, but landslide occurrences were predicted based on all the landslide pixels. The sets of prediction classes were compared to the distribution of landslide occurrences in the validation group. Two approaches produced different prediction maps, and both of them generated acceptable results. The results, together with the hazard zonation maps, would allow planners to determine the actions for mitigating landslide effects, to avoid development in susceptible areas, and recommend adjustments to existing land use and restrictions for future land use. The map may also be used as a basis for the landslide risk assessment studies to be applied in the study area in the future. The introduced model can also be applied for mountainous areas with similar features including Alborz, Zagros and Caucuses Mountains.
Keyword: Landslide, Prediction, Likelihood, Validation, Accuracy and Hazard
(c) 2006 - 2018, Earth and Environmental Team
Design by Adrian Dorin