|
Landslides can result in enormous casualties and huge economic losses in mountainous regions. The article describes the statistical method used for the assessment of the landslide risk in the landslide-prone pilot regions of Uzbekistan. The study area comprises the territory of 250 square kilometers of mountainous area caused by different landslide-inducing factors and is divided into a grid with 70 m70 m cell-points. The occurrence of the landslide at every point of the study area is influenced by a number of factors and the probability of occurrence of landslide at every point of the study area is considered as function depending on the probabilities from each factor:
P=F (P1 ,P2,...,Pn) (1)
Where P- probability of the occurrence of landslide, Pi-probability of occurrence the landslide from i -factor only, n - the number of factors considered.
Pi- probability of occurrence of the landslide for each factor is a function of the value corresponding to the factor and varies throughout the study area:
Pi=Pi( Xi) (2)
Where Xi- the value of the factor.
For the assessment of the landslide hazards, the following factors were selected: elevation, slope, aspect, rivers, springs, arable lands, geological factors, rainfall. Based on statistical analysis of geospatial data on landslides and selected factors, the methodology for determining probability function (2) for each factor and probability of the occurrence of the landslide (1) was developed. The developed methodology was implemented using GIS ArcVeiw with Avenue scripting language and Visual C++ environment. The map of probability of the occurrence of landslides and probability maps for each factor were developed for study area.
|
