|
Accuracy Investigation of Analytical Hierarchy Process (AHP) and Multivariate Regression (MR) Methods in Landslide Hazard Zonation, Using GIS (Case Study: Marbor River Basin)
|
Kourosh Shirani *   , Abdollah Saif , Masoud Alimoradi |
| University of Isfahan |
|
|
Abstract: (8519 Views) |
| A combination of natural and human factors have caused numerous landslide related damaged. One of the main strategies for reducing the damages of landslides is to avoid these regions. For this purpose it is necessary to be prepared landslide hazard zonation map precisely among available methods for these areas. At first, the object of this research is the relative weight determination of effective parameters, using Multivariate Regression (MR) in substitute of parameters priority in Analytical Hierarchy Process (AHP). For implementing this research, first of all, by aerial photos and satellite images, geological maps and studies of the field by using GPS, landslide inventory map in Marbor River Basin (sub-basin of the upper part of North Karoon Basin), an 800 square kilometers area located on the lower part of Southern Isfahan province, were prepared. By Using field studies and reviewing the related studies about similar areas with basins of Karoon and Dez Upper North and the existing thematic maps, 9 factors including lithology, slope, land use, rainfall, vegetation, aspect, and lineaments elements, containing fault and drainages, the action of factors in determining landslide (54 parameters in the consist of eight main factors) were measured. For the purpose of enhancing accuracy, accelerating and easing the analysis, all spatial and descriptive data entered into GIS system and were overlayed ArcGIS version9.3 software as vector. By overlaying these maps, 27466 homogeneous units were resulted, that were based upon calculating two methods including MR an AHP related to Multivariate decision analysis the results of analysis showed that both methods have almost had the same accuracy in separation landslide hazard zones, and the landslide hazard index was obvious too. Of course, it should be noticed that hierarchical approach based on regression data, on the basis the sum of quality index as an indicator of the accuracy, is more desirable. |
|
| Keywords: Landslide, Hazard zonation, Marbor watershed, Multivariate regression, Analytical Hierarchy Process, GIS |
|
|
Full-Text [PDF 2646 kb]
(3178 Downloads)
|
Type of Study: Research |
Received: 2015/02/19 | Accepted: 2015/02/19 | Published: 2015/02/19
|
|
|
|
|
|
|
| Send email to the article author |
|
|
