Валидација карата подложности клизиштима (Студија случаја: урбано подручје општине Бања Лука, БиХ)
Validation of Landslide Susceptibility maps (Case study: Urban Area of the Municipality of Banja Luka - B&H)
Author(s): Novica Lovrić, Radislav TošićSubject(s): Physical Geopgraphy, Applied Geography, Maps / Cartography
Published by: Српско географско друштво
Keywords: landslide; landslide; susceptibility maps; GIS matrix method; validation; B&H
Summary/Abstract: Landslides represent a serious geo-hazard in many areas of the world. They are one of the most damaging and most significant geo-hazards in Bosnia and Herzegovina. In the previous research, three GIS-based methodologies (Index based method, Statistical index method, and Landslide susceptibility analysis), have been used to assess the landslide susceptibility in urban area of the Municipality of Banja Luka. Validation technique is performed by comparing existing landslide data from 2012 with obtained landslide susceptibility maps. In this research, the landslide susceptibility maps were supplemented by landslide susceptibility map which is prepared using a GIS Matrix method. The area and percentage distribution of the susceptibility classes in the study area were determined as a result of the four different techniques. The Statistical index method has provided the most satisfying results. As a consequence of heavy rains during the period from May to August 2014, 126 landslides occurred in the study area, and they offer a good opportunity to validate obtained landslide susceptibility maps of the study area with landslides that occurred in different periods of time. Validation was performed by using a ‘degree of fit’ method. The values of the degree of fit in high and very high category in all used methods are over 80%, except for GIS matrix method, where the percentage is 60%. The validation of obtained landslide susceptibility maps suggests that the applications of all the used techniques provide a good basis for creation landslide susceptibility maps, but the best results are given by using a Statistical index method.
Journal: Гласник Српског географског друштва
- Issue Year: 97/2017
- Issue No: 1
- Page Range: 19-34
- Page Count: 16
- Language: English