Building models for automatic landslide-susceptibility analysis, mapping and validation in ArcGIS

被引:47
作者
Jimenez-Peralvarez, J. D. [1 ]
Irigaray, C. [1 ]
El Hamdouni, R. [1 ]
Chacon, J. [1 ]
机构
[1] Univ Granada, Dept Civil Engn, ETSICCP, E-18071 Granada, Spain
关键词
Bivariate statistical analysis; Landslide susceptibility; GIS matrix method; ArcGIS; BETIC-CORDILLERA; GIS; MAPS; PREDICTION; EXAMPLES; GRANADA;
D O I
10.1007/s11069-008-9305-8
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
In this paper, ModelBuilder(TM) in ArcGIS (ESRI) has been applied to landslide-susceptibility analysis, mapping and validation. The models (scripts), available for direct downloading as an ArcGIS tool, allow landslide susceptibility to be computed in a given region, providing a landslide-susceptibility map, with the GIS matrix method, and ensuring a quality validation. The paper details the steps needed for the model-building process, enabling users to build their own models and to become more familiar with the tool. The susceptibility model leads the user first through a Digital Elevation Model (DEM), depicting the morphological and morphometric features of the study area, and then through a Digital Terrain Model (DTM), useful as a source of landslide-determinant factors, such as slope elevation, slope angle and slope aspect. In addition, another determinant factor is the lithological unit, independent of the DEM. Once the determinant landslide factors are reclassified and in a vectorial format, all the combinations between the classes of these factors are determined using the geoprocessing abilities of ArcGIS. The next step for the development of the landslide-susceptibility model consists of identifying the areas affected by a given surface of rupture (i.e. source area) in every combination of the determinant-factor classes. This step leads to the landslide matrix based on a previously georeferenced landslide database of the region, in which the slopes are distinguished into two simple classes: with or without landslides. In the last stage, to build a landslide-susceptibility model, the user computes the percentages of area affected by landslides in every combination of determinant factors. In the resulting landslide-susceptibility map a progressive zonation of areas or slopes increasingly prone to landslides is performed. A model for the validation of the resulting landslide-susceptibility map is also presented, based on the determination of the degree of fit, which is calculated from the cross tabulation between a set of landslides (not included in the susceptibility analysis) and the corresponding susceptibility map.
引用
收藏
页码:571 / 590
页数:20
相关论文
共 67 条
[1]  
Agterberg F.P., 1993, COMPUTERS GEOLOGY 25, P13
[2]  
AGTERBERG FP, 1989, GEOL SURV CAN, V89, P171
[3]  
Aleotti P., 1999, B ENG GEOL ENVIRON, V58, P21, DOI DOI 10.1007/S100640050066
[4]  
[Anonymous], NAT HAZARDS REV
[5]  
[Anonymous], 1980, THRESHOLDS GEOMORPHO
[6]  
*ARCGIS, ESRI ARCMAPTM 9 0 LI
[7]   The application of GIS-based logistic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan [J].
Ayalew, L ;
Yamagishi, H .
GEOMORPHOLOGY, 2005, 65 (1-2) :15-31
[8]  
Bonham-Carter G., 1994, GEOGRAPHIC INFORM SY
[9]  
BONHAMCARTER GF, 1988, PHOTOGRAMM ENG REM S, V54, P1585
[10]  
CANUTI P, 1996, PUBBLICAZIONE CNR GN, V846