Development of Web-based Spatial Decision Support System for Landslide Susceptibility Assessment in Xishuangbanna, China

Abstract

Climate change, human activities such as rapid and unplanned urban expansion, increase the risk of natural hazards. Landslide is one of the most severe that cause imponderable loss of human life and property and showing a trend of increasing occurrence worldwide. Both national and provincial scales of landslide-prone area zonation studies have been found but no regional such study yet for Xishuangbanna Prefecture, Yunnan Province, China. The Frequency Ratio (FR) is selected as the mapping model due to its popularity, simplicity, and understandability. Fourteen most frequently used landslide conditioning factors (LCFs) have been extracted from a comprehensive literature review (52 studies) firstly, then 27 scenarios have been designed to produce the 27 landslide susceptibility maps, and the Area Under the Curve (AUC) of each map has been calculated and compared. Finally, seven of the fourteen factors are identified as the more effectively decisive LCF for the study area, the final landslide susceptibility mapping (LSM) was done based on these 7 LCFs. The AUC values of the final landslide susceptibility map with inputting landslide inventory dataset and verification dataset are 85.8% and 84.0%, respectively, which means the model acquired a very good result for both success and prediction rate. Innumerable landslide susceptibility assessment (LSA) studies have been done by researchers in recent decades, but most of the outcomes of these works have been archived and rare of them have been well used for supporting decision-makers in real situations. The last objective of the study is to adopt the outcomes of the study to develop a Web-based Spatial Decision Support System (Web-based SDSS), which not only can be used as an effective aiding tool to support the decision-making processes but its broad potential for further development also provides a reconsideration of making better use of the outcomes of these GIS-based studies.

-