UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Landslide Risk Zoning Applying Kohonen's Self-Organizing Map Neural Network Technique

Ahmed, B; Forte, R; (2016) Landslide Risk Zoning Applying Kohonen's Self-Organizing Map Neural Network Technique. In: Rahman, MS, (ed.) Proceedings of the 1st Bangladesh Planning Research Conference (BPRC). Department of Urban and Regional Planning, Jahangirnagar University: Dhaka, Bangladesh. Green open access

[thumbnail of 404_Ahmed.pdf]
Preview
Text
404_Ahmed.pdf - Published Version

Download (2MB) | Preview

Abstract

Every year during the monsoon period, the vulnerable communities living in the dangerous hill slopes in Cox's Bazar Municipality (CBM), Bangladesh face landslide hazards. The frequency and intensity of landslide hazards are increasing due to torrential rainfall in short period of time. In addition, being the most attractive tourist city of Bangladesh, CBM is facing acute population pressure. The rate of urbanization is also high in CBM. The local people are building residential houses by cutting the hills and making the landslide disaster scenario worse. Frequent landslide hazards are causing human casualties, and property, infrastructure and roads are damaged. The economic loss due to landslides is also high in CBM. Therefore, it is necessary to prepare the landslide risk zoning maps to help mitigate the adverse impacts. This article has adopted the Kohonen's Self-Organizing Map (SOM) neural network technique. To produce the SOM for landslide risk mapping, a total of 12 factor maps (i.e. slope, land cover, geology, geomorphology, NDVI, soil moisture, rainfall pattern, and distance from-existing buildings; stream, road and drainage network, and faults and lineaments) are selected. A detail landslide inventory map is prepared for network training and model validation purpose. The performance of the SOM method is validated using the Area Under the relative operating characteristic Curve (AUC) method. The AUC value is calculated 86.60%. The SOM image is reclassified; applying natural breaks (Jenks) method, in four different risk zones-very high, high, medium and low. The landslide risk-zoning map as produced by SOM technique is found scientifically significant and representing the ideal situation. The concerned urban planners, engineers and stakeholders can use this kind of risk-zoning map for formulating policies related to landslide disaster risk reduction in the hill districts of Chittagong, Bangladesh.

Type: Proceedings paper
Title: Landslide Risk Zoning Applying Kohonen's Self-Organizing Map Neural Network Technique
Event: 1st Bangladesh Planning Research Conference (BPRC)
Location: Dhaka, Bangladesh
Dates: 05 February 2016 - 06 February 2016
Open access status: An open access version is available from UCL Discovery
DOI: 10.13140/RG.2.1.2748.7766/1
Publisher version: http://www.juniv.edu/department/urp
Language: English
Additional information: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Inst for Risk and Disaster Reduction
URI: https://discovery.ucl.ac.uk/id/eprint/1475788
Downloads since deposit
105Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item