Geo-deformation failures leading landslides cause huge loss to socio-economic infrastructures and personal life and property all over the world, specifically in hilly terrains. It is a localized phenomenon usually gets triggered by rainfall. In this paper, a framework for a warning system based on prediction of failure of geo-deformation has been proposed. The system warns against rainfall-triggered landslides making use of a Knowledge Based System including a warning module. The warning process is generalized enough to be used for any other kind of natural hazard warning The system consists of the input module, the understanding module, rainfall prediction module, the expert module, the output module and the warning module. The input module accepts scanned images of thematic maps of contributing factors of landslides as well as output from rainfall prediction module, based on field observed GPS data. The understanding module interprets input information to extract relevant information as required by expert module. The expert module consists of a Knowledge Base (KB) and Inference strategy to categorize the given region into different intensities of landslide hazard, output module provides a warning message based on decision of inference module and finally, warning module send the message to user. The system stands tested and validated for landslide susceptibility categorization at five sites. Currently implementation of rainfall and warning module is underway. The major outcomes of the proposed system will lead to the development of a warning system towards regional scale geo-hazard analyses and provide a support system for landslide disaster preparedness and mitigation activities.

A popular usage-based hazard warning system / Bhattacharya, D.; Ghosh, J. K.; Boccardo, P.; Samadhiya, N. K.. - (2010), pp. 63-71. (Intervento presentato al convegno 6th International Symposium on Geo-Information for Disaster Management, Gi4DM 2010 tenutosi a Centro Congressi "Torino Incontra", ita nel 2010).

A popular usage-based hazard warning system

Bhattacharya D.;Boccardo P.;
2010

Abstract

Geo-deformation failures leading landslides cause huge loss to socio-economic infrastructures and personal life and property all over the world, specifically in hilly terrains. It is a localized phenomenon usually gets triggered by rainfall. In this paper, a framework for a warning system based on prediction of failure of geo-deformation has been proposed. The system warns against rainfall-triggered landslides making use of a Knowledge Based System including a warning module. The warning process is generalized enough to be used for any other kind of natural hazard warning The system consists of the input module, the understanding module, rainfall prediction module, the expert module, the output module and the warning module. The input module accepts scanned images of thematic maps of contributing factors of landslides as well as output from rainfall prediction module, based on field observed GPS data. The understanding module interprets input information to extract relevant information as required by expert module. The expert module consists of a Knowledge Base (KB) and Inference strategy to categorize the given region into different intensities of landslide hazard, output module provides a warning message based on decision of inference module and finally, warning module send the message to user. The system stands tested and validated for landslide susceptibility categorization at five sites. Currently implementation of rainfall and warning module is underway. The major outcomes of the proposed system will lead to the development of a warning system towards regional scale geo-hazard analyses and provide a support system for landslide disaster preparedness and mitigation activities.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2915450