Integrated spatio-temporal data mining for forest fire prediction.
Transactions in GIS
Forests play a critical role in sustaining the human environment. Most forest fires not only destroy the natural environment and ecological balance, but also seriously threaten the security of life and property. The early discovery and forecasting of forest fires are both urgent and necessary for forest fire control. This article explores the possible applications of Spatio-temporal Data Mining for forest fire prevention. The research pays special attention to the spatio-temporal forecasting of forest fire areas based upon historic observations. An integrated spatio-temporal forecasting framework - ISTFF - is proposed: it uses a dynamic recurrent neural network for spatial forecasting. The principle and algorithm of ISTFF are presented, and are then illustrated by a case study of forest fire area prediction in Canada. Comparative analysis of ISTFF with other methods shows its high accuracy in short-term prediction. The effect of spatial correlations on the prediction accuracy of spatial forecasting is also explored. © 2008 The Author. Journal compilation © 2008 Blackwell Publishing Ltd.
|Title:||Integrated spatio-temporal data mining for forest fire prediction|
|Keywords:||Dynamic recurrent neural network, Elmanneural network, Forest fire prevention, Local Moran's Index, Spatio-temporal data mining, Spatio-temporal forecasting|
|UCL classification:||UCL > School of BEAMS > Faculty of Engineering Science
UCL > School of BEAMS > Faculty of Engineering Science > Civil, Environmental and Geomatic Engineering
UCL > School of BEAMS > Faculty of the Built Environment
UCL > School of BEAMS > Faculty of the Built Environment > Centre for Advanced Spatial Analysis
Archive Staff Only