:: Volume 5, Issue 3 (12-2017) ::
jgit 2017, 5(3): 99-122 Back to browse issues page
Modeling the spreading of forest fire based on a cellular automata using the markov chain and MOLA with a neighborhood filter
Parham Pahlavani * , Hamid Reza Sahraiian , Amin Raei
University of Tehran
Abstract:   (7613 Views)
Nowadays, to reduce the damages and high costs of forest fire, there is a need for identifying the factors affecting forest fire, modeling the spread of the fire, as well as specifying the actions to extinguish forest fire. In this research, we tried to identify the biophylsical and human factors affecting spread of the fire in a study area using the geographically weighted regression (GWR) integrated with a genetic algorithm. Subsequently, spread of the forest fire was modeled using the cellular automata (CA), markov chain, and multi-objective land allocation (MOLA) with various neighborhood filters for calibration of transition rules of the CA. Moreover, a combination of the CA and logistic regression was used  to compare with the results of the method mentioned above. Results showed that for the fire that happened  on the study area on November 17, 2010, the proposed CA algorithm using Markov chain and MOLA with a 3×3 neighborhood filter and 30 m pixel size is more precise than those of the other neighborhood filters and pixel sizes. In this case, the kappa index, the overall accuracy, and the relative operating characteristic (ROC) were equalled to 88.8 %, 95.1 %, and 89.0 %, respectively. Also, comparison of two proposed methods of this research indicated that the CA algorithm using the Markov chain and MOLA reached more precise and accurate results than those achieved by the CA algorithm using the logistic regression.
Keywords: Spreading of Forest Fire, Geographically Weighted Regression, Cellular Automata, Markov Chain, MOLA, Logistic Regression
Full-Text [PDF 1977 kb]   (1370 Downloads)    
Type of Study: Research | Subject: GIS
Received: 2016/08/28 | Accepted: 2017/06/18 | Published: 2018/01/10

XML   Persian Abstract   Print

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 5, Issue 3 (12-2017) Back to browse issues page