International Journal of Wildland Fire International Journal of Wildland Fire Society
Journal of the International Association of Wildland Fire
RESEARCH ARTICLE

Artificial neural network approach for modeling the impact of population density and weather parameters on forest fire risk

Li-Ming Li A , Wei-Guo Song A B , Jian Ma A and Kohyu Satoh A
+ Author Affiliations
- Author Affiliations

A State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei, Anhui, 230027, PR China.

B Corresponding author. Email: wgsong@ustc.edu.cn

International Journal of Wildland Fire 18(6) 640-647 https://doi.org/10.1071/WF07136
Submitted: 13 September 2007  Accepted: 7 November 2008   Published: 22 September 2009

Abstract

The risk of forest fire occurrence is affected by the interactions among forest fuels, weather, human activities, etc. In the present paper, we try to build a method to model and forecast forest fire risk based on artificial neural networks. The data considered include population density and several weather parameters, i.e. average relative humidity, wind velocity and daily sunshine hours. With an interpolation method, these data have been expanded into 1 by 1 km meshes that are calculated according to the standard mesh code system in Japan, where the Japanese territory is divided into a lattice by latitude and longitude. Different parameter combinations and corresponding fire probabilities are computed. The correlations between forest fire probability and population density, and sequentially that between forest fire probability and combinations of population density together with one or several weather parameters are analyzed with three back-propagation neural networks in comparison with polynomial regression investigations. The results indicate that non-linear relationships exist among the influential factors and forest fire probability; artificial neural networks could better capture the non-linearity and give closer results to the test set compared with polynomial regression. The proposed method may be used to investigate and forecast forest fire risk providing there are enough data.

Additional keywords: fire danger rating, fire probability.


Acknowledgements

The study was supported by the China National Natural Science Foundation (No.30400344), National Science & Technology Pillar Program (2006BAK01A02–06) and Research and Development Special Fund for Public Welfare Industry (Forestry, 200704027).


References


Arrue BC, Ollero A , de Dios JRM (2000) An intelligent system for false alarm reduction in infrared forest-fire detection. IEEE Intelligent Systems and Their Applications  15, 64–73.


Bishop CM (1995) ‘Neural Networks for Pattern Recognition.’ (Oxford University Press: Oxford)

Bradstock RA, Gill AM, Kenny BJ , Scott J (1998) Bushfire risk at the urban interface estimated from historical weather records: consequences for the use of prescribed fire in the Sydney region of south-eastern Australia. Journal of Environmental Management  52, 259–271.
CrossRef |

Derbel F (2004) Performance improvement of fire detectors by means of gas sensors and neural networks. Fire Safety Journal  39, 383–398.
CrossRef | CAS |

Deeming JE, Burgan RE, Cohen JD (1977) The National Fire-Danger Rating System – 1978. USDA Forest Service, Intermountain Forest and Range Experiment Station, General Technical Report INT-39. (Ogden, UT)

Fausett L (1994) ‘Fundamentals of Neural Networks: Architecture, Algorithms and Applications.’ (Prentice-Hall, Inc.: Upper Saddle River, NJ)

Fernandes AM, Utkin AB, Lavrov AV , Vilar RM (2004) Development of neural network committee machines for automatic forest fire detection using Lidar. Pattern Recognition  37, 2039–2047.
CrossRef |

Haykin S (2004) ‘Neural Networks – A Comprehensive Foundation.’ 2nd edn. (China Machine Press: Beijing)

Lee BS, Alexander ME, Hawkes BC, Lynham TJ, Stocks BJ , Englefield P (2002) Information systems in support of wildland fire management decision making in Canada. Computers and Electronics in Agriculture  37, 185–198.
CrossRef |

Özesmi SL, Tan CO , Özesmi U (2006) Methodological issues in building, training, and testing artificial neural networks in ecological applications. Ecological Modelling  195, 83–93.
CrossRef |

Rose-Pehrsson SL, Shaffer RE, Hart SJ, Williams FW, Gottuk DT, Strehlen BD , Hill SA (2000) Multi-criteria fire detection systems using a probabilistic neural network. Sensors and Actuators  69, 325–335.

CrossRef |

Rumelhart DE, McClelland JL (1986) ‘Parallel Distributed Processing: Explorations in the Microstructure of Cognition.’ (MIT Press: Cambridge, MA)

Satoh K, Kitamura S, Komurasaki S, Kurahara K (2002) A system to predict occurrence and development of forest fires – computer simulation of forest fires based on weather data [in Japanese]. In ‘Proceedings of Thermal Engineering Conference’, 7–8 November 2002, Okinawa, Japan. pp. 457–458. (The Japan Society of Mechanical Engineers: Tokyo)

Satoh K, Kitamura S, Kuwahara K, Yang KT (2003) An analysis to predict forest fire danger and fire spread study to develop a fire danger rating and fire spread. In ‘Proceedings of the 2003 ASME Summer Heat Transfer Conference’, 21–23 July 2003, Las Vegas, NV. pp. 1–8. (ASME Press: New York)

Satoh K, Song WG, Yang KT (2004) A study of forest fire danger prediction system in Japan. In ‘Proceedings of 15th International Conference on Database and Expert Systems Applications (DEXA ’04)’, 30 August–3 September 2004, Zaragoza, Spain. (Eds F Galindo, M Takizawa, R Traunmüller) pp. 598–602. (Springer Verlag)

Song WG, Satoh K , Wang J (2004) Distribution analysis of forest fire-related data in Japan. Fire Safety Science  13, 180–185.


Song WG, Wang J , Satoh K (2005) Distribution and correlation analysis of wildfire and weather data in Japan. Progress in Natural Science  15, 363–367.
CrossRef |

Song WG, Wang J, Satoh K , Fan WC (2006) Three types of power-law distribution of forest fires in Japan. Ecological Modelling  196, 527–532.
CrossRef |

Song WG, Zheng HY, Wang J, Ma J , Satoh K (2007) Weather-driven model indicative of spatiotemporal power laws. Physical Review E: Statistical, Nonlinear, and Soft Matter Physics  75, 016109.
CrossRef |

Syphard AD, Radeloff VC, Keeley JE, Hawbaker TJ, Clayton MK, Stewart SI , Hammer RB (2007) Human influence on California fire regimes. Ecological Applications  17(5), 1388–1402.
CrossRef | PubMed |

Vasconcelos MJP, Silva S, Tomé M, Alvim M , Pereira JMC (2001) Spatial prediction of fire ignition probabilities: comparing logistic regression and neural networks. Photogrammetric Engineering and Remote Sensing  67(1), 73–81.


Vasilakos C, Kalabokidis K, Hatzopoulos J, Kallos G , Matsinos Y (2007) Integrating new methods and tools in fire danger rating. International Journal of Wildland Fire  16(3), 306–316.
CrossRef |

Van Wagner CE (1987) Development and structure of the Canadian Forest Fire Weather Index System. Canadian Forestry Service, Forestry Technical Report 35. (Ottawa, ON)

Vega-Garcia C, Lee BS, Wooddard PM , Titus SJ (1996) Applying neural network technology to human-caused wildfire occurrence prediction. AI Applications  10(3), 9–18.



Export Citation Cited By (15)