A Simulation Method Of Forest Fire Spread Based On WRF And Cellular Automata

Forest fires are characterized by high suddenness,randomness,rapid spread and destructiveness,etc.If they are not controlled in time,they will directly change the surface vegetation cover and ecosystem structure,and even have a great impact on people’s life safety and social economy.Therefore,timely and accurate forest fire spread prediction at the time of fire is of great significance for forest fire fighting and fire spread control.Cellular automata(CA)is widely used in the study of forest fire spread,previous research focused on the simulation of the spread of forest fires after the occurrence of forest fires,due to the lack of accurate meteorological data,the above methods rarely consider the trend prediction of forest fire spread.A large number of studies have shown that the Weather Research and Forecasting Model(WRF)can better predict the change trend of near-surface wind field within a certain time and space range,so this paper proposes a forest fire spread simulation method integrating CA and WRF.The specific research work is as follows:(1)Different parameterization schemes were selected to numerically simulate the wind field in the forest area of Muli County,and the simulation results were compared with the automatic weather station data for verification,and the effects of different schemes on the simulation of the wind field in the forest area were analyzed.The results show that the combination of Noah land surface scheme,YSU boundary layer scheme and WSM6 microphysical scheme is more suitable for the numerical simulation of the wind field in the forest area of Muli County(2)The mathematical models of forest fire spread and computer simulation techniques were compared,and by comparing the advantages and disadvantages of different spread models and simulation methods,the Wang Zhengfei model was selected as the base rate model for the forest fire spread simulation study in this paper,and the conversion rule of metacell was defined accordingly.Based on this conversion rule,a validation test of CA method for forest fire spread under different wind field conditions was conducted,and the results showed that the CA algorithm constructed based on Wang Zhengfei’s velocity equation is feasible for forest fire spread simulation.(3)A forest fire spread simulation method based on WRF and CA was established.The predicted wind field of WRF is used as the input factor of wind field of CA for real-time forest fire spread prediction,and the forest fire in March 2020 in Qiaowa Town,Muli County,Sichuan Province is used as the validation example to simulate forest fire spread based on constant wind field,wind field of actual weather station at the forest fire site and predicted wind field of WRF respectively,and the boundary length and area of forest fire spread area are selected as the comparison index for The results showed that the wind field based on the WRF predicted wind field and the wind field of the forest fire were simulated.The results show that the simulation results based on the predicted wind field of WRF and the simulation results based on the wind field of the actual weather station are very close,and the difference in the area of forest fire spread between the two conditions is only 3.2 percentage points,and the difference in the perimeter length is 4.6 percentage points,which shows that the proposed method of forest fire spread simulation based on WRF and CA can be used for the practical application of forest fire spread prediction and provide a scientific basis for forest fire rescue.