论文部分内容阅读
收集2006年至2011年4-10月每日的雷击定位数据、气象数据、可燃物数据及雷击引发的火灾数据,结合黑龙江大兴安岭地区的地形特征及森林雷击火发生的特点,引进加拿大火险天气指标(FWI),拟合出黑龙江大兴安岭林区雷击引发持续闷燃火的概率模型。对该模型进行HL检验,结果表明拟合效果良好,模型拟合精度为55%。2006-2011年每日雷击火发生数量的观测值与预测值的拟合关系为N(观测)=0.8368 N(预测),R2=0.732。基于该模型开发出黑龙江大兴安岭森林雷击火预测系统,系统试运行结果表明,该系统在雷击火高发季节,预测雷击火发生的准确率达31.5%。
Lightning strike data, meteorological data, combustibles data and lightning data from April 2006 to April 2011 were collected daily. Based on the topographic characteristics of Daxinganling area in Heilongjiang Province and the characteristics of forest fires, a fire risk index of Canada was introduced (FWI), the probability model of continuous smoldering fire caused by lightning strike in Daxing’anling forest area of Heilongjiang Province was fitted. The HL test results show that the fitting effect is good, the model fitting accuracy is 55%. The fitted relationship between the number of daily lightning fires in 2006-2011 and the predicted value is N (observation) = 0.8368 N (prediction), R2 = 0.732. Based on the model, a forest lightning fire forecasting system was developed in Daxing’anling, Heilongjiang Province. The results of system commissioning showed that the system was able to predict the occurrence of lightning fires by 31.5% during the peak season of lightning strikes.