Moisture Content Prediction Model Of Fine Dead Fuel In Typical Forests And Ditch Meadows In Daxing'an Mountains

The moisture content of forest fuel is one of the important indicators of forest fire forecasting.The changing law of the moisture content of dead fuel of forest land is of great significance for forecasting the occurrence of forest fires and controlling forest fires.Building a forest fuel moisture content model is a key technology for forest fire prediction and forecasting.Meteorological factors are the main influencing factors that control the change in the moisture content of the dead fuel in the forest.The systematic and in-depth study of meteorological factors on the changes in the moisture content of fine dead fuel in wild forests is very important to construct a prediction model for the moisture content of fine dead fuel in forests.At present,there are some forecast models for forest fire prediction and forecast based on meteorological factors,but the calculation method and application range of the forecast model are very limited,there is still a long way to go for the forest fire weather warning and monitoring system in China.The Daxing 'an Mountains area is the main area of the forest ecosystem in northern China,and it is also an area with a high incidence of forest fires.Forest fires are the main factors affecting the balance of the ecosystem in this area.This paper focuses on the key technical issues of forest fire prediction and forecast in two forest fire danger periods in the Daxing 'an Mountains.The moisture content data of fine dead fuel collected from May and June of 2015 to 2016 in typical forests(L.gmelinii-B.platyphylla mixed forest,L.gmelinii forest,and Q.mongolica forest)and ditch meadows in Daxing 'an Mountains and the meteorological data outside the forest were used to construct a prediction model for the moisture content of fine dead fuel based on meteorological factors.The model construction method was easy to popularize,and the test results had high accuracy,providing a theoretical basis for forest fire forecasting in Daxing 'an Mountains.The main findings are as follows:(1)Prediction model for real-time changes of moisture content of fine dead fuel in typical forests and ditch meadows in Daxing 'an MountainsAccording to the statistical analysis method,under the assumption that the meteorological factors were consistent at a certain time on the regional scale,the moisture content data of the fine dead fuel in each forest stand and ditch meadows and the meteorological data from the meteorological station outside the forest were used to analyze by regression,correlation analysis,least squares method and field theory thought to construct the prediction model of real-time changes in moisture content of fine dead fuel based on meteorological factors in typical forests and ditch meadows in Daxing 'an Mountains.That was,the moisture content of the fine dead fuel at time t,the temperature difference between time t and t-1 outside the forest,the relative humidity difference between time t and t-1 and the rainfall between time t and t-1 were used to estimate the moisture content of the fine dead fuel at time t+1.The least square method was used to estimate the parameters in the corresponding models of typical forests and ditch meadows.The model test results showed that the accuracy rates of real-time change prediction model of the moisture content of fine dead fuel in typical forests(L.gmelinii-B.platyphylla mixed forest,L.gmelinii forest,and Q.mongolica forest)and ditch meadows were 91.1%,90.0%,91.0%and 81.0%,respectively(relative error did not exceed 5%),the average accuracy rate was 88.3%.The research results showed that the constructed real-time change prediction model for the moisture content of fine dead fuel had a good prediction effect,and the model had good practicability,and could be used in the forest fire real-time forecasting system in Daxing 'an Mountains.(2)Prediction model of moisture content of fine dead fuel based on rainfall in typical forests and ditch meadows in Daxing'an MountainsAssuming that the meteorological factors were consistent at a certain time on the regional scale,regression analysis,correlation analysis and least square method were used to construct a real-time prediction model for the moisture content of fine dead fuel.On the basis of this model,a prediction model of the moisture content of fine dead fuel based on rainfall in typical forests and ditch meadows in Daxing 'an Mountains was derived.The prediction model could describe the variation of the moisture content of fine dead fuel in the forest under constant rainfall intensity when the meteorological factors outside the forest were constant.The main meteorological factors selected included temperature,relative humidity,wind speed and rainfall outside the forest.The least square method was used to estimate the parameters in the corresponding models of the typical forests and ditch meadows.The model test results showed that the accuracy rates of the prediction models for the moisture content of fine dead fuel based on rainfall in typical forests(L.gmelinii-B.platyphylla mixed forest,L.gmelinii forest,and Q.mongolica forest)and ditch meadows were 87.7%,83.9%,93.9%and 81.8%,respectively(relative error does not exceed 5%).The average accuracy rate was 86.8%.The results of the study showed that the constructed rainfall-based prediction model for the change in moisture content of fine dead fuel was correct,and it was feasible to describe the "moisture absorption"process of fine dead fuel in the wild forest.This model had a better effect on predicting the change of moisture content of fine dead fuel when the rainfall duration of forest land was long.(3)Prediction model for the daily change of moisture content of fine dead fuel in typical forests and ditch meadows in Daxing 'an MountainsAssuming that the meteorological factors were consistent at a certain time on the regional scale,using regression analysis,correlation analysis theory,least squares method and interpolation theory,and meteorological data from the ecological protection station outside the forest,a prediction model based on meteorological factors for the daily variation of the moisture content of fine dead fuel in typical forests and ditch meadows in Daxing 'an Mountains was constructed.The model included multiple sub-models:the daily average value model of moisture content of fine dead fuel,the daily maximum value model of moisture content of fine dead fuel,the daily minimum value model of moisture content of fine dead fuel,the daily maximum value occurrence time model of moisture content of fine dead fuel,the daily minimum value occurrence time model of moisture content of fine dead fuel,and the daily variation curve model of moisture content of fine dead fuel.That was,using the meteorological factors outside the forest on that day to estimate the daily maximum and minimum moisture content of the fine dead fuel in the forest on the next day,and the time when the daily maximum and daily minimum values appeared,and then used the spline interpolation theory to give the 24 hourly fine dead fuel moisture content change curve,and the least square method to estimate the parameters of the corresponding model of each typical forests and ditch meadows.The test results showed that the accuracy rates of the prediction models for the daily changes in the moisture content of fine dead fuel in typical forests(L.gmelinii-B.platyphylla mixed forest,L.gmelinii forest,and Q.mongolica forest)and ditch meadows were 80.0%,83.3%,83.3%and 83.3%,respectively(relative error does not exceed 10%),and the average accuracy rate was 82.5%.The research results showed that the established prediction model for the daily variation of the moisture content of fine dead fuel was feasible,and the model had good practicability and could be used in the forest fire day forecasting system.In summary,the prediction model for the moisture content of the fine dead fuel in typical forests and ditch meadows in Daxing 'an Mountains was constructed and simulated on a time scale,revealing the changing law of the moisture content of the fine dead fuel on the forest surface.The research results will provide important data basis and theoretical support for the prediction and forecast of forest fires during the two fire prevention periods in the Daxing'an Mountains,and will be of positive significance for the protection of forest ecosystems in north China.