A Modified Model of Direct Estimation Method for Fine Fuel Moisture Content Prediction by Considering Crown Density

Abstract

Moisture content of forest fine fuel is one of the most important factors for forest fire risk assessment and fire spread prediction. The weather parameters above the fuel surface used in Catchpole's existing Direct Estimation Model are not suitable for forecast. The crown density of the forest area in the model, which may has great influence to the fine fuel moisture content in the forest, is not considered. A series of outdoor experiments were carried out in order to study the effect of the crown density, including 3 simulated crown densities and 2 real tests. In the experiments, the collected Pinus Sylvestris needles in the Da Hinggan Mountains of China are used. The dynamic variations of the weights of the needles are captured in 0.5h or 1h interval, and the dry weights of them are measured after 24h kiln dry. During the experiments, the local meteorological data including the temperature, the humidity and the winds velocity, are recorded from the homepage of Weather China at the same time. Based on the experiments, a modified model for fine fuel moisture content prediction is achieved, in which the crown density is used to adjust the temperature and the humidity near the fuel surface. The comparison results show that under different conditions of crown densities, the Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE) of the modified model are decreased significantly, and the Mean Absolute Percentage Errors (MAPE) for different crown densities are all less than 6%. The contrast validation between forest field observation and the prediction values of the modified method indicates that the absolute errors are less than 10%. This modified method can effectively reduce the prediction errors due to different crown densities; meanwhile, the calculation is using meteorological forecast data directly, so it has better applicability.