步长对白栎和马尾松凋落物含水率预测精度的影响

Effect of step length on the prediction accuracy of the moisture content of forest litter of Quercus fabri and Pinus massoniana

【目的】凋落物含水率的高低,决定林火发生的可能性和火行为指标,对预测森林火险具有十分重要的意义。缩短预测步长是提高含水率预测精度的重要途径,但步长过小,则会耗费大量人力物力。因此,选择合适的步长来预测凋落物的含水率,对于研究步长对预测凋落物含水率的准确度是必不可少的。【方法】本研究以西南林区常见的阔叶树白栎和针叶树马尾松为研究对象,在森林防火期以1 h为间隔监测凋落物含水率,得到不同的步长含水率变化情况,对变化情况的影响因子进行分析,选择气象要素回归法和直接估计法建立预测模型,确定最适宜的步长预测含水率。【结果】1)监测步长低于4 h时,基本能够反映凋落物含水率日动态变化情况。2)随着采样时间延长,空气温度和相对湿度与凋落物含水率的相关性先增加后下降。白栎凋落物含水率动态变化主要受空气温湿度的影响,而马尾松凋落物含水率的变化则主要受风速的影响。3)对于气象要素回归法,两种不同类型的凋落物选择不同步长时的含水率预测精度都没有显著差异;对于直接估计法,白栎凋落物含水率以1 h为步长时预测精度显著高于其他步长,而不同步长对马尾松凋落物则没有显著影响。【结论】本研究得到的步长适用于西南林区典型凋落物含水率预测,有利于理解凋落物含水率动态变化,对构建凋落物含水率预测模型和火险预测具有重要的现实意义。

【Objective】The moisture content of the forest litter determines the possibility of forest fire and a series of fire behavior indicators,which is very import for fire risk.Shortening the prediction step is an important way to improve the accuracy of moisture content prediction,but if the step is too small,it will consume a lot of manpower and materials resources.Therefore,it is very important to study the effect of step length on the prediction accuracy of litter moisture content and select appropriate step length to predict litter moisture content.【Method】This study took the common broad-leaved oak and coniferous in the southwest forest region as the research object,monitored the moisture content of litter at the interval of one-hour during the forest fire prevention period,obtained the moisture content changes with different step lengths,analyzed the influence factors,and selected the meteorological element regression model and the direct estimation model to predict moisture content,and determined the most suitable step length.【Result】1)When the monitoring step length was less than four-hour,it reflected basically the daily dynamic change of litter moisture content.2)With the longer the sampling time,the correlation coefficients between air temperature,relative humidity and litter moisture content first increased and then decreased.The dynamic change of the moisture content of litter of the Quercus fabri was mainly affected by the air temperature and relative humidity,while the wind speed mainly affected the dynamic change of the moisture content of litter of the Pinus massoniana.3)For the meteorological element regression model,different steps length were selected to predict the moisture content of litter,and there was no significant difference in the prediction accuracy of the two types of litter.For the direct estimation method,the prediction accuracy of the moisture content of the Q.fabri litter was significantly higher than that of other steps when the step length was one-hour,while the step length has no significant impact on the P.massoniana.【Conclusion】Through this study,the appropriate length of typical litter moisture content prediction in the southwest forest region is obtained,which is helpful to understand the dynamic change of litter moisture content,and has important practical significance for the construction of litter moisture content prediction model and fire risk prediction research.