直接估计法预测不同层凋落物含水率的适用性分析

Applicability analysis of direct estimation method for predicting litter moisture content in different layers

【目的】森林不同层凋落物的含水率决定了林火垂直蔓延和地下火发生的可能性,建立不同层凋落物含水率预测模型具有重要意义。直接估计法作为目前应用最广的枯落物含水率预测方法,使用该方法预测不同层凋落物的含水率,并对其适用性进行分析,对凋落物含水率研究具有重要意义。【方法】以蒙古栎-红松混交林和白桦林下4层凋落物为研究对象,在春季防火期内以日为步长监测其水分动态变化情况,并使用直接估计法建立预测模型,分析该方法的适用性。【结果】1)蒙古栎-红松混交林和白桦林的OLn层和OLv层凋落物含水率并没有显著差异,而白桦林内OF层和OH层的凋落物含水率显著高于蒙古栎-红松混交林;2)两个林分内OLn层凋落物含水率动态变化与Olv层、OF层的含水率变化具有相关性,而与OH层不相关。OLn层和OLv层凋落物含水率动态变化与空气温度都不相关,与相对湿度、风速和降雨具有相关性,且随着距离采样时间的越长,相关性逐渐减弱;3)对于蒙古栎-红松混交林,4层凋落物不同含水率预测模型误差没有显著差异,但表层凋落物使用Nelson模型效果较好,而下层凋落物使用气象要素回归模型较好,Simard模型对于各层凋落物含水率预测均不适用。对于白桦林,气象要素回归模型预测效果优于直接估计法,但与Nelson模型都没有显著差异,随着距离表面越远,直接估计法预测凋落物含水率越不适用。【结论】在今后的研究中,应该对直接估计法中关键参数:平衡含水率和时滞等进行校正,提高直接估计法进行不同层凋落物含水率的预测精度,对凋落物含水率研究和森林火险预报具有重要意义。

【Objective】The moisture content of litter in different layers of forest determines the possibility of vertical spread of forest fire and underground fire.It is of great significance to establish the prediction model of moisture contents in different litter layers.The direct estimation method is currently the most widely used prediction method for litter moisture contents.Using this method to predict the litter moisture content of different layers and analyzing its applicability is important for the research on litter moisture contents.【Method】In this study,the four-layer litter under the mixed forest of Quercus mongolica-Pinus koraiensis and Betula platyphylla was used as the research object,and the daily water dynamic changes were monitored during the spring fire prevention period,the direct estimation method was used to establish the prediction model,and the applicability of the method was finally analyzed.【Result】1)There was no significant difference in the litter moisture content between the OLn layer and OLv layer of the mixed forest of Quercus mongolica-Pinus koraiensis and Betula platyphylla,while the litter moisture content of the OF layer and OH layer in the Betula platyphylla forest was significantly higher than that of the Quercus mongolica-Pinus koraiensis mixed forest.2)The dynamic changes of litter moisture content in the OLn layer of the two stands were correlated with the moisture content changes in the OLv and OF layers instead of the OH layer.The dynamic changes of litter moisture content in the OLn and OLv layers were related to relative humidity,wind speed and rainfall instead of air temperature,and the correlation gradually weakened with the extension of sampling time.3)For the Quercus mongolica-Pinus koraiensis mixed forest,the prediction model errors of litter moisture contents of the four layers showed no significant difference,while the Nelson model was better for the prediction of surface litter,and the meteorological element regression model was better for lower litter.The Simard model was not applicable to the prediction of litter moisture content of either layer.For the Betula platyphylla forest,the meteorological element regression model showed a better prediction effect than the direct estimation method,which was similar to the Nelson model.As the distance from the surface was farther,the direct estimation method became less suitable for predicting litter moisture contents.【Conclusion】In the future research,the key parameters in the direct estimation method,such as equilibrium moisture content and time lag,should be corrected to improve the prediction accuracy of the litter moisture content in different layers,which is of great significance to the study of litter moisture content and forest fire prediction.