摘要
黑龙江省平原和山区交错带植被受长期的人为破坏,林相差,人为活动频繁,火险等级高,火灾频发。加强该区域可燃物含水率动态和预测研究,有利于提高火险预报准确性。以处于该交错区的黑龙江省庆安县典型地表死可燃物为研究对象,对其含水率和气象要素进行了动态观测,分析了影响含水率的因子,并以气象要素、FWI指标及两者的混合为预报因子分别建立了地表死可燃物含水率的预测模型。结果表明,可燃物不同含水率时,其影响因子不同,低含水率时,受湿度影响大;高含水率时,受降雨影响大。FWI指标,主要是FFMC与全范围的含水率相关,但与≥35%的含水率相关差。该指标可用于预测全范围的含水率,但误差大于气象要素回归法,不适合降雨后的含水率预测。FWI指标与气象要素混合建模对于绝大多数模型并没有提高精度。气象要素回归模型误差在〈35%时与FWI模型差异不显著,但对于全范围的含水率预测,其误差低于后者。对于该地区除红松林〈35%的含水率以外的地表死可燃物含水率的预测,应采用气象要素回归模型,MAE为2.0%～7.8%,平均5.4%;MRE为10.6%～28.1%,平均15.8%。对〈35%的红松林含水率预测,加入FWI指标能够改进预测精度,采用混合模型最好。
The plain and mountain ecotones in Heilongjiang province suffered from long time human disturbance, causing forest severely damaged, such as bad forest form and high fire danger rate. The enhancements on studying dynamic prediction of fuel moisture contents can increase fire danger forecast accuracy. By taking the typical dead land surface fuels in the ecotones in Qing＇ an county of Heilongjiang province as the tested objects, the moisture of and weather variables in the region were successively observed, and the factors affecting fuel moisture dynamics changes were analyzed. Fuel moisture prediction models were established by using weather variables, FWI indexes, and the two aspects variables as predictive factors, respectively. The results show that the different factors affected the fuel moisture at different levels; for the lower moisture, the air humidity was the major affecting factor while the precipitation was for higher moisture; FWt indexes, mainly FFMC, were correlated with the fuel moisture evaluated for full moisture range, but not for moisture ≥ 35%, which indicated that the indexes can be used for predicting fuel moisture for full range, but its accuracy was lower than that of vapor exchange models; the vapor exchange models have lower errors than FWI models for full range fuel moisture prediction but not for 〈 35% moisture prediction; the model accuracies were not improved when FWl indexes were incorporated; the vapor exchange models should be used for moisture prediction for all fuels in the region except 〈 35% moisture prediction of fuel in Korean pine stand, which accuracy was MAE 2.0%-7.8%, averaged 5.4%, and MRE 10.6%～28.1%, averaged 15.8%. For the prediction of fuel moisture 〈 35% in Korean pine stand, mixed variable models was the best.
出处
《中南林业科技大学学报》
CAS
CSCD
北大核心
2014年第12期27-34,共8页
Journal of Central South University of Forestry & Technology
基金
国家林业公益性行业科研专项(201204508)
关键词
农林交错区
地表死可燃物
可燃物含水率
黑龙江省庆安县
Forest-agriculture ecotones
land surface dead combustible fuel
fuel moisture content
Qing＇an county of Heilongjiangprovince
