北京山区森林枯落物层水文过程模拟研究

发布时间：2018-01-27 11:30

本文关键词： 北京山区枯落物层水文过程模拟研究　出处：《西北农林科技大学》2017年硕士论文　论文类型：学位论文

【摘要】：森林枯落物作为森林土壤层的重要组成部分,具有涵养水源、阻延地表径流等水文生态功能。以往的研究多数重点关注枯落物层的持水蓄水能力,无法真正模拟枯落物层在森林坡面的水文过程,不能明确了解枯落物层的水文生态效益。针对上述问题,本文以北京山区典型林分枯落物为研究对象,采用“室内浸泡+人工模拟降雨”相结合的方法,通过设置不同坡度和不同降雨条件,分析不同林分枯落物在降雨过程中的动态水文过程并建立相应的过程模型,探讨枯落物截留、入渗、径流等水文分量对坡度和降雨因子的响应机制,研究结果可为准确评价森林水文功能、深刻理解森林坡面产流机制提供理论和数据支撑。全文主要研究结果如下:(1)在浸泡实验中,4种典型林分枯落物的最大持水量范围在2.27~25.94 t·hm-2之间,未分解层和半分解层枯落物的最大持水量、最大持水率、有效拦蓄量、有效拦蓄率和有效拦蓄深度等均表现出相同的规律:栓皮栎松栎混交油松侧柏。除油松和侧柏之间最大持水量和最大持水率无显著性差异,其余各组林分之间均表现出显著性差异。各林分枯落物的动态持水量与浸泡时间满足对数函数关系,吸水速率与浸泡时间呈幂函数关系。(2)人工降雨试验中,恒定雨强和变化雨强条件下枯落物的降雨截留过程都经历了三个阶段:快速截留、稳定增长和雨后滴落。在变化雨强试验中,栓皮栎枯落物层的瞬时截留速率波动较大,较油松表现出更好的“蓄水-排水”功能。(3)栓皮栎的最大截留量Cmax和最小截留量Cmin的范围分别为1.56~3.22 mm和1.08~1.68 mm,油松的Cmax和Cmin的范围为1.3~2.73 mm和0.72~1.14 mm。坡度因子对两种枯落物的Cmax和Cmin均有显著影响,雨强因子对两树种的Cmax有显著影响,对栓皮栎的Cmin有显著影响,但对油松的Cmin无显著影响;变雨强对两种枯落物的Cmax和Cmin均无明显影响。(4)根据动态实验数据,提出了适用于次降雨条件下基于坡度和降雨因子的枯落物截留动态模型,相对均方根误差RRMSE普遍小于20%,模型精度良好。此外,提出了基于坡度和降雨因子的枯落物雨后滴落量动态模型,相对均方根误差RRMSE普遍小于10%且纳什效率系数NSE普遍大于0.9,模型模拟精度很好。相比浸泡实验,人工降雨实验可以更真实、更准确地描述枯落物层蓄水持水的物理过程。(5)在次降雨中,枯落物截留量占比最小,小于5%,侧向流占比次之(3.59%~44.58%),垂向流占比最大(53.90%~99.17%),由于枯落物结构的差异,栓皮栎的侧向流占比明显高于油松。坡度和雨强因子对各水文分量的分配比例有显著的影响,变雨强对降雨分配特征没有影响。在变雨强试验中,栓皮栎枯落物较油松枯落物表现出更好的径流调节作用。
[Abstract]:Forest litter, as an important part of forest soil layer, has the hydroecological functions of conserving water source and delaying surface runoff. Most previous studies have focused on the water holding capacity of litter layer. The hydrological process of litter layer on the forest slope can not be really simulated, and the hydro-ecological benefits of litter layer can not be clearly understood. In view of the above problems, this paper takes the typical forest litter in Beijing mountainous area as the research object. By using the method of "indoor immersion artificial simulation rainfall", different slope and different rainfall conditions were set up. The dynamic hydrological processes of litter in different stands during rainfall were analyzed and the corresponding process models were established to explore the response mechanism of hydrological components such as litter interception infiltration and runoff to slope and rainfall factors. The results can provide theoretical and data support for accurate evaluation of forest hydrological function and deep understanding of forest slope runoff production mechanism. The main results of this paper are as follows: 1) in immersion experiments. The maximum water holding capacity of litter in four typical stands ranged from 2.27 to 25.94 t 路hm-2, and the maximum water holding capacity, maximum water holdup and effective storage capacity of litter in undecomposed and semi-decomposed layers. There was no significant difference in maximum water holding capacity and maximum water holding rate between Pinus tabulaeformis and Platycladus orientalis except Pinus tabulaeformis and Platycladus orientalis. There were significant differences among the other groups. The dynamic water holding capacity of litter and the soaking time of each stand met the logarithmic function. The relationship between water absorption rate and soaking time is power function. 2) in the artificial rainfall experiment, the rainfall interception process of litter under the condition of constant rain intensity and changing rain intensity has undergone three stages: rapid interception. The transient interception rate of litter layer of Quercus variabilis fluctuated greatly in the experiment of changing rain intensity. Compared with Pinus tabulaeformis, it has a better function of "water storage and drainage". The ranges of maximum Cmax and minimum Cmin of Quercus variabilis were 1.56 ~ 3.22 mm and 1.08 ~ 1.68 mm, respectively. The ranges of Cmax and Cmin of Pinus tabulaeformis were 1.3 ~ 2.73 mm and 0.72 ~ 1.14 mm respectively. The slope factor had a significant effect on Cmax and Cmin of the two litter species. Rain intensity factor had significant influence on Cmax of two species, Cmin of Quercus variabilis, but not on Cmin of Pinus tabulaeformis. According to the dynamic experimental data, a dynamic model of litter interception based on slope and rainfall factors was proposed. The relative root mean square error (RRMSE) is generally less than 20 and the model accuracy is good. In addition, a dynamic model of litter drop after rain based on slope and rainfall factor is proposed. The relative root mean square error (RRMSE) is generally less than 10% and the Nash efficiency coefficient (NSE) is generally greater than 0.9.The simulation accuracy of the model is very good. A more accurate description of the physical process of water retention in litter layer. (5) in the rainfall, the litter interception ratio is the smallest, less than 5, and the lateral flow ratio is 3.59 and 44.58). The vertical flow ratio is the largest, 53.90% and 99.17%, due to the difference of litter structure. The ratio of lateral flow of Quercus variabilis was obviously higher than that of Pinus tabulaeformis. The slope and rain intensity factors had significant influence on the distribution ratio of each hydrological component, but the change of rain intensity had no effect on the distribution characteristics of rainfall. The litter of Quercus variabilis showed better regulation of runoff than that of Pinus tabulaeformis.
【学位授予单位】：西北农林科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：S714

【相似文献】