Liu Yuanhao, Wang Fei, Zhang Shuyu, Ding Wenbin, Li Rongqi, Han Jianqiao, Ge Wenyan, Chen Hao, Shi Shangyu
College of Soil and Water Conservation Science and Engineering, Northwest A&F University, Yangling, 712100, Shaanxi Province, China.
College of Soil and Water Conservation Science and Engineering, Northwest A&F University, Yangling, 712100, Shaanxi Province, China; Institute of Soil and Water Conservation, Chinese Academy and Sciences and Ministry of Water Resources, Yangling, Shaanxi, 712100, China.
J Environ Manage. 2024 Nov;370:122455. doi: 10.1016/j.jenvman.2024.122455. Epub 2024 Sep 8.
Interception loss (IL) is an important process in the hydrological cycle within semi-arid forest ecosystems, directly affecting the amount of effective rainfall. However, the factors influencing IL during individual rainfall events remain to be quantified. This study collected rainfall, vegetation, and interception data during the 2022 and 2023 growing seasons in a typical black locust forest within the Zhifanggou watershed. It employed the Random Forest Regression (RFR) and back-propagation neural network (BPNN) methods to quantitatively evaluate the contribution rates of various factors to the IL and interception loss percentage (ILP). The IL among the 48 effective rainfall events was 172.05 mm, accounting for 19.54% of the rainfall amount. IL and ILP increased as the distance from the trunk decreased. During all rainfall events, both IL and ILP were significantly negatively correlated with the leaf area index (LAI) and canopy cover (CC); IL is significantly positively correlated with total rainfall (TR) and rainfall intensity (RI), while ILP is significantly negatively correlated with TR, RI, and rainfall duration (RD). The BPNN and RFR results indicated that rainfall, canopy, and tree characteristics contributed 43.06%, 44.79%, and 12.15% to IL, respectively, and 57.27%, 34.09%, and 8.63% to ILP, respectively. TR, CC, and LAI represented the primary influencing factors. Rainfall and canopy characteristics were the main factors affecting IL (ILP). As rainfall event magnitude increases, canopy contributions to IL and ILP decrease. In semi-arid areas, managing forest canopies to control IL helps address water imbalances in ecosystems.
截留损失（IL）是半干旱森林生态系统水文循环中的一个重要过程，直接影响有效降雨量。然而，单次降雨事件中影响截留损失的因素仍有待量化。本研究收集了纸坊沟流域典型刺槐林2022年和2023年生长季的降雨、植被和截留数据。采用随机森林回归（RFR）和反向传播神经网络（BPNN）方法，定量评估各因素对截留损失和截留损失率（ILP）的贡献率。48次有效降雨事件中的截留损失为172.05毫米，占降雨量的19.54%。截留损失和截留损失率随距树干距离的减小而增加。在所有降雨事件中，截留损失和截留损失率均与叶面积指数（LAI）和冠层覆盖度（CC）显著负相关；截留损失与总降雨量（TR）和降雨强度（RI）显著正相关，而截留损失率与总降雨量、降雨强度和降雨持续时间（RD）显著负相关。BPNN和RFR结果表明，降雨、冠层和树木特征对截留损失的贡献率分别为43.06%、44.79%和12.15%，对截留损失率的贡献率分别为57.27%、34.09%和8.63%。总降雨量、冠层覆盖度和叶面积指数是主要影响因素。降雨和冠层特征是影响截留损失（截留损失率）的主要因素。随着降雨事件强度的增加，冠层对截留损失和截留损失率的贡献降低。在半干旱地区，管理森林冠层以控制截留损失有助于解决生态系统中的水分失衡问题。
