旬河流域景观格局变化对泥沙输出的影响
梅嘉洺1,2,3, 刘 洋3,4, 岳朋芸3,4, 李 东1,2,3, 姜庆虎3, 刘 峰3

(1.西藏大学 青藏高原生态与环境研究中心, 拉萨 850000; 2.西藏大学 理学院, 拉萨 850000; 3.中国科学院 武汉植物园水生植物与流域生态重点实验室, 武汉 430074; 4.中国科学院大学, 北京 100049)

土壤侵蚀; 景观格局; SWAT模型; 植被恢复; 旬河流域

Effects of Landscape Pattern Change on Sediment Tranport in the Xunhe River Basin
MEI Jiaming1,2,3, LIU Yang3,4, YUE Pengyun3,4, LI Dong1,2,3, JIANG Qinghu3, LIU Feng3

(1.Research Center for Ecology and Environment of Qinghai-Tibetan Plateau, Tibet University, Lhasa 850000, China; 2.College of Science, Tibet University, Lhasa 850000,China; 3.Key Laboratory of Aquatic Botany and Watershed Ecology, Wuhan Botanical Garden, Chinese Academic of Sciences, Wuhan 430074, China; 4.University of Chinese Academy of Sciences, Beijing 100049, China)

soil erosion; landscape pattern; SWAT model; vegetation restoration; Xunhe River Basin

备注

为探究河流输沙量对景观格局变化的响应,以旬河流域为研究区,选取旬河上、中、下游12个典型流域,借助SWAT模型对各支流泥沙进行了模拟分析; 在此基础上,通过1995年和2015年遥感影像解译及景观指数分析,探究流域内部景观格局变化及其与河流输沙量的关系。结果 表明:SWAT在径流、泥沙模拟过程中,率定期和验证期的决定系数(R2)均大于0.75,Nash-Sutteliffe系数(Ens)为0.74～0.85,模型的适用性良好。通过遥感影像解译,旬河流域土地利用类型主要以林地和农用地为主。受“退耕还林还草”和“植树造林”等植被恢复措施的影响,2015年流域景观的优势化程度比1995年显著加深,景观类型之间的连通性趋于优良,斑块分布更加集中,景观格局朝着单一化方向发展,导致河流输沙量显著降低。其中,景观水平上的景观分离度指数(DIVISION)和类型水平上水体的斑块密度指数(PD5)对输沙量解释度最大。此外,相较于景观水平,以类型水平指数表征对输沙量的解释度,拟合效果优良。该研究可为旬河流域水土流失治理及减缓土壤侵蚀等方面提供一定的科学依据。

Landscape pattern change in a watershed is an important factor controlling sediment transportation of a river system. The Xunhe River, a major tributary of the Han River, was chosen as the study site. 12 typical watersheds in the upper, middle and lower reaches of the Xunhe River were selected to simulate and analyze the sediment transportation in each tributary using SWAT model. We interpreted remote sensing images from 1995 and 2015 to obtain land cover and calculate the landscape indices. The relationship between landscape pattern change and sediment transportation was examined. SWAT model performed well in simulating sediment transportation. The coefficients of determination(R2)in calibration and validation periods were both greater than 0.75. The Nash-Sutteliffe indices(Ens)were between 0.74 and 0.85. Forest and agricultural land were the two major land cover types in the Xunhe River watershed. Due to forestation and the conversion of agricultural lands into forests and grasslands, the landscape in 2015 had a significantly higher dominant degree, connection among landscape classes was stronger, distribution of patches was more concentrated than that in 1995. The landscape pattern was evolved to be more simplified, resulting in the significant reduction in river sediment transportation. The DIVISION of the landscape level and the PD5 of the class level for water were the two indices that contributed the most to sediment transportation. In addition, the class level indices performed better in explaining sediment transportation than the landscape level indices. This study can provide the scientific basis for soil erosion control and mitigation in the Xunhe River area.