【作者】 孙艳玲；

【导师】 刘洪斌；

【作者基本信息】 西南农业大学 ， 土壤学， 2004， 硕士

【摘要】 随着社会经济的快速发展，环境质量的恶化正在威胁着人类的生存与发展。其中，土壤侵蚀和非点源污染问题正成为目前人们关注的话题。降雨径流模拟是土壤侵蚀、非点源污染、地表水量平衡计算等众多研究领域的非常重要的技术手段。无论在流域侵蚀产沙过程中，还是在非点源污染运移过程中，水都是主要因素，所以在以流域为尺度进行土壤侵蚀研究和非点源污染研究时，流域的水文过程是研究的主要问题。 随着计算机技术和一些交叉学科的发展，流域水文模拟的研究方法产生了根本性的变化。其突出趋势主要反映在计算机技术、空间技术、遥感技术等的应用方面，分布式物理模型被广泛提出，遥感(RS)、地理信息系统(GIS)在水文模拟中的应用给传统的研究方法带来了创新，特别是数字高程模型(DEM)技术的完善为数字水文学的发展和数字水文模型的诞生提供了坚实的基础。数字高程模型(DEM)存储了丰富的地形信息，通过GIS可以提取流域的基本特征，包括下垫面特征、水系、河网等，并可以依据河网等级确定流域的边界，进行子流域的划分。为数字水文模拟提供了平台。 本研究应用地理信息系统和WMS模型系统，以三峡库区晏家河流域为例，由DEM提取流域数字特征，再与降雨径流模型有机结合进行了降雨径流数字模拟研究，主要的研究内容和结果包括以下几个方面： (1)建立了流域空间数据库。本研究以Windows 2000 Professional为操作系统，采用ESRI公司的地理信息系统产品Arc／Info 8为工作平台，建立了晏家河流域的空间数据库，包括流域数字高程模型、流域水系、土壤、土地利用等空间信息。流域内土地利用类型主要以水田、早地为主，共占流域面积的68.05％。林地分布也较多，占流域面积的27.47％。土壤类型主要以沙夹泥田、黄沙泥土为主，共占流域面积的57.59％。其中，沙夹泥田占流域面积的41.47％，主要分布在流域的中部。流域内＜3°和7°-25°的坡地分布较多，分别占了流域总面积的41.58％、34.5％，而＞25°坡地面积为13.59Km~2，所占比例为18.74％。 (2)构建了数字流域。在数字高程模型(DEM)的基础上，进行了流域河网的生成、子流域的划分以及流域边界的生成研究，并进行了与地形相关、具有水文学意义的部分参数的提取。本研究将晏家河流域离散成为6个子流域，计算得到子流域1至子流域6的面积(单位为km~2)分别为19.72、7.31、9.03、18.55、10.04、7.91，占整个流域的百分比(％)分别为27.18、10.07、12.44、25.57、13.84、10.90。在6个子流域中，流域长度以子流域4最大，为8.18km，平均高程以子流域2最高，为588.76m，河道长度以子流域4最长，为11.5km，河道坡度以子流域2最大，为0.0814。 (3)采用降雨径流模型进行了地表径流模拟。在生成数字流域的基础上，首先建立每一个子流域的降雨径流模型；然后采用马斯京根方法，建立河网汇流模型。对于产生的每一个集西南农业大学硕士学位论文摘要水子流域，应用SCS径流曲线数方法计算产流，应用SCS无因次单位线计算坡面汇流。模拟结果表明，研究区内各个子流域和整个流域的地表径流的逐日变化大体上相同且追随降雨的变化，并表现出一定的滞后效应。各个子流域和整个流域的地表径流数值均在5月12日达到最大值，而降雨量在这天也达到最多，为44.5mm。子流域!至子流域6的日平均径流量(mm)分别为1.39、0.49、0.58、1.27、0.72、0.63，月总径流量(mm)分别为43.04、15.33、18.08、39.51、22.44、19.47。整个流域出口的日平均径流量为5.Ogrnm，总径流量为157.slmm。 本研究的降雨径流模拟技术是一种具有物理基础的包含大容量信息的现代模拟技术。流域水系、分水线及地形参数信息是根据地形由计算机自动生成。子流域划分的方法具有一定的物理基础，避免了以往水文模型中根据降雨站划分子流域单元的缺陷。在这些子流域上根据各自的下垫面特性分别构建降雨径流模型，然后采用马斯京根方法，建立河网汇流模型，从而实现了流域水文模型从集总式向分布式的转化。更多还原

【Abstract】 With the rapid development of economy, the deterioration of the environmental quality is threatening the human survival and development in the world. Now soil erosion and non-point source pollution are becoming the attentive topic among the people. Rainfall-runoff simulation is the very important technique for soil erosion, non-point source pollution, surface water balance calculation and some other study fields. Water is always the main factor not only during the course of soil eroding but also during the process of non-point source pollution moving. So hydrological process of watershed is the central problem in the study of soil erosion and non-point source pollution with watershed in scale.With the development of computer technology and some cross knowledge, the study method of watershed hydrological simulation has changed ultimately. The outstanding trends are that computer technology, spatial technology and remote sensing technology are applied in hydrological model and distributed physical hydrological model is brought about widely. In the hydrological simulation, RS and GIS technology bring innovation for the traditional study methods. Digital elevation model (DEM) technology provides stable foundation for the development of digital hydrology and the naissance of the digital hydrological model in particular. DEM has abundant information about terrain, and we can extract the watershed basic characters, which include land surface characters, drainage network et al based on GIS, and also can make certain the watershed boundary and set off sub watersheds from DEM, which provide the flat roof for digital hydrological simulation.Taken Yanjia river watershed as an example, this study extracted watershed digital characters from DEM, and simulated rainfall-runoff with the rainfall-runoff model based on GIS and WMS. The main contents and results include:(1) The building of the spatial database of watershed. Windows 2000 Professional was used here for operating system, and Arc/Info 8 from ESRI Company was used for the work flat floor. This study builds the spatial database of Yanjia river watershed, which includes digital elevation model, watershed drainage network, soil, land use et al. In the Yanjia river watershed, paddy field and dry land are the main types in land use, which are 68.05% of the whole watershed. And woodland is 27.47% of the whole watershed. The primary soil in the watershed include sand land, yellow land, which is 57.59% of the whole watershed. And sand land is 41.47% of the whole watershed, which mainly distributes in the middle of the study area. There are many areas that slope is < 3° and slope is between 7° and 25°, which is 41.58% and 34.5% of the whole watershed respectively. While in the study area there are 13.59 Km2 that slope is >25°, which is 18.74% of the whole watershed.(2) The construction of the digital watershed. This study extracted watershed drainage network, set off sub-watersheds and made certain watershed boundary. And this study also extracted some parameters about hydrology and terrain based on DEM. This study set off six sub-watersheds from the Yanjia river watershed, and calculated the areas of each sub-watershed. The areas from the first to the sixth watershed are 19.72 km2, 7.31 km2, 9.03 km2, 18.55 km2, 10.04 km2, 7.91 km2 respectively, and the percentage of each sub-watershed from the first to the sixth is 27.18%, 10.07%, 12.44%, 25.57%, 13.84%, 10.90% respectively. Among the six sub-watersheds, the forth is the biggest about watershed length, which is 8.18 km, the second sub-watershed is the highest in average elevation, which is 588.76m, the forth sub-watershed is the longest about drainage network length, which is 11.5 km, the second is the biggest about drainage network slope, which is 0.0814.(3) The rainfall-runoff simulation applying rainfall-runoff model. Firstly, this study set up rainfall-runoff model in each sub-watershed, and then set up conflux model using Muskingum method. In each sub-watershed from DEM, we used SCS runoff curve更多还原