基于遥感和GIS的洪水淹没范围估测与模拟研究

CORC > 海洋研究所 > 中国科学院海洋研究所 > 海洋地质与环境重点实验室

题名	基于遥感和GIS的洪水淹没范围估测与模拟研究
作者	毕京佳1,2
学位类别	硕士
答辩日期	2016-05-19
授予单位	中国科学院大学
授予地点	北京
导师	黄海军研究员
关键词	洪水淹没范围洪水痕迹遥感种子淹没算法 DEM
学位专业	环境科学
中文摘要	我国的洪水灾害发生十分频繁，每年因洪水灾害造成数百亿元的直接经济损失。洪水淹没作为洪水灾害最重要的性质，一直是洪水研究重点关注的问题，也是水文调查、工厂选址、城市规划等研究的重点关注对象。使用遥感手段解译提取洪水淹没范围相较于野外观测具有很大的优势，既节省了观测成本，又降低了观测风险。GIS作为一个功能强大的系统集成工具，既可以综合管理遥感、地形、水文等不同来源的数据，又可以提供强大的数据处理分析方法，实现洪水淹没模拟分析。本文针对洪水发生时刻卫星影像数据缺乏的情况，提出了一种延时估测洪水淹没范围的方法。首先采用植被指数法和土壤含水量变化法提取洪水发生前、后植被、土壤遗留下的变化和痕迹；然后利用DEM高程数据进行淹没推算，估测洪水淹没的最大范围。文章选取大通河流域洪水和五里河流域洪水作为研究实例，使用Landsat TM、ETM数据以及SRTM DEM数据进行试验。结果表明，在植被覆盖率高、植被种类复杂的平原地区，植被指数变化痕迹可用于估测洪水淹没范围；而在植被覆盖率低、植被种类单一的河谷地区，土壤含水量变化痕迹更加适用。将估算结果与现场调访数据或洪水发生后一天的影像数据对比显示，不同类型的地貌区选取相适宜的估算方法能够获得较好的洪水淹没范围估算效果。将估测结果同基于DEM的最高洪水位模拟结果对比，发现估测结果更加准确。最后文章针对此方法设计了一个洪水淹没范围估测与模拟系统，使用.NET框架下AE与IDL一体化开发技术实现了系统的部分功能，为同时使用到遥感与GIS技术的专业软件系统构建提供了一种可行的解决方案。
英文摘要	China is a country in which flood is one of the severe disaster and occurred frequently. As one of the most important properties of flood disaster, flood inundation has always been the focus of research on flood. Using remote sensing method to extract flood inundated area has great advantages compared with field observation, which not only saves the cost of observation, but also reduces the risk. GIS is a useful system integration tool that can manage remote sensing, topography, hydrology and other sources of data comprehensively, and can provide powerful data processing and analysis methods and simulation analysis. In this paper, a new method for estimating the flood inundation is proposed, which is based on the situation of lack satellite image data. Vegetation index method and soil water content method were used to extract the differences of vegetation and soil before and after the flood, and then estimate the maximum range of flood inundation by seed flooding algorithm using DEM data. The Landsat TM, ETM data and DEM data are used in this paper. The results showed that in plain region with high vegetation cover and more vegetation types, the vegetation index method can be used to estimate the flood inundated area; and in the river valley area with low vegetation cover and less vegetation types, the soil water content method is better. Compared the results with the data from the field survey and the data from the satellite image of the second day after the flood shows that the estimation method can be used to extract the flood inundation. Compared the simulation result based on DEM using 8-direction seed-spread algorithm, the estimation result had better accuracy. This method can be used in the study of historical flood which is lack of archive data. Finally, we designed a software system to accomplish this method called Flood Inundation Estimate and Simulate System. This System built by integration development technology of AE and IDL under .NET framework. The design provided a feasible solution for the professional software systems which will use remote sensing and GIS technology simultaneously.
内容类型	学位论文
源URL	[http://ir.qdio.ac.cn/handle/337002/112561]
专题	海洋研究所_海洋地质与环境重点实验室
作者单位	1.中科院海洋研究所 2.中国科学院大学
推荐引用方式 GB/T 7714	毕京佳. 基于遥感和GIS的洪水淹没范围估测与模拟研究[D]. 北京. 中国科学院大学. 2016.