IMHE OpenIR  > 数字山地与遥感应用中心
川中丘陵区土壤侵蚀遥感评价方法研究
Alternative TitleResearch on the Remote Sensing Techniques for Soil Erosion Assessment in the Hilly Region of Central Sichuan
Language中文
曾珍
Thesis Advisor范建容
2014
Degree Grantor中国科学院研究生院
Place of Conferral北京
Degree Name硕士
Degree Discipline地图学与地理信息系统
KeywordRapideye Trmm 决策树 像元二分模型 空间耦合 Rusle
Abstract川中丘陵区位于长江上游,四川盆地中部,嘉陵江中下游地区,是我国重要的粮食产地,耕地面积较大,水力侵蚀较为严重,对该区域进行土壤侵蚀评价具有重要的科学意义。以往对于川中丘陵区的土壤侵蚀遥感评价效果不佳,因此本研究选取川中丘陵区典型区域李子溪流域为研究区,以RS、GIS为技术依托,获取土壤侵蚀影响因子,进行了川中丘陵区土壤侵蚀遥感评价方法研究与改进,并得到李子溪流域2009年土壤侵蚀状况,为川中丘陵区土壤侵蚀遥感评价提供了科学的技术支撑。主要研究内容与结果如下: 1、土壤侵蚀影响因子遥感提取方法:利用2011年TRMM数据和实测降雨数据进行日雨量精度检验,表明利用TRMM日雨量数据代替实测日雨量数据具有一定的可靠性,并采用日雨量模型计算得到2009年李子溪流域降雨侵蚀力;采用了面向对象分割法和决策树相结合的方法对土地利用进行计算机自动分类,结合人工目视解译法对错分像元进行修改,得到研究区土地利用类型信息;将Rapideye遥感数据与多光谱相机相结合,获取了研究区林地和撂荒地纯像元的NDVI值,采用像元二分模型得到李子溪流域林地和撂荒地的植被覆盖度,并根据植被覆盖度带入蔡崇法C因子计算模型中得到李子溪流域林地和撂荒地的植被和经营管理因子C值。 2、土壤侵蚀遥感评价方法研究:基于知识库与空间信息耦合法与基于模型参数化法对李子溪流域2009年的土壤侵蚀状况进行了遥感评价,从空间分布、侵蚀强度面积分布比例及精度检验等方面对两种方法得到的评价结果进行对比分析后认为,前者实现容易,参数获取简单,但因为未考虑水土保持措施对旱地土壤侵蚀的改善作用,该方法对旱地土壤侵蚀评价不合理,评价结果正确率较低;后者评价结果较为合理,评价结果正确率较高,但众多参数获取具有一定的难度,无法进行土壤侵蚀快速评价。 3、方法改进与土壤侵蚀评价:本研究将两种方法的优势结合,在基于知识库与空间信息耦合法评价的基础上,结合植被和经营管理因子确定修正系数,利用修正系数对旱地的评价结果进行修正并得到李子溪流域2009年土壤侵蚀状况,以实测土壤侵蚀强度对该方法得到的评价结果进行精度检验,评价结果正确率达91%,能够为川中丘陵土壤侵蚀快速评价提供科学的依据。遥感评价结果表明,该区域土壤侵蚀以微轻度侵蚀为主,强烈及极强烈侵蚀极少,二十年来政府采取的一系列生态措施取得了一定的成效,减缓了土壤侵蚀强度,改善了该区域的生态环境。
Other AbstractThe hilly region of central Sichuan which is the major grain producing area in China is located in the middle and lower reaches of the Jialing River, a tributary of the Yangtze River. Soil erosion is a serious concern in this area. The soil erosion evaluation in this area has important scientific significance. For the past studies, the results were poor. This thesis chose Lizixi basin which is a typical region in the hilly region of central Sichuan as the study area. Based on the techniques of RS (remote sensing) and GIS (geographical information system), the influence factors of soil erosion were acquired. The study of the methods of remote sensing assessment for soil erosion was fulfilled. The soil erosion information in Lizixi basin in 2009 was analyzed based on the improving method of remote sensing assessment for soil erosion. The main research content and results are: 1. The remote sensing extractions of the influence factors of soil erosion: The accuracy assessment was made between precipitation estimation from TRMM (Tropical Rainfall Measurement Mission) and precipitation data from rain gauges which are from 3 meteorological station in the study area in 2011 and the results showed that the daily precipitation data from rain gauges can be replaced by the daily precipitation data from TRMM data. Based on the TRMM daily precipitation data and daily precipitation model the rainfall erosivity of Lizixi basin in 2009 was calculated. Objected-oriented image analyzing and decision tree classification were used to complete the computer automatic classification. With the auxiliary of visual interpretation the error classifications were corrected and obtained the land use information of study area. Combination of the Rapideye remote sensing data and the ADC multi-spectral camera, the NDVI of pure pixel of forest and abandoned cultivated land were obtained and based on the binary pixel model the vegetation coverage of forest and the abandoned cultivated land were acquired. According to the water and soil conservation measures C factor calculation model, the C factor of forest and the abandoned cultivated land were calculated. 2. The study on the methods of soil erosion remote sensing assessment: The soil erosion assessment of Lizixi basin in 2009 was made based on the coupling of repository and spatial information and model parameterization approach. Compared analyzing two of the soil erosion remote sensing assessment methods from spatial distribution, area ratio of different erosion intensity and accuracy verification showed that the influence factors of the former method can be easily obtained and the former method can be easily fulfilled. However, based on the former method, the assessment of farm land was unreasonable because of the lack of the consideration of the water and soil conservation measures which can slow down the erosion intensity and the evaluation results had low accuracy. The assessment results based on the latter method was reasonable and had high accuracy, but the influences factors cannot be easily obtained and the fast soil erosion assessment cannot be implemented. 3. Method improvement and soil erosion assessment: In order to take the full use of the two methods based on the results of soil erosion monitoring method of coupling of repository and spatial information, the influence factor of water and soil conservation measures was joined to revise the soil erosion assessment result of the farm land. In this way, the erosion information of Lizixi basin was fulfilled and the results were verified by the investigation results. The accuracy can reach 91% so that it can provide a scientific method for fast evaluation of soil erosion in hilly region of central Sichuan. Based on this revised method, the assessment results showed that mainly area was none erosion or slight erosion. The area of severe erosion or very severe erosion can hardly been seen. A series of ecological measures taken in the last two decades ha
Document Type学位论文
Identifierhttp://ir.imde.ac.cn/handle/131551/8025
Collection数字山地与遥感应用中心
Recommended Citation
GB/T 7714
曾珍. 川中丘陵区土壤侵蚀遥感评价方法研究[D]. 北京. 中国科学院研究生院,2014.
Files in This Item:
File Name/Size DocType Version Access License
曾珍:川中丘陵区土壤侵蚀遥感评价方法研究(3628KB)学位论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[曾珍]'s Articles
Baidu academic
Similar articles in Baidu academic
[曾珍]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[曾珍]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 曾珍:川中丘陵区土壤侵蚀遥感评价方法研究.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.