黄河黑山峡无植被区的地表组成物质遥感信息模型
收稿日期: 2008-07-17
修回日期: 2008-11-27
网络出版日期: 2009-07-25
基金资助
国家科技支撑计划(2006BAB07);绿洲生态教育部省部共建重点实验室开放基金
Research on mass ingredient model based on remote sensing technology in non-vegetation area of Heishan Gorge basin
Received date: 2008-07-17
Revised date: 2008-11-27
Online published: 2009-07-25
Supported by
国家科技支撑计划(2006BAB07);绿洲生态教育部省部共建重点实验室开放基金
王树东, 杨胜天, 温志群, 曾红娟, 王玉娟 . 黄河黑山峡无植被区的地表组成物质遥感信息模型[J]. 地理研究, 2009 , 28(4) : 1128 -1135 . DOI: 10.11821/yj2009040027
It is very important to extract various kinds of underlying surfaces for soil erosion model because of various contributions of soil, vegetation, desert and rock under the same natural condition. Currently, traditional classification and information extraction methods based on remote sensing data have been widely applied in eco-hydrologic process field. But due to similarity and complexity of spectrums of soil, rock and desert, it was hard to distinguish soil, desert and rock in the same area. Rock land mountain, desert and soil mountain are widely distributed in the middle and upper reaches of Yellow River basin. In this paper, real spectrums of rock, soil and desert measured in lab using ASD (Analysis Spectrum Device) are analyzed, the result indicates that they could be well distinguished. On account of complex topographic changes and underlying surface roughness, spectrums from Landsat TM 5 become more complex and uncertain, but characteristics of surface texture of the rock land mountain are obvious and could be well differentiated from that of soil mountain and desert. For problem-solving of spectral complexity, normalized spectral index (NSI) is presented: NSI=(R4+R3+R2-3×R1)/(R5-R1)(R1,R2,R3,R4 and R5 individually refer to reflectance of the Langsat TM bands from 1 to 5). Then, the rock land mountain index (RMI) is presented according to the characteristics of normalized spectral index and texture: RMI=(R4+R3+R2-3×R1)/(R5-R1)+Rt(Rt refers to homogeneity index of texture), and the result indicates that information extraction precision of rock land mountain is 82.7% through set of threshold. Finally, we analyze spectral normalized spectral information of desert and soil and establish desert-exposed soil difference model (DS-Def): =DS-Def=(R4-R1)/(R5-R1)+R1+R2, and the result indicates that desert information extraction precision is 73.1%, and that of exposed soil is 72.8%. The above results indicate that the information extraction precision is higher than that by methods of traditional classification.
[1] 何书金.中国西部典型地区土地利用变化对比分析.地理研究,2006,25(1):71~75.
[2] 程天文.黄土高原中、小流域降雨径流变化与人类活动影响.地理研究,1997,16(增刊):104~108.
[3] 沈振荣,张瑜芳. 水资源科学实验与研究.北京:中国科学技术出版社,1992.240~305.
[4] 张丽萍,张登荣,张锐波,等. 小流域土壤水蚀强度生态预测模型及实验模拟.自然灾害学报,2007,16(2):51~55.
[5] 张光辉. 黄河流域降雨侵蚀力对全球变化的响应.山地学报,2005,23(4):420~424.
[6] 刘宝元,谢云,张科利. 土壤侵蚀预报模型.北京:中国科学技术出版社,2001.2~15.
[7] 胡刚,吴永秋,刘宝元,等. 东北漫港黑土区切沟侵蚀发育特征.地理学报,2007,62(11):1165~1173.
[8] 王志强,刘宝元,海春兴. 土壤厚度对天然草地植被盖度和生物量的影响.水土保持学报,2007,21(4):164~167.
[9] Friedl M A. Relationships among remotely sensed data surface energy balance energy balance, and area-averaged fluxes over partially vegetated land surface.Journal Applied Meteorology,1996,35(11):2091~2103.
[10] 凌美华,刘振声.流域地物识别的多谱段遥感数据分析.地理研究,1982,1(3):39~43.
[11] Anys H, Bonn F, Merzouk A.Remote sensing and GIS based mapping and modeling of water erosion and sediment yield in a semi-arid watershed of Morocco. Geocarto International,1994,9(1):31~40.
[12] Wang G, Gertner G, Fang S, Anderson A B. Mapping multiple variables for predicting soil loss by geostatistical methods with TM images and a slope map. Photogrammetric Engineering and Remote Sensing, 2003,69 (8), 889~898.
[13] Adams J B, Smith M O, Johnsen P E.Spectral mixture modeling: A new analysis of rock and soil types at he Viking Lander-1 site. Journal of Geophysical Research, 1986, 91 (B8): 8098~8112.
[14] Langran K J. Potential for monitoring soil erosion features and soil erosion modelling components from remotely sensed data. Proceedings of IGARSS’83. IEEE, San Francisco, CA.1983.21~ 24.
[15] 赵英时.遥感应用原理与方法.北京:科学出版社,2003.25~26.
[16] 陈述彭,赵英时.遥感地学分析.北京:测绘科学出版社,1990.55~67.
[17] Campbell James B. Introduction to Remote Sensing. New York:the Guilford Press,1987:3~7.
[18] 李培军.用ASTER图像和地统计学纹理进行岩性分类.矿物岩石,2004,24(3):116~120.
[19] H RM.Filtering for texture classification comparative study. IEEE,1979,67(5):786~804.
[20] Yun Zhang. Optimization of building detection in satellite image by combining multi-spectral classification and texture filtering. ISPRS Journal of Photo Grammetry & Remote Sensing, 1999,54: 50~60.
[21] 吴均,赵忠明.利用基于小波的尺度共生矩阵进行纹理分析.遥感学报,2001,5(2):100~105.
[22] 陈君颖,田庆九.高分辨率遥感植被分类研究.遥感学报,2007,11(2):221~226.
[23] 许妙忠,余志惠.高分辨率卫星影像中阴影的自动提取与处理.测绘信息与工程,2003,28(1):20~22.
/
| 〈 |
|
〉 |