基于IRS-6的理县森林资源调查研究
|
摘要
从大量的信息提取和空间决策研究方法中,遥感(RS)的强大图像处理技术和地理信息系统(GIS)的强大空间分析功能提供了强有力的技术支持。本文以“如何从IRS-6遥感影像中获取有用的森林资源信息”为主线,以理县森林资源调查为原型,主要研究了传统信息提取技术和基于新算法的信息提取技术的优缺点及对于IRS-6遥感影像的实用性。论文的主要研究内容和取得的主要研究成果为如下:
(1)多源空间数据的集成,为森林资源调查提供数据支持。
论文研究选取1:5万的IRS-6遥感图像作为调查底图,以历史森林资源调查数据作为参考,结合1:1万标准分幅影像图,以及部分林地征占、林地规划资料进行森林资源总体调查。从以上资料可得出,论文涉及多种GIS数据源,主要有MAPGIS格式、SHP格式和CAD格式,且比例尺不统一、投影类型不同等问题,针对数据源的多样化,主要研究了根据数据格式转换、坐标转换和投影变换这三种数据集成方法。
(2)传统信息提取技术和基于新算法的信息提取优缺点对比。
传统的信息提取技术主要指目视解译、非监督分类和监督分类,这些分类方法简单、易理解、易操作,但也缺乏实用性,特别是对于高山峡谷地区的理县;基于新算法的信息提取技术层出不穷,论文选取其中的分层综合分类法、光谱信息分类法和纹理分类法,但针对地貌类型为高山峡谷区的理县,研究各自的优缺点以及对于理县遥感影像的实用性和可靠性。
(3)森林资源调查结果的分析与评价,主要包括森林面积和蓄积总量分析、森林质量、结构和分布及动态变化。
理县林业资源丰富,林业用地占全县总土地面积的53%,活立木蓄积量2216万立方米,森林覆盖率为40.7%,而上年度森林资源调查的覆盖率为30.03%,可见森林覆盖率增加了近10%,且森林及林木的质量有所提升。从全县整体布局出发,林业用地主要沿河谷分布,随着海拔的升高树种和数量呈减少趋势。相对低海拔地区以灌丛、混交林、未成林地、未成林封育地、林业辅助生产用地为主;相对中高海拔地区以阔叶林和针叶林为主;高海拔地区主要以针叶林和疏林地为主。
从理县森林资源调查过程中,将遥感和GIS等空间信息技术引入,能够快速完成许多常规手段无法完成的工作,大大节约人力、财力和物力,提高工作效率,为调查研究和结果分析提供强大的技术支持。
From a large amount of information extraction and space policy research methods of remote sensing (RS)supplied powerful image processing technology and geographical information system (GIS) analysis of space capabilities to provide strong support. this article to the Remote Sensing Images from the IRS -6 how to extract useful information for forest resources, "one in charge in the LiXian county forest resources survey as archetypes, Main studied traditional information extraction algorithm based on new technology and information technology to take the strengths and weaknesses, and the IRS -6 remote sensing images. the applicability of the main contents of the study and research for the following :
(1) More than one source of space data integration, forest resources survey data provides research support.
the thesis summarizes , the IRS -6 Remote Sensing images which scale is 1:50000 for investigation. in history as a reference to forest resources survey data, the 1:10000 standard picture image maps and woodlands for part of the planning data on forests and woodlands resources from the survey. above information must be out of paper on gis data sources, mainly mapgis format, pattern CAD and format shp, And a lack of uniformity, the projection different types of problems, data sources, mainly studied according to data format conversion, the switch and the transformation of the three data integration.
(2) traditional information extraction algorithm based on new technology and information extraction strengths and weaknesses of the contrast.
Traditional information technology mainly means to extract by eyes can interpret and classified supervision and monitoring, these classification simple, easy to understand, casy to operate, but also lacks practicality, especially in the mountains in the area counties;information extraction algorithm based on new technology and tap in the hierarchical comprehensive system, the spectrum of information classification system, and texture, On the landscape but a type of mountain and the truth, the respective merits and shortcomings and to pay attention to the applicability of remote sensing images and reliability.
(3) Forest resources survey results of analysis and evaluation, including the forest area and cumulation of the analysis, the forest quality, structure and distribution and dynamic change.
Pay attention to the county level of forestry resources, forestry land on the whole land area of the total of 53%, standing forest stock 2216w cubic metres, the forest coverage is 40.7%, the annual forest resources of the coverage is 30.03%, the forest coverage rate of increase of nearly 10%, and the forest and forest quality has been promoted from the county set, the overall composition of the forestry land along the valley. the primary level up the species and number is reduced. Relatively low altitude berek area, make a forest, woodlands and yong afforestation land and forestry supplementary productivity is relatively high level areas ; in coniferous forests with broad-leaved forest and; high altitude, mainly in coniferous forests and woodlands.
From county forest resources for the investigation process and gis and remote sensing and space information technology, with the many general method can not finish the work and save the manpower, financial and material resources, enhancing work efficiency and for investigation and analysis of supply good technical support.
(1)多源空间数据的集成,为森林资源调查提供数据支持。
论文研究选取1:5万的IRS-6遥感图像作为调查底图,以历史森林资源调查数据作为参考,结合1:1万标准分幅影像图,以及部分林地征占、林地规划资料进行森林资源总体调查。从以上资料可得出,论文涉及多种GIS数据源,主要有MAPGIS格式、SHP格式和CAD格式,且比例尺不统一、投影类型不同等问题,针对数据源的多样化,主要研究了根据数据格式转换、坐标转换和投影变换这三种数据集成方法。
(2)传统信息提取技术和基于新算法的信息提取优缺点对比。
传统的信息提取技术主要指目视解译、非监督分类和监督分类,这些分类方法简单、易理解、易操作,但也缺乏实用性,特别是对于高山峡谷地区的理县;基于新算法的信息提取技术层出不穷,论文选取其中的分层综合分类法、光谱信息分类法和纹理分类法,但针对地貌类型为高山峡谷区的理县,研究各自的优缺点以及对于理县遥感影像的实用性和可靠性。
(3)森林资源调查结果的分析与评价,主要包括森林面积和蓄积总量分析、森林质量、结构和分布及动态变化。
理县林业资源丰富,林业用地占全县总土地面积的53%,活立木蓄积量2216万立方米,森林覆盖率为40.7%,而上年度森林资源调查的覆盖率为30.03%,可见森林覆盖率增加了近10%,且森林及林木的质量有所提升。从全县整体布局出发,林业用地主要沿河谷分布,随着海拔的升高树种和数量呈减少趋势。相对低海拔地区以灌丛、混交林、未成林地、未成林封育地、林业辅助生产用地为主;相对中高海拔地区以阔叶林和针叶林为主;高海拔地区主要以针叶林和疏林地为主。
从理县森林资源调查过程中,将遥感和GIS等空间信息技术引入,能够快速完成许多常规手段无法完成的工作,大大节约人力、财力和物力,提高工作效率,为调查研究和结果分析提供强大的技术支持。
From a large amount of information extraction and space policy research methods of remote sensing (RS)supplied powerful image processing technology and geographical information system (GIS) analysis of space capabilities to provide strong support. this article to the Remote Sensing Images from the IRS -6 how to extract useful information for forest resources, "one in charge in the LiXian county forest resources survey as archetypes, Main studied traditional information extraction algorithm based on new technology and information technology to take the strengths and weaknesses, and the IRS -6 remote sensing images. the applicability of the main contents of the study and research for the following :
(1) More than one source of space data integration, forest resources survey data provides research support.
the thesis summarizes , the IRS -6 Remote Sensing images which scale is 1:50000 for investigation. in history as a reference to forest resources survey data, the 1:10000 standard picture image maps and woodlands for part of the planning data on forests and woodlands resources from the survey. above information must be out of paper on gis data sources, mainly mapgis format, pattern CAD and format shp, And a lack of uniformity, the projection different types of problems, data sources, mainly studied according to data format conversion, the switch and the transformation of the three data integration.
(2) traditional information extraction algorithm based on new technology and information extraction strengths and weaknesses of the contrast.
Traditional information technology mainly means to extract by eyes can interpret and classified supervision and monitoring, these classification simple, easy to understand, casy to operate, but also lacks practicality, especially in the mountains in the area counties;information extraction algorithm based on new technology and tap in the hierarchical comprehensive system, the spectrum of information classification system, and texture, On the landscape but a type of mountain and the truth, the respective merits and shortcomings and to pay attention to the applicability of remote sensing images and reliability.
(3) Forest resources survey results of analysis and evaluation, including the forest area and cumulation of the analysis, the forest quality, structure and distribution and dynamic change.
Pay attention to the county level of forestry resources, forestry land on the whole land area of the total of 53%, standing forest stock 2216w cubic metres, the forest coverage is 40.7%, the annual forest resources of the coverage is 30.03%, the forest coverage rate of increase of nearly 10%, and the forest and forest quality has been promoted from the county set, the overall composition of the forestry land along the valley. the primary level up the species and number is reduced. Relatively low altitude berek area, make a forest, woodlands and yong afforestation land and forestry supplementary productivity is relatively high level areas ; in coniferous forests with broad-leaved forest and; high altitude, mainly in coniferous forests and woodlands.
From county forest resources for the investigation process and gis and remote sensing and space information technology, with the many general method can not finish the work and save the manpower, financial and material resources, enhancing work efficiency and for investigation and analysis of supply good technical support.
引文
[1]刘新东,杨福成,李荣平.2009.11.3S集成技术在森林资源调查中的应用.吉林林业科技.2009.11,38(6):30-33.
[2]吴见,彭道黎. 2009.8.土地沙化遥感信息提取技术研究进展.世界林业研究, 2009.8, 22(4):47-52.
[3]牛宝茹,刘俊蓉,王政伟.2005.1.干旱半干旱地区植被覆盖度遥感信息提取研究.武汉大学学报.信息科学版.2005.1.30(1):27-30.
[4]胡振琪,陈涛.2008.基于ERDAS的矿区植被覆盖度遥感信息提取研究——以陕西省榆林市神府煤矿区为例.西北林学院学报.2008.23(2):164-167.
[5]徐春迪,2008.3,基于ETM+土地利用/土地覆盖遥感信息提取研究——以宝鸡市金台、渭滨两区为例.宝鸡文理学院学报(自然科学版).2008.3.28(1):72-76.
[6]许辉熙,2008.6.<空间信息技术在水电开发工程预可研究中的决策支持—以黄河玛尔挡水电站为例>.博士论文.
[7]胡文英,角媛梅.2007.5.遥感图像纹理信息提取方法综述.云南地理环境研究.2007.519(3):66-70.
[8]张寿选,段洪涛,谷孝鸿.2008.基于水体透明度反演的太湖水生植被遥感信息提取.湖泊科学.2008.20(2).184-190.
[9]欧立业,何忠焕,马海州,郑金海.2008.1.基于知识的分层综合分类法在土地利用/土地覆盖遥感信息提取中的应用.测绘科学.2008.1.33(1):173-175.
[10]郜丽静,骆剑承,沈占锋,刘雯,陈生.2009.3.基于指数计算的多层次遥感信息提取模.地理与地理信息科学.2009.3.25(2):39-45.
[11]张芳,李鑫,陈佳楠.2009.8.卫星遥感影像目视解译在森林资源调查中的应用分析.内蒙古林业调查设计.2009.8.32(4):72-77.
[12]李春干,陈琦,谭必增.2009.4.基于卫星遥感数据空中抽样的大尺度森林资源动态监测.林业资源管理.2009.4.(2):106-110.
[13]刘晓双,黄建文,鞠洪波.2009.4高空间分辨率遥感森林参数提取探讨.林业资源管理.2009.4.(2):111-117.
[14]井艳波,张梁.2009.遥感图像(RS)在黑龙江省森林资源连续清查中的应用.林业科技情报.2009.41(2):32-33.
[15]王正兴,刘闯,Huete Alfredo.植被指数研究进展:从AVHRR-ND-VI到MODIS-EVI[J].生态学报,2003,23(5):979-987.
[16]郭凯,孙培新,李卫国.利用ERDAS IMAGINE从遥感影像中提取植被指数[J].西部探矿工程.2005.6.
[17]党安荣等.ERDAS IMAGINE遥感图像处理方法[M].北京:清华大学出版社,2002.
[18]郑元润,周广胜.基于NDVI的中国天然森林植被净第一性生产力模型.植物生态学报[J],2000,24(1):9-12.
[19]ATKINSON P M,TATNALL A R L.Introduction neural net-works in remote sensing [J]. International Journal of Remote Sensing,1997,18(4):699-709.
[20]MCFEETERS S K.The use of Normalized Difference Water In-dex (NDWI) in thedelineation of open water features[J].Inter-national Journal of Remote Sensing, 1996, 17(7): 1425-1432.
[21]BAUMQARTNER A,MAYER H,STEGER C,et al.Auto-matic road extraction based on multi-scale,grouping,and con-text[J]. Photogrammetric Engineering & Remote Sensing,1999, 65(7):777-782.
[22]蔡栋,魏巍.2009.基于决策树的水生植被遥感信息提取研究.安徽农业科学. 2009. 37(16): 7615-7616.
[23]张微,杨金中,王小红.基于特征地貌类型的湖州市第四纪地质遥感信息提取方法探讨.国土资源遥感.2009.9.15.81(3):45-48.
[24]王钦军,蔺启忠,黎明晓等,一种突出目标的多光谱遥感信息提取方法,光谱学与光谱分析,2009.4.4(29):1018-1022.
[25]张克军,遥感图像特征提取方法研究[D].西安:西北工业大学.2007.3-5.
[26]张龙其,刘平,邢光成.关于北京54和西安80坐标系相互转换的探讨[J].工程地球物理学报,2005,2(1):69-72.
[27]赵焕臣,许树柏.层次分析法---一种简易的新决策方法[M].北京:科学出版社,1986.
[28]赵英时,遥感应用分析原理与方法[M].北京:科学出版社,2003.
[29]钟耳顺,王康弘,宋关福,等.G1S多源数据集成模式评述[A].中国G1S年会(深圳)论文集[C],1999.
[30]曾宏达.基于DEM和地统计的森林资源空间格局分析.地理信息科学.2005,7(2):82-88.
[31]邬伦,刘瑜,张晶等.地理信息系统原理、方法和应用[M],北京:科学出版社,2001.
[32]王解先,王军,陆彩萍.WGS-84与北京54坐标的转换问题[J].大地测量与地球动力学,2003,23(3):70-73.
[33]汤国安,杨昕.ArcGIS地理信息系统空间分析教程[M].北京:科学出版社,2006.
[34]孙庆先,李茂堂,路京选,等.地理空间数据的尺度问题及其研究进展[J].地理与地理信息科学,2007,23(4):53-57.
[35]孙华.SPOT5在森林资源调查中的应用研究[D].长沙:中南林业科技大学,2006.
[36]崔铁军,郭黎.多源地理空间矢量数据集成与融合方法探讨[J].测绘科学技术学报,2007,24(1):1-4.
[37]戴昌达,姜小光,唐伶俐.遥感图像应用处理与分析[M].北京:清华大学出版社,2004.
[38]党安荣,王晓栋,陈晓峰,等.ERDAS IMAGINE遥感图像处理方法[M].北京:清华大学出版社,2002.
[39]邓辉.高精度卫星遥感技术在地质灾害调查与评价中的应用[D].成都:成都理工大学,2007.
[40]邓良基.遥感基础与应用[M].北京:中国农业出版社,2002.
[41]房世波,“3S”技术在川西干旱河谷退耕还林(草)监测与生态还林决策中的应用—以甘孜州九龙县为例[D].成都:成都理工大学,2004.
[42]冯秀丽,基于SPOT5影像的1:1万土地利用更新调查关键技术研究[D].杭州:浙江大学,2006.
[43]龚建周,张鹏,杨景荣,等.IKONOS数据在植被景观制图和景观空间格局分析中的应用[J].干旱区资源与环境,2004,18(2):130-135.
[44]焦超卫,,赵牡丹,汤国安等.基于GIS的植被空间格局特征与地形因子的相关关系-以陕西省耀县为例[M].水土保持通报,2005,12(6):19-23.
[45]田建林,雷平,石军南. Quick Bird影像目视判读在土地利用类型调查中的应用研究[J].四川林勘设计,2006(1):45-48.
[46]XueZ, LiSZ, Teoh E K. Ai2eigensnake: An Affine2invariantDeformable ContourModel for ObjectMatching[ J ]. Im age andV ision Com putting, 2002,20 (2): 77—84.
[47]Joy S M, Reich R M, Reynolds R T. A Non2parametricSupervised Classification of Vegetation Types on the KaibabNational Forest Using Decision Trees [ J ]. Int. J. Rem oteSensing,2003,24 (9): 1835—1852.
[48]FriedlM A, Mclver D K, Hodges J C F, et al. Global LandCoverMapp ing from MOD IS: Algorithms and Early Results[ J ].Rem ote Sensing of Environm ent, 2002, 11 (83) : 287—302.
[49]王伯荪,廖宝文,王勇军等.深圳湾红树林生态系统及其持续发展[M ].北京:科学出版社, 2002.
[50]李爽,张二勋.基于决策树的遥感影像分类方法研究[ J ].地域研究与开发, 2003, 22 (1) : 17—21.
[2]吴见,彭道黎. 2009.8.土地沙化遥感信息提取技术研究进展.世界林业研究, 2009.8, 22(4):47-52.
[3]牛宝茹,刘俊蓉,王政伟.2005.1.干旱半干旱地区植被覆盖度遥感信息提取研究.武汉大学学报.信息科学版.2005.1.30(1):27-30.
[4]胡振琪,陈涛.2008.基于ERDAS的矿区植被覆盖度遥感信息提取研究——以陕西省榆林市神府煤矿区为例.西北林学院学报.2008.23(2):164-167.
[5]徐春迪,2008.3,基于ETM+土地利用/土地覆盖遥感信息提取研究——以宝鸡市金台、渭滨两区为例.宝鸡文理学院学报(自然科学版).2008.3.28(1):72-76.
[6]许辉熙,2008.6.<空间信息技术在水电开发工程预可研究中的决策支持—以黄河玛尔挡水电站为例>.博士论文.
[7]胡文英,角媛梅.2007.5.遥感图像纹理信息提取方法综述.云南地理环境研究.2007.519(3):66-70.
[8]张寿选,段洪涛,谷孝鸿.2008.基于水体透明度反演的太湖水生植被遥感信息提取.湖泊科学.2008.20(2).184-190.
[9]欧立业,何忠焕,马海州,郑金海.2008.1.基于知识的分层综合分类法在土地利用/土地覆盖遥感信息提取中的应用.测绘科学.2008.1.33(1):173-175.
[10]郜丽静,骆剑承,沈占锋,刘雯,陈生.2009.3.基于指数计算的多层次遥感信息提取模.地理与地理信息科学.2009.3.25(2):39-45.
[11]张芳,李鑫,陈佳楠.2009.8.卫星遥感影像目视解译在森林资源调查中的应用分析.内蒙古林业调查设计.2009.8.32(4):72-77.
[12]李春干,陈琦,谭必增.2009.4.基于卫星遥感数据空中抽样的大尺度森林资源动态监测.林业资源管理.2009.4.(2):106-110.
[13]刘晓双,黄建文,鞠洪波.2009.4高空间分辨率遥感森林参数提取探讨.林业资源管理.2009.4.(2):111-117.
[14]井艳波,张梁.2009.遥感图像(RS)在黑龙江省森林资源连续清查中的应用.林业科技情报.2009.41(2):32-33.
[15]王正兴,刘闯,Huete Alfredo.植被指数研究进展:从AVHRR-ND-VI到MODIS-EVI[J].生态学报,2003,23(5):979-987.
[16]郭凯,孙培新,李卫国.利用ERDAS IMAGINE从遥感影像中提取植被指数[J].西部探矿工程.2005.6.
[17]党安荣等.ERDAS IMAGINE遥感图像处理方法[M].北京:清华大学出版社,2002.
[18]郑元润,周广胜.基于NDVI的中国天然森林植被净第一性生产力模型.植物生态学报[J],2000,24(1):9-12.
[19]ATKINSON P M,TATNALL A R L.Introduction neural net-works in remote sensing [J]. International Journal of Remote Sensing,1997,18(4):699-709.
[20]MCFEETERS S K.The use of Normalized Difference Water In-dex (NDWI) in thedelineation of open water features[J].Inter-national Journal of Remote Sensing, 1996, 17(7): 1425-1432.
[21]BAUMQARTNER A,MAYER H,STEGER C,et al.Auto-matic road extraction based on multi-scale,grouping,and con-text[J]. Photogrammetric Engineering & Remote Sensing,1999, 65(7):777-782.
[22]蔡栋,魏巍.2009.基于决策树的水生植被遥感信息提取研究.安徽农业科学. 2009. 37(16): 7615-7616.
[23]张微,杨金中,王小红.基于特征地貌类型的湖州市第四纪地质遥感信息提取方法探讨.国土资源遥感.2009.9.15.81(3):45-48.
[24]王钦军,蔺启忠,黎明晓等,一种突出目标的多光谱遥感信息提取方法,光谱学与光谱分析,2009.4.4(29):1018-1022.
[25]张克军,遥感图像特征提取方法研究[D].西安:西北工业大学.2007.3-5.
[26]张龙其,刘平,邢光成.关于北京54和西安80坐标系相互转换的探讨[J].工程地球物理学报,2005,2(1):69-72.
[27]赵焕臣,许树柏.层次分析法---一种简易的新决策方法[M].北京:科学出版社,1986.
[28]赵英时,遥感应用分析原理与方法[M].北京:科学出版社,2003.
[29]钟耳顺,王康弘,宋关福,等.G1S多源数据集成模式评述[A].中国G1S年会(深圳)论文集[C],1999.
[30]曾宏达.基于DEM和地统计的森林资源空间格局分析.地理信息科学.2005,7(2):82-88.
[31]邬伦,刘瑜,张晶等.地理信息系统原理、方法和应用[M],北京:科学出版社,2001.
[32]王解先,王军,陆彩萍.WGS-84与北京54坐标的转换问题[J].大地测量与地球动力学,2003,23(3):70-73.
[33]汤国安,杨昕.ArcGIS地理信息系统空间分析教程[M].北京:科学出版社,2006.
[34]孙庆先,李茂堂,路京选,等.地理空间数据的尺度问题及其研究进展[J].地理与地理信息科学,2007,23(4):53-57.
[35]孙华.SPOT5在森林资源调查中的应用研究[D].长沙:中南林业科技大学,2006.
[36]崔铁军,郭黎.多源地理空间矢量数据集成与融合方法探讨[J].测绘科学技术学报,2007,24(1):1-4.
[37]戴昌达,姜小光,唐伶俐.遥感图像应用处理与分析[M].北京:清华大学出版社,2004.
[38]党安荣,王晓栋,陈晓峰,等.ERDAS IMAGINE遥感图像处理方法[M].北京:清华大学出版社,2002.
[39]邓辉.高精度卫星遥感技术在地质灾害调查与评价中的应用[D].成都:成都理工大学,2007.
[40]邓良基.遥感基础与应用[M].北京:中国农业出版社,2002.
[41]房世波,“3S”技术在川西干旱河谷退耕还林(草)监测与生态还林决策中的应用—以甘孜州九龙县为例[D].成都:成都理工大学,2004.
[42]冯秀丽,基于SPOT5影像的1:1万土地利用更新调查关键技术研究[D].杭州:浙江大学,2006.
[43]龚建周,张鹏,杨景荣,等.IKONOS数据在植被景观制图和景观空间格局分析中的应用[J].干旱区资源与环境,2004,18(2):130-135.
[44]焦超卫,,赵牡丹,汤国安等.基于GIS的植被空间格局特征与地形因子的相关关系-以陕西省耀县为例[M].水土保持通报,2005,12(6):19-23.
[45]田建林,雷平,石军南. Quick Bird影像目视判读在土地利用类型调查中的应用研究[J].四川林勘设计,2006(1):45-48.
[46]XueZ, LiSZ, Teoh E K. Ai2eigensnake: An Affine2invariantDeformable ContourModel for ObjectMatching[ J ]. Im age andV ision Com putting, 2002,20 (2): 77—84.
[47]Joy S M, Reich R M, Reynolds R T. A Non2parametricSupervised Classification of Vegetation Types on the KaibabNational Forest Using Decision Trees [ J ]. Int. J. Rem oteSensing,2003,24 (9): 1835—1852.
[48]FriedlM A, Mclver D K, Hodges J C F, et al. Global LandCoverMapp ing from MOD IS: Algorithms and Early Results[ J ].Rem ote Sensing of Environm ent, 2002, 11 (83) : 287—302.
[49]王伯荪,廖宝文,王勇军等.深圳湾红树林生态系统及其持续发展[M ].北京:科学出版社, 2002.
[50]李爽,张二勋.基于决策树的遥感影像分类方法研究[ J ].地域研究与开发, 2003, 22 (1) : 17—21.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |