基于RS的盘县煤矿区森林生态环境动态监测技术研究
|
摘要
近年来,遥感技术手段逐渐从定性转向定性与定量相结合。森林生态环境动态监测具有实时性、长期性、综合性和周期性等特点,要求调查成果必须具有较高的精度和准确度。采用遥感技术对森林生态环境有诸多优点,遥感为能森林生态环境动态监测提供技术和精度成果保证,遥感覆盖面积大、实时性和现实性强、速度快、周期性和准确可靠,同时,遥感还省时、省力、费用低等。
本研究利用盘县1974年、1999年和2005年MSS、ETM+和CBERS遥感图像作为遥感数据源,结合其他地面调查数据和历史资料,通过对遥感图像的地质构造解译,分析盘县煤矿区的地质分布特征,并基于先验知识的人机交互解译手段,提取盘县煤矿区森林资源信息,对30年间盘县煤矿区森林资源进行动态分析,并对盘县煤矿区森林生态环境的动态变化提出科学、有效的评价。
本研究的主要结论有：
①研究煤矿区地质构造解译的ETM+最佳波段组合为7波段、5波段和4波段组合,此三波段组合的最佳波段指数最大,达42.2679；
②在ETM+遥感图像上,煤矿区分布与地质构造分布具有明显的对应关系,一般而言,在遥感图像上,煤矿区主要分布于盘县南部、西部、西北部和北部等地质构造复杂多样地区；主要分布于三叠系下统(T1)、二叠系下统(P2)、石炭系下统(C1)地质时期；主要分布于断层结构不发育,或由西向北的煤矿分布于同一断层的一边；
③基于先验知识的人机交互解译提取森林资源信息,具有良好的科学性和准确性,能满足遥感技术对于图像信息提取的精度要求,本研究对于盘县煤矿区三期图像的信息提取精度分析得知,三期森林资源信息提取结果的分析精度总体上均达到了相当满意的一致程度,总kappa系数均在0.9以上,总精度也在90%以上；
④1974年-1999年,这25年期间,针叶林大幅减少,减少31.442km2；阔叶林也有大量减少,减少约27.032km2；灌林地和经济林有所增加,其中灌林地增幅最大,增量为28.601km2；主要城镇增加41.945km2；
森林资源总体呈下降趋势,针叶林、阔叶林及灌林地等有林地大部分转变为荒山荒地,部分有林地与荒山荒地转变为城镇建设用地。这个时期为盘县地区‘城镇化建设期、环境破坏期、森林资源锐减期；
⑤1999-2005年,5年间盘县煤矿区森林资源状况表现为：针叶林继续大幅度减小,减小约49.143km2,阔叶林面积有所增长(增额为28.369km2),基本恢复到1974年阔叶林水平；经济林、灌林地继续大幅增加,二者总量占整个煤矿区森林资源的32.35%；荒山荒地大幅减小,由1974年的27.42%下降至2005年的14.05%：
森林资源总体上有所回升,有林地面积逐渐增加,荒山荒地治理效果良好,此阶段为盘县煤矿区的“环境保护期、森林资源恢复期”；
⑥盘县煤矿区在30年的建设发展中,由环境破坏期转变到环境保护期,森林资源恢复与保护取得良好效果、成果显著。
Due to the population and resource pressure and ignorer of forest ecological environment protection, forest ecological environment of our country has been destroyed and deteriorated seriously, especially in large drought and semiarid grassland and desert area, which account for 1/3 of our total area. Hence, there are many studies on monitoring of desert forest ecological environment but little on dynamic changes of forest ecological environment in mining area. Recently, with the development of remote sensing technology from qualitative analysis to qualitative and quantitative analysis combination. The dynamic monitoring of forest ecological environment has real time, protracted nature, comprehensiveness and periodicity characteristics, and the survey results must have high precision and accuracy. There are many advantages using remote sensing technology in dynamic monitoring of forest ecological environment, it can provide technique and precision guarantee, and has large coverage, better real time and reality, high speed, time saving, labour saving and cost saving ect.
Based on the MSS,ETM+and CBERS remote sensing images in the year of 1974,1999and 2005,and combined other investigation data and history information, according to geological structure interpretation of remote sensing images, this study analyzed the geological distribution characteristics in mining area, and extracted panxian forest resource information though man-computer interactive interpretation means, the forest resource of 30 years in mining area was dynamically analyzed, and presented scientific and effective evaluation for dynamic changes of mining area forest ecological environment.
The main conclusions of this study are as follows:
(1) The optimal wave band combination of ETM+for geological structure interpretation are 7 wave band,5 wave band and 4 wave band, and the optimal wave band index of combination of these three wave band reach to 42.2679
(2) There are obvious corresponding relation between mine distribution and geological structure distribution in ETM+remote sensing images. Generally, mining areas are mainly in south, west, northwest and north of panxian with complex geological structures, and mainly in geological periods of Triassic lower series(T1),Permian lower series(P2), and Lame-based lower series(C1), and mainly distributed in no growth faultage structure and in one side of the same faultage.
(3) Based on the three-periods information extraction precision analysis of images in panxian mining areas, we obtained that the analysis precision as a whole reach ed to better consistent, and total kappa coefficient was above0.99,and total precision was above 90%.
(4) During the years of 1974 and 1999, there was largely decrease in coniferous forest and broadleaf forest, which reduced 31.442km2 and 27.032km2.But the shrub forest, with the highest increment of 28.601km2,and economic forest were increased. The main towns increased 41.945km2.
The trend of forest resource is decreasing, most coniferous forest, broadleaf forest and shrub forest transformed to barren hills and wasteland, some forest land and barren hills and wasteland turned to town construction land. This period is urbanization construction period, environment destruction period and forest resource fall sharply period of panxian.
(5) During the years of 1999 and 2005, coniferous forest continued decreased sharply about 49.143km2,broadleaf forest increased 28.369km2 and basically reached to the level in 1974, the total economic forest and shrub forest, which accounted for 32.35% of the whole mining area forest resource, continued increased sharply. Barren hills and wasteland decreased sharply from 27.42% in 1974 to 14.05% in 2005.
The forest resource is recovering as a whole, and forest land is increasing, and there was excellent treating effects on barren hills and wasteland. This period is environment protection period, forest resource recovery period.
(6) During the 30 years of mining area construction and development in panxian, from environment destruction period transformed into environment protection period, and there was significant results in forest resource recover and protection.
本研究利用盘县1974年、1999年和2005年MSS、ETM+和CBERS遥感图像作为遥感数据源,结合其他地面调查数据和历史资料,通过对遥感图像的地质构造解译,分析盘县煤矿区的地质分布特征,并基于先验知识的人机交互解译手段,提取盘县煤矿区森林资源信息,对30年间盘县煤矿区森林资源进行动态分析,并对盘县煤矿区森林生态环境的动态变化提出科学、有效的评价。
本研究的主要结论有：
①研究煤矿区地质构造解译的ETM+最佳波段组合为7波段、5波段和4波段组合,此三波段组合的最佳波段指数最大,达42.2679；
②在ETM+遥感图像上,煤矿区分布与地质构造分布具有明显的对应关系,一般而言,在遥感图像上,煤矿区主要分布于盘县南部、西部、西北部和北部等地质构造复杂多样地区；主要分布于三叠系下统(T1)、二叠系下统(P2)、石炭系下统(C1)地质时期；主要分布于断层结构不发育,或由西向北的煤矿分布于同一断层的一边；
③基于先验知识的人机交互解译提取森林资源信息,具有良好的科学性和准确性,能满足遥感技术对于图像信息提取的精度要求,本研究对于盘县煤矿区三期图像的信息提取精度分析得知,三期森林资源信息提取结果的分析精度总体上均达到了相当满意的一致程度,总kappa系数均在0.9以上,总精度也在90%以上；
④1974年-1999年,这25年期间,针叶林大幅减少,减少31.442km2；阔叶林也有大量减少,减少约27.032km2；灌林地和经济林有所增加,其中灌林地增幅最大,增量为28.601km2；主要城镇增加41.945km2；
森林资源总体呈下降趋势,针叶林、阔叶林及灌林地等有林地大部分转变为荒山荒地,部分有林地与荒山荒地转变为城镇建设用地。这个时期为盘县地区‘城镇化建设期、环境破坏期、森林资源锐减期；
⑤1999-2005年,5年间盘县煤矿区森林资源状况表现为：针叶林继续大幅度减小,减小约49.143km2,阔叶林面积有所增长(增额为28.369km2),基本恢复到1974年阔叶林水平；经济林、灌林地继续大幅增加,二者总量占整个煤矿区森林资源的32.35%；荒山荒地大幅减小,由1974年的27.42%下降至2005年的14.05%：
森林资源总体上有所回升,有林地面积逐渐增加,荒山荒地治理效果良好,此阶段为盘县煤矿区的“环境保护期、森林资源恢复期”；
⑥盘县煤矿区在30年的建设发展中,由环境破坏期转变到环境保护期,森林资源恢复与保护取得良好效果、成果显著。
Due to the population and resource pressure and ignorer of forest ecological environment protection, forest ecological environment of our country has been destroyed and deteriorated seriously, especially in large drought and semiarid grassland and desert area, which account for 1/3 of our total area. Hence, there are many studies on monitoring of desert forest ecological environment but little on dynamic changes of forest ecological environment in mining area. Recently, with the development of remote sensing technology from qualitative analysis to qualitative and quantitative analysis combination. The dynamic monitoring of forest ecological environment has real time, protracted nature, comprehensiveness and periodicity characteristics, and the survey results must have high precision and accuracy. There are many advantages using remote sensing technology in dynamic monitoring of forest ecological environment, it can provide technique and precision guarantee, and has large coverage, better real time and reality, high speed, time saving, labour saving and cost saving ect.
Based on the MSS,ETM+and CBERS remote sensing images in the year of 1974,1999and 2005,and combined other investigation data and history information, according to geological structure interpretation of remote sensing images, this study analyzed the geological distribution characteristics in mining area, and extracted panxian forest resource information though man-computer interactive interpretation means, the forest resource of 30 years in mining area was dynamically analyzed, and presented scientific and effective evaluation for dynamic changes of mining area forest ecological environment.
The main conclusions of this study are as follows:
(1) The optimal wave band combination of ETM+for geological structure interpretation are 7 wave band,5 wave band and 4 wave band, and the optimal wave band index of combination of these three wave band reach to 42.2679
(2) There are obvious corresponding relation between mine distribution and geological structure distribution in ETM+remote sensing images. Generally, mining areas are mainly in south, west, northwest and north of panxian with complex geological structures, and mainly in geological periods of Triassic lower series(T1),Permian lower series(P2), and Lame-based lower series(C1), and mainly distributed in no growth faultage structure and in one side of the same faultage.
(3) Based on the three-periods information extraction precision analysis of images in panxian mining areas, we obtained that the analysis precision as a whole reach ed to better consistent, and total kappa coefficient was above0.99,and total precision was above 90%.
(4) During the years of 1974 and 1999, there was largely decrease in coniferous forest and broadleaf forest, which reduced 31.442km2 and 27.032km2.But the shrub forest, with the highest increment of 28.601km2,and economic forest were increased. The main towns increased 41.945km2.
The trend of forest resource is decreasing, most coniferous forest, broadleaf forest and shrub forest transformed to barren hills and wasteland, some forest land and barren hills and wasteland turned to town construction land. This period is urbanization construction period, environment destruction period and forest resource fall sharply period of panxian.
(5) During the years of 1999 and 2005, coniferous forest continued decreased sharply about 49.143km2,broadleaf forest increased 28.369km2 and basically reached to the level in 1974, the total economic forest and shrub forest, which accounted for 32.35% of the whole mining area forest resource, continued increased sharply. Barren hills and wasteland decreased sharply from 27.42% in 1974 to 14.05% in 2005.
The forest resource is recovering as a whole, and forest land is increasing, and there was excellent treating effects on barren hills and wasteland. This period is environment protection period, forest resource recovery period.
(6) During the 30 years of mining area construction and development in panxian, from environment destruction period transformed into environment protection period, and there was significant results in forest resource recover and protection.
引文
[1]邓良基.遥感基础与应用[M].2002.8：355.
[2]朱明德.森林资源动态监测原理与方法[J].南京林业大学学报(自然科学版).1991.(04)：77-82.
[3]孙震,苏尚典,益建芳.遥感综合技术在城市环境监测中的应用[J].影像技术.2006.(02)：92-95.
[4]侯文良,方川.RS和GIS结合下的森林资源动态监测研究——以巴楚县为例[J].新疆农业科学.2006.(S1)：70-71.
[5]J.D.Colby and P. L. Kenting.Landcover Classification UsingLandsat TM Imagery in The Tropical Highlands The Influence of Anisotropic Reflectance [J].Remote Sensing,1998,(8):407～418.
[6]Kohl, M, Sutter, R. Application of aerial photographs in the estimation of standing volume in the Swiss national forest inventory[A]. Forest inventories in Europe with special reference to statistical methods[C]. Swiss Fed Inst For Snow and Landscape Res,1992.176-191.
[7]Bo Ranneby.Designing a new national forest survey for Sweden[J]. Sudia Forestalia Suecica,1987(1771):1-29.
[8]周立.森林虫害监测与防治综合管理系统中3S集成技术实现[J].林业资源管理.1999(1)：59-62.
[9]寇文正.利用遥感技术进行森林资源动态监测验之中国遥感进展[M].中国学术出版社.1992.
[10]刘培均.森林资源动态研究[M].科学出版社,1988.
[11]肖化顺.森林资源监测中林业3S技术的应用现状与展望[J].林业资源管理.2004.(02)：53-58.
[12]姜小光,王长耀,王成.成像光谱数据的光谱信息特点及最佳波段选择—以北京顺义区为例[J].干旱区地理.2000.23(3)：215—220.
[13]章孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州：浙江大学出版社.2003：174-175
[14]杨金红,尹球,顾松山等.城区高光谱遥感数据假彩色波段组合研究[J].南京气象学院学报.2005.28(3)：289-295.
[15]李石华,王金亮,陈姚等.多光谱遥感数据最佳波段选择方法试验研究[J].云南地理环境研究.2005.11(6)：17.
[16]陈述彭,赵英时.遥感地学分析[M].测绘出版社,1990.
[17]朱亮璞.普通高等教育地质矿产类规划教材—遥感地质学[M].地质出版社,1994,3.
[18]王瑞雪.丽江地区遥感影像线环结构特征与场地稳定性分析[J].世界地震工程.2005.21(1).
[19]李永颐,李斌山,陆成.遥感地质学[M].重庆大学出版社.1990.6.
[20]韩文清.遥感地质进展及其应用[J].新疆工学院学报,1998,(04).
[21]游先祥.遥感原理及在资源环境中的应用[M].中国林业出版社.2003.9：232.
[22]赖震刚,王继.利用ERDAS IMAGINE进行影像的几何精校正[J].现代测绘,2003,(02)：38-39.
[23]汤安国,杨昕.ArcGIS地理信息系统空间分析实验教程[M].北京：科学出版社.2006.4.
[24]孙家抦.遥感原理与应用[M].武汉大学出版社.2003.2：163.
[25]王金辉.遥感技术在潍坊地区活动断裂研究中的应用[D].山东：山东科技大学.2005：31-32.
[26]Faust, Nickolas L. Image Enhancement [J]. In Volume 20, Supp lement 5 of Encyclopedia of Computer Science and Technology,New York:Marcel Dekker,Inc.1989.
[27]赵英时.遥感应用分析原理与方法[M].北京：科学出版社,2006：168-169.
[28]李海洋.遥感图象分类方法综述[J].林业科技情报,2008,(01)：4-5.
[29]R ichards J A. Remote Sensing D igital Image A nalysis, Sp r2inger2V erlag [M].Berlin,1994.340.
[30]杨鑫.浅谈遥感图像监督分类与非监督分类[J].四川地质学报,2008,(03)：251-254.
[30]韦玉春,汤国安,杨昕,等.遥感数字图像处理教程[M].北京：科学出版社.2008：226-227
[32]黄宁,刘小军,朱敏慧,张守融.遥感图像分类技术研究[J].华北工学院测试技术学报,2001,(02)：86-92.
[33]罗开华.遥感图像分类后处理3种实现方法比较[J].林业调查规划,2008,(02)：24-26.
[2]朱明德.森林资源动态监测原理与方法[J].南京林业大学学报(自然科学版).1991.(04)：77-82.
[3]孙震,苏尚典,益建芳.遥感综合技术在城市环境监测中的应用[J].影像技术.2006.(02)：92-95.
[4]侯文良,方川.RS和GIS结合下的森林资源动态监测研究——以巴楚县为例[J].新疆农业科学.2006.(S1)：70-71.
[5]J.D.Colby and P. L. Kenting.Landcover Classification UsingLandsat TM Imagery in The Tropical Highlands The Influence of Anisotropic Reflectance [J].Remote Sensing,1998,(8):407～418.
[6]Kohl, M, Sutter, R. Application of aerial photographs in the estimation of standing volume in the Swiss national forest inventory[A]. Forest inventories in Europe with special reference to statistical methods[C]. Swiss Fed Inst For Snow and Landscape Res,1992.176-191.
[7]Bo Ranneby.Designing a new national forest survey for Sweden[J]. Sudia Forestalia Suecica,1987(1771):1-29.
[8]周立.森林虫害监测与防治综合管理系统中3S集成技术实现[J].林业资源管理.1999(1)：59-62.
[9]寇文正.利用遥感技术进行森林资源动态监测验之中国遥感进展[M].中国学术出版社.1992.
[10]刘培均.森林资源动态研究[M].科学出版社,1988.
[11]肖化顺.森林资源监测中林业3S技术的应用现状与展望[J].林业资源管理.2004.(02)：53-58.
[12]姜小光,王长耀,王成.成像光谱数据的光谱信息特点及最佳波段选择—以北京顺义区为例[J].干旱区地理.2000.23(3)：215—220.
[13]章孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州：浙江大学出版社.2003：174-175
[14]杨金红,尹球,顾松山等.城区高光谱遥感数据假彩色波段组合研究[J].南京气象学院学报.2005.28(3)：289-295.
[15]李石华,王金亮,陈姚等.多光谱遥感数据最佳波段选择方法试验研究[J].云南地理环境研究.2005.11(6)：17.
[16]陈述彭,赵英时.遥感地学分析[M].测绘出版社,1990.
[17]朱亮璞.普通高等教育地质矿产类规划教材—遥感地质学[M].地质出版社,1994,3.
[18]王瑞雪.丽江地区遥感影像线环结构特征与场地稳定性分析[J].世界地震工程.2005.21(1).
[19]李永颐,李斌山,陆成.遥感地质学[M].重庆大学出版社.1990.6.
[20]韩文清.遥感地质进展及其应用[J].新疆工学院学报,1998,(04).
[21]游先祥.遥感原理及在资源环境中的应用[M].中国林业出版社.2003.9：232.
[22]赖震刚,王继.利用ERDAS IMAGINE进行影像的几何精校正[J].现代测绘,2003,(02)：38-39.
[23]汤安国,杨昕.ArcGIS地理信息系统空间分析实验教程[M].北京：科学出版社.2006.4.
[24]孙家抦.遥感原理与应用[M].武汉大学出版社.2003.2：163.
[25]王金辉.遥感技术在潍坊地区活动断裂研究中的应用[D].山东：山东科技大学.2005：31-32.
[26]Faust, Nickolas L. Image Enhancement [J]. In Volume 20, Supp lement 5 of Encyclopedia of Computer Science and Technology,New York:Marcel Dekker,Inc.1989.
[27]赵英时.遥感应用分析原理与方法[M].北京：科学出版社,2006：168-169.
[28]李海洋.遥感图象分类方法综述[J].林业科技情报,2008,(01)：4-5.
[29]R ichards J A. Remote Sensing D igital Image A nalysis, Sp r2inger2V erlag [M].Berlin,1994.340.
[30]杨鑫.浅谈遥感图像监督分类与非监督分类[J].四川地质学报,2008,(03)：251-254.
[30]韦玉春,汤国安,杨昕,等.遥感数字图像处理教程[M].北京：科学出版社.2008：226-227
[32]黄宁,刘小军,朱敏慧,张守融.遥感图像分类技术研究[J].华北工学院测试技术学报,2001,(02)：86-92.
[33]罗开华.遥感图像分类后处理3种实现方法比较[J].林业调查规划,2008,(02)：24-26.
© 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 |