近30年西藏尼玛地区草地遥感动态监测研究
发布时间：2019-03-12 17:56
【摘要】：青藏高原地处中低纬度带,地域面积辽阔,地势高低起伏较大,平均海拔4000多米,被誉为“世界屋脊”和地球的“第三极”,主要以其奇异的地貌,独特的自然风光、丰富的自然资源及其对周边地区乃至全球气候与环境的影响,一直为国内外各界所瞩目。草地资源是发展第一产业(畜牧业)的物质基础,草地资源对于以牧业为主的尼玛地区的经济社会发展具有举足轻重的作用。目前,西藏尼玛地区天然草地面临着诸多问题,主要以草地面积减少、质量下降、草畜矛盾尖锐等问题,及时准确地了解西藏尼玛地区天然草地资源的时空分布状况至关重要,而现代的检测技术对于天然草场的时空变化监测具有快速、及时、方便等的优势,以遥感监测为主要监测手段,其具有监测面积大尺度、时间小尺度的优势,能够实时把握草地资源的变化状况,为草地的动态监测与管理提供了有力的技术支持。本研究采用了路线调查与遥感解译相结合的方法,结合前人资料,对本区草地生态环境现状、草地退化原因作了详细的调查和分析。本次研究区地理位置特殊,生态环境基础十分脆弱,由于长期以来对该区域的草地资源缺乏科学的利用与管理,过牧、滥牧现象已经十分的突出,已经导致研究区的草地严重退化和草地生态失调,甚至部分区域出现大面积荒漠化和沙漠化等现象,保护研究区草地的生态环境,开展其生态环境建设,已经刻不容缓。本研究选取西藏尼玛地区为对象,选取Landsat不同时间段的影像数据,先对原始数据进行预处理,然后通过分析草地资源不同季节(主要选取夏季遥感影像,原因：夏季草场生长最为茂盛易于解译,可以将不同的用地区分开来,尤其是草地与荒漠化土地分开,而秋季枯黄的草地与荒漠化的土地是无法区分开的。)的光谱特征,采用NDVI、监督分类等方法提取草地资源相关的信息,并通过获取的相关信息,对草地进行等级划分,监测、分析了研究区1988年、1999年、2009年和2015年草地的年际时空动态变化。(1)遥感数据的预处理Landsat遥感原始数据预处理主要有以下几个步骤,首先是辐射校正,消除传感器自身运行所产生的误差,并将DN转换为辐射亮度值,然后采用FLAASH模型进行大气纠正,消除大气和光照等因素对地物反射的影响。然后采用多项式纠正模型重新定位,并进行三次卷积内插法重新采样完成几何精纠正。最后在ENVI软件中进行投影转换,保证所有研究数据为同一投影。结果显示：辐射纠正和FLAASH模型大气纠正较好的消除了大气等对遥感数据的干扰,保证了NDVI指数的精确度。(2)草地资源信息提取将预处理后的Landsat数据进行选择训练样区,分析其可分离度及分离精度,进行监督分类,并通过逻辑运算求得草地的NDVI指数,结合NDⅥ指数大小及其分布,综合考虑植被覆盖度概况,将草地分为4个等级,保证等级内差异最小,等级外差异最大,尽最大努力的提取同一草地在不同时空分布的等级信息,为后期的动态监测与保护提供数据基础。(3)草地资源动态监测对不同年份的草地面积进行统计和不同等级草地面积及地理位置变化统计。从草地资源NDVI值角度分析：最近30年来,研究区草地总面积呈递减趋势,然而,不同等级的各种草地随着时间的变化呈现出变化不一的趋势,低等级草地面积表现为先减后增,中低等级草地和中等级草地面积都是表现出先增后减。(4)草地变化原因分析和草地发展建议结合遥感数据的解译及数学运算分析和实地调查、验证获取的数据信息分析得出草地变化的主要原因有严重的超载过牧、研究区生态失衡、鼠害严重、毒草猖獗和一系列不合理的人类活动等,草地优化发展的对策主要有加强对沙化退化草场综合治理、防止超载过牧、加强自然保护区建设、草原水利建设和草原可持续发展体系和保护制度建设等。本研究以西藏尼玛地区为对象,采用多时相的Landsat影像数据,充分发挥草地遥感的优势,分析了近30年西藏尼玛地区草地时空特征。本研究可以为尼玛地区的草地资源信息提取与时空动态变化特征分析提供了切实可行的技术方法,对政府及草地资源使用者及时准确地了解草地资源的时空分布现状及变化趋势,对促进地方畜牧业科学生产和保护生态环境的平衡与改善具有极其重要的参考价值及指导意义。
[Abstract]:The Qinghai-Tibet Plateau is located in the middle and low-latitude zone with a vast area, with a large and high terrain, with an average altitude of more than 4000 meters, and is known as the "third pole" of the "the roof of the world" and the earth, and is mainly characterized by its strange landforms and unique natural scenery. The rich natural resources and its impact on the surrounding area and even the global climate and environment have been attracting the attention of various circles at home and abroad. Grassland resources are the material base of developing the first industry (animal husbandry). At present, there are many problems in the natural grassland in the Nima area in Tibet, which is mainly based on the problems such as the decrease of the area of the grassland, the decline of the quality, the sharp contradiction of the grass and the livestock, and the timely and accurate understanding of the spatial and temporal distribution of the natural grassland resources in the Nima area of Tibet. The modern detection technology has the advantages of fast, timely and convenient monitoring on the space-time change of the natural grassland, and the remote sensing monitoring is the main monitoring means, which has the advantages of large scale and small time scale of the monitoring area, can control the change condition of the grassland resources in real time, And provides powerful technical support for the dynamic monitoring and management of the grassland. In this study, the method of combination of route survey and remote sensing interpretation is adopted, and the present situation of grassland ecological environment and the cause of grassland degradation are investigated and analyzed in detail in the light of the previous data. The geographical location of the research area is special, the ecological environment is very fragile, because of the lack of scientific utilization and management of the grassland resources in the region for a long time, the phenomenon of overgrazing and overgrazing has been very prominent, which has led to the serious degradation of the grassland in the research area and the ecological imbalance of the grassland, It is urgent to protect the ecological environment of the grassland in the research area and to carry out its ecological environment construction even in some areas, such as the large area of desertification and desertification. In this study, the image data of the Nama area in Tibet is selected, and the image data of the Landsat time period is selected, the original data is pre-processed firstly, and then the summer remote sensing image is mainly selected by analyzing the different seasons of the grassland resources (main selection of the summer remote sensing image). It is possible to separate different areas of use, in particular the grass and the desertification land, and the grass and desertification land in the fall of the autumn cannot be distinguished. ) The spatial and temporal dynamics of the grassland in 1988,1999,2009 and 2015 were analyzed by means of NDVI and supervised classification. (1) preprocessing Landsat remote sensing raw data of remote sensing data mainly comprises the following steps: firstly, the radiation correction is carried out, the error generated by the operation of the sensor is eliminated, the DN is converted into a radiation brightness value, the FLASH model is adopted for atmospheric correction, And the influence of factors such as atmosphere and light on the reflection of the figure is eliminated. Then the polynomial correction model is used for re-positioning, and the three-time convolution interpolation method is used for re-sampling to complete the geometric rectification. And finally, the projection conversion is carried out in the ENVI software to ensure that all the research data is the same projection. The results show that the radiation correction and the air correction of the FLASH model eliminate the interference of the atmosphere and the like on the remote sensing data, and the accuracy of the NDVI index is guaranteed. (2) the grassland resource information extracts the pre-processed Landsat data to select a training sample area, analyzes the separable degree and the separation precision, carries out supervision classification, and obtains the NDVI index of the grassland through a logical operation, combines the size of the ND VI index and the distribution thereof, and comprehensively considers the vegetation coverage profile, The grassland is divided into 4 grades, the difference in the guarantee level is the smallest, the difference between the grades is the largest, and the level information of the same grassland at different time and time is extracted to the best efforts, and the data base is provided for later dynamic monitoring and protection. (3) The dynamic monitoring of the grassland resources can be used for statistics of the area of the grassland in different years and the change of the area of the grassland and the geographical position. From the angle of NDVI of the grassland resources, the total area of the grassland in the research area is decreasing in the last 30 years. However, the different grades of the grassland show different trends over time. The area of medium-low-grade grassland and medium-grade grassland is the first increase and decrease. (4) The cause of the change of the grassland and the suggestion of the development of the grassland combine the interpretation of the remote sensing data and the analysis of the mathematical operation and the field investigation, and the analysis of the obtained data information shows that the main causes of the change of the grassland have serious overload and overgrazing, the ecological imbalance of the research area and the serious damage of the mouse, The countermeasures of grassland optimization and development mainly include strengthening the comprehensive treatment of the degraded grassland, preventing overgrazing, strengthening the construction of the nature reserve, the construction of the grassland water conservancy and the sustainable development system of the grassland and the construction of the protection system. In this study, the spatial and temporal characteristics of the grassland in the Nima area of Tibet were analyzed by using Landsat image data of a long time and using Landsat image data of the time. The study can provide a feasible technical method for the information extraction and spatial-temporal dynamic analysis of the grassland resources in the Nima area. The present situation and the changing trend of the spatial and temporal distribution of the grassland resources are known to the government and the users of the grassland resources in a timely and accurate manner. It is of great reference value and guiding significance to the promotion of the balance and improvement of the scientific production of local animal husbandry and the protection of the ecological environment.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：S812.5
本文编号：2439022
[Abstract]:The Qinghai-Tibet Plateau is located in the middle and low-latitude zone with a vast area, with a large and high terrain, with an average altitude of more than 4000 meters, and is known as the "third pole" of the "the roof of the world" and the earth, and is mainly characterized by its strange landforms and unique natural scenery. The rich natural resources and its impact on the surrounding area and even the global climate and environment have been attracting the attention of various circles at home and abroad. Grassland resources are the material base of developing the first industry (animal husbandry). At present, there are many problems in the natural grassland in the Nima area in Tibet, which is mainly based on the problems such as the decrease of the area of the grassland, the decline of the quality, the sharp contradiction of the grass and the livestock, and the timely and accurate understanding of the spatial and temporal distribution of the natural grassland resources in the Nima area of Tibet. The modern detection technology has the advantages of fast, timely and convenient monitoring on the space-time change of the natural grassland, and the remote sensing monitoring is the main monitoring means, which has the advantages of large scale and small time scale of the monitoring area, can control the change condition of the grassland resources in real time, And provides powerful technical support for the dynamic monitoring and management of the grassland. In this study, the method of combination of route survey and remote sensing interpretation is adopted, and the present situation of grassland ecological environment and the cause of grassland degradation are investigated and analyzed in detail in the light of the previous data. The geographical location of the research area is special, the ecological environment is very fragile, because of the lack of scientific utilization and management of the grassland resources in the region for a long time, the phenomenon of overgrazing and overgrazing has been very prominent, which has led to the serious degradation of the grassland in the research area and the ecological imbalance of the grassland, It is urgent to protect the ecological environment of the grassland in the research area and to carry out its ecological environment construction even in some areas, such as the large area of desertification and desertification. In this study, the image data of the Nama area in Tibet is selected, and the image data of the Landsat time period is selected, the original data is pre-processed firstly, and then the summer remote sensing image is mainly selected by analyzing the different seasons of the grassland resources (main selection of the summer remote sensing image). It is possible to separate different areas of use, in particular the grass and the desertification land, and the grass and desertification land in the fall of the autumn cannot be distinguished. ) The spatial and temporal dynamics of the grassland in 1988,1999,2009 and 2015 were analyzed by means of NDVI and supervised classification. (1) preprocessing Landsat remote sensing raw data of remote sensing data mainly comprises the following steps: firstly, the radiation correction is carried out, the error generated by the operation of the sensor is eliminated, the DN is converted into a radiation brightness value, the FLASH model is adopted for atmospheric correction, And the influence of factors such as atmosphere and light on the reflection of the figure is eliminated. Then the polynomial correction model is used for re-positioning, and the three-time convolution interpolation method is used for re-sampling to complete the geometric rectification. And finally, the projection conversion is carried out in the ENVI software to ensure that all the research data is the same projection. The results show that the radiation correction and the air correction of the FLASH model eliminate the interference of the atmosphere and the like on the remote sensing data, and the accuracy of the NDVI index is guaranteed. (2) the grassland resource information extracts the pre-processed Landsat data to select a training sample area, analyzes the separable degree and the separation precision, carries out supervision classification, and obtains the NDVI index of the grassland through a logical operation, combines the size of the ND VI index and the distribution thereof, and comprehensively considers the vegetation coverage profile, The grassland is divided into 4 grades, the difference in the guarantee level is the smallest, the difference between the grades is the largest, and the level information of the same grassland at different time and time is extracted to the best efforts, and the data base is provided for later dynamic monitoring and protection. (3) The dynamic monitoring of the grassland resources can be used for statistics of the area of the grassland in different years and the change of the area of the grassland and the geographical position. From the angle of NDVI of the grassland resources, the total area of the grassland in the research area is decreasing in the last 30 years. However, the different grades of the grassland show different trends over time. The area of medium-low-grade grassland and medium-grade grassland is the first increase and decrease. (4) The cause of the change of the grassland and the suggestion of the development of the grassland combine the interpretation of the remote sensing data and the analysis of the mathematical operation and the field investigation, and the analysis of the obtained data information shows that the main causes of the change of the grassland have serious overload and overgrazing, the ecological imbalance of the research area and the serious damage of the mouse, The countermeasures of grassland optimization and development mainly include strengthening the comprehensive treatment of the degraded grassland, preventing overgrazing, strengthening the construction of the nature reserve, the construction of the grassland water conservancy and the sustainable development system of the grassland and the construction of the protection system. In this study, the spatial and temporal characteristics of the grassland in the Nima area of Tibet were analyzed by using Landsat image data of a long time and using Landsat image data of the time. The study can provide a feasible technical method for the information extraction and spatial-temporal dynamic analysis of the grassland resources in the Nima area. The present situation and the changing trend of the spatial and temporal distribution of the grassland resources are known to the government and the users of the grassland resources in a timely and accurate manner. It is of great reference value and guiding significance to the promotion of the balance and improvement of the scientific production of local animal husbandry and the protection of the ecological environment.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：S812.5
【参考文献】
相关期刊论文 前1条
1 王承军,胡新博,顾详,候继东;利用卫星遥感技术改进草地估产方法[J];草食家畜;1998年01期
相关硕士学位论文 前1条
1 李新一;基于3S技术的内蒙古镶黄旗草原生产力研究[D];中国农业大学;2005年
，本文编号：2439022
本文链接：https://www.wllwen.com/yixuelunwen/dongwuyixue/2439022.html
