Knowledge Resource Center for Ecological Environment in Arid Area
2000-2011年东北亚地区植被覆盖度变化研究 | |
其他题名 | Change of Fractional Vegetation Cover in Northeast Asia from 2000 to 2011 |
韩佶兴 | |
出版年 | 2012 |
学位类型 | 硕士 |
导师 | 王宗明 |
学位授予单位 | 中国科学院大学 |
中文摘要 | 植被覆盖度(Fractional Vegetation Cover, FVC) 是描述生态系统的重要基础数据,也是全球变化检测、水文、土壤侵蚀等研究中的重要参数指标。东北亚地区作为我国重要的生态屏障及全球变化的敏感区研究其植被覆盖度变化有极其重要的意义。本文以东北亚地区为研究区,结合野外实测数据,Landsat-TM/ETM+数据以及2000-2011年MODIS-NDVI数据,综合运用遥感技术、GIS空间分析以及统计分析方法,讨论了不同尺度下植被覆盖度算法的选取,分析了过去12a植被覆盖度时空变化的特征,主要结论如下: \n(1)基于Landsat-TM/ETM+数据的像元二分法估算精度小于线性混合分解法,像元二分法对植被覆盖度高值估算结果精度较低。基于MODIS-NDVI数据的植被覆盖度计算结果比较也同样可以看到NDVI过饱和对估算结果的影响,相比之下,采用累计NDVI频率为5%和95%所对应的NDVI值作为参数估算东北亚地区植被覆盖度精度较高。 \n(2)东北亚地区植被覆盖度总体呈现山区高平原低的特点,具有从南向北、从东向西逐渐减小的变化特征。过去12a大部分地区呈基本稳定趋势,植被覆盖度呈退化趋势的地区主要分布在中西伯利亚高原、贝加尔湖东部雅布洛诺夫山脉以及内蒙古中东部。植被覆盖度呈增加趋势的地区主要分布在中国的东北平原区、蒙古高原东部以及阿尔泰山脉东麓。 \n(3)日本、朝鲜、韩国的平均年最大植被覆盖度最大,俄罗斯、中国次之,蒙古最小。过去12a,中国、蒙古平均年最大植被覆盖度具有轻微上升趋势,俄罗斯、日本、朝鲜均呈轻微下降趋势,韩国年最大植被覆盖度基本保持稳定,各国植被覆盖度变化趋势都均不显著。 \n(4)森林植被的年最大植被覆盖度在东北亚地区各植被类型中最大,苔原、草原最小;草原地区年最大植被覆盖度波动性最大,耕地次之。过去12a森林地区的植被覆盖度呈显著降低趋势,植被覆盖度呈上升趋势比例最大的植被类型为耕地,主要分布在中国东北平原区。东北亚地区各植被覆盖类型植被活动都从4月份开始明显增强,除耕地是在6月份外,其他类型植被的植被覆盖度增长速度都在5月份达到最大,耕地和草原年植被覆盖度达到峰值的时间比其他植被晚一个月,各植被类型都在10月进入植被枯黄期,生长季结束。 \n(5)东北亚东南样带植被覆盖度分布沿纬度变化差异明显。除耕地外,各典型植被区植被覆盖度从南向北呈现明显的先迅速增加后缓慢降低的分布规律。中高纬度带植被覆盖度增加速度最大值出现时间比低纬度带和高纬度带提前一个月,其出现时间早于低纬度荒漠草原区,主要受水分梯度变化影响,早于高纬度苔原区主要受温度梯度变化影响。 |
英文摘要 | Fractional Vegetation Cover (FVC) is a underlying data for describing ecosystem and an important parameter for researches on the global change detection, hydrology, soil erosion etc. The Northeast Asia, as a critical ecological shelter of China and the sensitive area of global change, analyzing its change characteristics of fractional vegetation cover has great significance. Combined with field observation data, Landsat TM/ETM+ data and MODIS-NDVI data, technology of Remote Sensing, GIS spatial analysis and statistical analysis method, this paper discussed the algorithm seletion of fractional vegetation cover extraction at different scales, estimated fractional vegetation cover from 2000 to 2011 in the Northeast Asia, and analyzed its spatial-temporal variations characteristics. Based on these researches, some basic conclusions are drawn as follows. \n(1) The Spectral Mixture Analysis (SMA) model and the NDVI-based SMA model were compared based on Landsat-TM/ETM+ data. The SMA model performed better for estimating fractional vegetation cover, especially in high NDVI value regions. The influence of supersaturated NDVI value on FVC estimation based on MODIS-NDVI data also can be seen. The NDVI-based SMA, taking accumulated frequency of NDVI corresponding to 5% and 95% as parameters, performed well for estimating FVC in the entire Northeast Asia. \n(2) In the Northeast Asia, the FVC is higher in mountain regions while lower in plain totally, which decreases from south to north and from east to west. In the past 12 years, the FVC in the most areas was basically steady. The regions where FVC declined were mainly located in the Central Siberian Plateau, Yablonovyy khrebet in the eastern of the Lake Baikal, and mid-eastern of the Inner Mongolia. The regions where FVC increased were mainly located in the Northeast Plain of China, the eastern part of the Mongolia Plateau and the eastern side of Altai Mountains. \n(3) The highest average annual maximum FVC was located in Japan, North Korea and South Korea, followed by Russia and China, the lowest is in Mongolia. In the past 12 years, the average annual maximum FVC increased slowly in China and Mongoliawhile decreased in Russia, Japan and North Korea, and that was basically steady in South Korea. Totally, the change of FVC in all countries was not significant. \n(4) The annual maximum FVC of forest is the highest in the study area, and that of tundra and grassland is the lowest. The annual maximum FVC of grassland fluctuates most severely, followed by arable land. In the past 12 years, the FVC of forest vegetation declined significantly, while that of arable land which was mainly located in the Northeast Plain of China increased fastest. The vegetation activity for all cover types intensed significantly from April in the Northeast Asia, the growth rate of FVC for all vegetation types reached the fastest in May except arable land which reached the fastest in June. The time of maximum annual FVC for arable land and grassland lagged one month than other vegetations. All vegetation types withered in October, which indicates that growing season ends. \n(5) The FVC in the North-South Transact of the Northeast Asia changes obviously along the latitude. Except for arable land, the FVC of all vegetation types increases rapidly first, then decreases slowly from south to north. The occurance of the fastest growth rate of FVC in the mid-high latitude region is one month before than that in the low and high latitude regions. |
中文关键词 | 植被覆盖度 ; 东北亚地区 ; 线型混合分解法 ; 像元二分法 |
英文关键词 | Fractional Vegetation Cover (FVC) Northeast Asia Normalized Difference Vegetation Index (NDVI) Spectral mixture analysis NDVI-based SMA model |
语种 | 中文 |
国家 | 中国 |
来源学科分类 | 地图学与地理信息系统 |
来源机构 | 中国科学院东北地理与农业生态研究所 |
资源类型 | 学位论文 |
条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/287066 |
推荐引用方式 GB/T 7714 | 韩佶兴. 2000-2011年东北亚地区植被覆盖度变化研究[D]. 中国科学院大学,2012. |
条目包含的文件 | 条目无相关文件。 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[韩佶兴]的文章 |
百度学术 |
百度学术中相似的文章 |
[韩佶兴]的文章 |
必应学术 |
必应学术中相似的文章 |
[韩佶兴]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。