| 应用高分辨率卫星数据估算阔叶红松林乔木多样性 |
| |
| 引用本文： | 解潍嘉,黄侃,李瑞平,孙浩,扈晶晶,黄华国.应用高分辨率卫星数据估算阔叶红松林乔木多样性[J].北京林业大学学报,2015,37(3):20-26. |
| |
| 作者姓名： | 解潍嘉 黄侃 李瑞平 孙浩 扈晶晶 黄华国 |
| |
| 作者单位： | 北京林业大学林学院，省部共建森林培育与保护教育部重点实验室;北京林业大学林学院，省部共建森林培育与保护教育部重点实验室;北京林业大学林学院，省部共建森林培育与保护教育部重点实验室;北京林业大学林学院，省部共建森林培育与保护教育部重点实验室;北京林业大学林学院，省部共建森林培育与保护教育部重点实验室;北京林业大学林学院，省部共建森林培育与保护教育部重点实验室 |
| |
| 基金项目： | “十二五”国家科技支撑计划项目(2012BAC01B03) |
| |
| 摘 要： | 应用遥感技术估算生物多样性是森林生态学的研究热点和难点。为探索高分辨率卫星数据在估算森林植被多样性上的应用潜力,以吉林蛟河市阔叶红松林大样地数据为支撑,提取了红松树冠的空间分布格局,并估算了乔木物种多样性。首先,基于冬季GeoEye-1影像提取红松树冠分布图,与样地每木调查数据吻合较好。然后,使用景观生态学和地统计学指数,分析红松种群空间分布规律,发现了最大自相关尺度30~40 m。最后,选择拥有红边波段的秋季RapidEye卫星图像,建立遥感影像5 m光谱值和30 m窗口大小纹理参数与生物多样性指数的多元线性回归模型。结果表明,红边信息并未显著改善多光谱数据估算植被生物多样性的能力,但所获大区域制图结果与林分龄级吻合较好。
|
| 关 键 词： | 遥感 森林生态系统 生物多样性 种群分布格局 |
| 收稿时间： | 2014-09-15 |
Applying high-resolution satellite images to estimate tree diversity of mixed broadleaf-Korean pine forest |
| |
| XIE Wei-jia
,
HUANG Kan
,
LI Rui-ping
,
SUN Hao
,
HU Jing-jing
,
HUANG Hua-guo.Applying high-resolution satellite images to estimate tree diversity of mixed broadleaf-Korean pine forest[J].Journal of Beijing Forestry University,2015,37(3):20-26. |
| |
| Authors: | XIE Wei-jia HUANG Kan LI Rui-ping SUN Hao HU Jing-jing HUANG Hua-guo |
| |
| Institution: | Beijing Forestry University, Beijing, 100083, P. R. China. |
| |
| Abstract: | The application of remote sensing technology to estimate forest biodiversity is a research hotspot and challenge in forest ecology. To explore the potential of high-resolution satellite data in estimating forest vegetation diversity, we extracted the spatial pattern of Korean pine ( Pinus koraiensis) canopies, and estimated the arbor species diversity, supported by field large plot data in Jiaohe, Jilin Province. First, we extracted the Korean pine canopy distribution based on a GeoEye-1 winter image, which agreed well with individual tree position measured in situ. Then, we analyzed the spatial population distribution of Korean pine using landscape ecology and geostatistics index, which was found to have a maximum autocorrelation scale of 30--40 m. Finally, we used a RapidEye satellite image with a special red edge band in autumn to establish multiple linear regression model with plot diversity data. The model linked remote sensing image spectral values ( 5 m ) and texture parameters ( 30 m window size ) with species diversity. Results showed that red edge information did not significantly improve the ability of multi-spectral data in estimating vegetation biodiversity, but regional mapping results matched well with the stand age class. |
| |
| Keywords: | remote sensing forest ecosystems biodiversity population distribution patterns |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《北京林业大学学报》浏览原始摘要信息 |
| 点击此处可从《北京林业大学学报》下载免费的PDF全文 |
|
