| 基于计算机模拟的植物返青期遥感监测方法比较研究 |
| |
| 引用本文： | 武永峰,何春阳,马瑛,李京,张业利.基于计算机模拟的植物返青期遥感监测方法比较研究[J].地球科学进展,2005,20(7):724-731. |
| |
| 作者姓名： | 武永峰 何春阳 马瑛 李京 张业利 |
| |
| 作者单位： | 北京师范大学资源学院,北京,100875;北京石油化工学院人文社科学院,北京,102617 |
| |
| 基金项目： | 北京师范大学校科研和教改项目 |
| |
| 摘 要： | 植物返青期变化与气象和气候因素密切相关，已成为研究全球变化对陆地生态系统影响和反馈机理的一个重要参数。利用遥感手段及时准确地监测区域及全球尺度植物返青期变化，是当前地学、生态学研究的一个前沿课题。但是目前国内外对植物返青期遥感监测方法缺乏有效比较，还没有形成统一认识。因此，在综述植物返青期传统测量方法和遥感监测方法最新研究进展的基础上，利用计算机模拟手段，对目前常用的4种遥感监测方法（滑动平均方法、NDVI比率阈值方法、最大变化斜率方法及Logistic函数拟合方法）进行了比较研究。模拟结果表明：4种方法的返青期计算值最大相差20天，突出反映了它们在构建原理方面的差别，即滑动平均方法将NDVI突增点对应的时期作为植物返青期；NDVI比率阈值方法将最大上升斜率点对应的时期作为植物返青期；最大变化斜率方法将NDVI数据斜率角变化最大的点对应的时期作为植物返青期；Logistic函数拟合方法则将拟合曲线上曲率变化最大的点对应的时期作为植物返青期；同时，4种方法所得返青期均满足一个共同的特点，即返青期之后NDVI曲线应保持最大持续增长。该研究也表明计算机模拟技术在帮助认识和理解植物返青期遥感监测方法中的巨大潜力。
|
| 关 键 词： | 植物物候 返青期 遥感监测方法 NDVI |
| 文章编号： | 1001-8166(2005)07-0724-08 |
| 收稿时间： | 2004-09-06 |
| 修稿时间： | 2004年9月6日 |
THE COMPARISON OF THE CURRENT REMOTE SENSING-BASED VEGETATION GREENUP DETECTION METHODS WITH THE COMPUTER SIMULATION |
| |
| WU Yong-feng,HE Chun-yang,Ma Ying,LI Jing,Zhang Ye-li.THE COMPARISON OF THE CURRENT REMOTE SENSING-BASED VEGETATION GREENUP DETECTION METHODS WITH THE COMPUTER SIMULATION[J].Advance in Earth Sciences,2005,20(7):724-731. |
| |
| Authors: | WU Yong-feng HE Chun-yang Ma Ying LI Jing Zhang Ye-li |
| |
| Institution: | 1.College of Resources Science and Technology, Beijing Normal University, Beijing 100875, China;2.School of Humanities and Social Sciences, Beijing Institute of Petrochemical Technology, Beijing 102617,China |
| |
| Abstract: | Vegetation greenup is closely related to seasonal dynamics of the lower atmosphere and is therefore an important variable in influences and feedback mechanisms of global climate variation to terrestrial ecosystem. Detecting vegetation greenup using remotely sensed data at regional or global scales has become an advanced topic in geography and ecology. However, current remote sensing based vegetation greenup detection methods are short of valid comparison and do not have a consistent understanding. So, This paper utilizes computer simulation technique to study four remote sensing based detecting methods (Moving average method, NDVI ratio threshold method, the greatest change slope method and Logistic function fitting method) after summarizing advanced developments of traditional measuring methods and remote sensing based detecting methods of vegetation greenup. Simulation results indicate that:①simulated results of greenup from the four methods make a difference of 20 days, which strongly reflect all method's differences in building principle. Moving average method defines vegetation greenup as the period beginning at the point where the smoothed time series data suddenly increases. NDVI Ratio Threshold method defines vegetation greenup as the period beginning at the point of the greatest slope. The greatest change slope method defines vegetation greenup as the period beginning at the point of the maximum change in the NDVI slope angle. Logistic Function Fitting method defines vegetation greenup as the period beginning at the point where the rate of change of curvature exhibits a local maximum in the fitting curve. In addition, the four methods have a same character that the NDVI time series data should maintains the greatest sustained increase after vegetation greenup begins. ②Computer simulating technique is helpful to understand the great potential of Remote sensing based vegetation greenup detection methods in the future. ③Validation of the results is very important in the course of using remote sensing based detecting methods, which this paper is short of because of lacking in practical observation data. |
| |
| Keywords: | NDVI |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《地球科学进展》浏览原始摘要信息 |
| 点击此处可从《地球科学进展》下载免费的PDF全文 |
|
