地理学报 ›› 2015, Vol. 70 ›› Issue (9): 1491-1502.doi: 10.11821/dlxb201509011

• • 上一篇    下一篇

洞庭湖流域植被动态变化的小波多分辨率分析

龙岳红1,2(), 秦建新1(), 贺新光1, 杨准1   

  1. 1. 湖南师范大学资源与环境科学学院,长沙 410081
    2. 湖南城市学院市政与测绘工程学院,益阳 413000
  • 收稿日期:2015-05-05 修回日期:2015-06-23 出版日期:2015-09-20 发布日期:2015-10-29
  • 作者简介:

    作者简介:龙岳红(1981-), 女, 湖南岳阳人, 博士研究生, 讲师, 主要从事资源环境遥感与GIS研究。E-mail: 913290941@qq.com

  • 基金资助:
    湖南省十二五重点学科地理学和湖南省教育厅科研项目(14A097)

Wavelet multi-resolution analysis of vegetation dynamic change in Dongting Lake Basin

Yuehong LONG1,2(), Jianxin QIN1(), Xinguang HE1, Zhun YANG1   

  1. 1. College of Resources and Environmental Science, Hunan Normal University, Changsha 410081, China
    2. College of Municipal and Surveying and Mapping Engineering, Hunan City University, Yiyang 413000, Hunan, China
  • Received:2015-05-05 Revised:2015-06-23 Online:2015-09-20 Published:2015-10-29
  • Supported by:
    Hunan Province 12th Five-Year Plan Key Subject of Geography The Scientific Research Project of Education Department of Hunan Province, No.14A097

摘要:

将离散小波多分辨率分析(MRA)应用于归一化植被指数(NDVI)时间序列研究,分解NDVI原数据序列成不同时间尺度的子序列,从而进行植被动态变化分析。针对洞庭湖流域的NDVI时间序列进行多尺度分解,挖掘这些数据中潜在的植被季节性和年际变化,对其进行评估,并结合土地覆盖变化与降水趋势变化分析引起该变化的可能原因。结果表明:小波多分辨率分析能提取洞庭湖流域植被动态的相关信息,如NDVI的年际成分均值、最低值、植被年内变化的振幅、NDVI最大值出现的月份和土地覆盖变化的趋势及幅度,这些信息有效刻画了流域植被动态变化特征。此外,将土地覆盖变化分析结果与降水数据相结合进行分析,发现流域植被覆盖的变化与降水变化有明显的关联性,其中西部地区和西北部地区植被动态变化对降水响应最为明显。但是有些地区降水没有明显减少趋势,而植被覆盖却存在减少趋势,则可能与该地地势较高、城镇化建设等其他因素相关。

关键词: 洞庭湖流域, MODIS NDVI, 离散小波变换, 多分辨率分析, 植被动态变化

Abstract:

The vegetation dynamic change can be analyzed by applying the discrete wavelet multi-resolution analysis (MRA) to decompose the normalized difference vegetation index (NDVI) time series into subseries at different time scales. In this thesis, the NDVI time series over Dongting Lake Basin was decomposed under different time scales by the wavelet transform. Then the latent seasonal and inter-annual variability of vegetation was mined and assessed. Finally, the possible causes of the changes were analyzed in combination with the land cover change analysis result and the rainfall data. The results show that by using the wavelet multi-resolution analysis, the information related to vegetation dynamics in Dongting Lake Basin can be extracted, which includes the mean of the inter-annual component of NDVI, the minimum level of NDVI, the amplitude of the annual phonological cycle, the timing of the maximum NDVI, the trend in the data series and the magnitude of the land cover change. The information effectively depicts the basin's vegetation dynamics characteristics. Furthermore, with the combination of land cover change analysis result and precipitation data, an obvious correlation was found between the change of vegetation cover and the precipitation change in the basin, that is, the vegetation dynamic changes response to precipitation is the most obvious in the western and northwestern parts of the basin. Although there is no obvious decrease in precipitation in some areas, the vegetation cover has a decreasing trend, which may be related to other factors, such as higher terrain and urbanization.

Key words: Dongting Lake Basin, MODIS NDVI, discrete wavelet transform, multi-resolution analysis, vegetation dynamic change