基于MODIS时序数据的长江三角洲地区植被覆盖时空变化分析

Spatio-temporal analysis of the vegetation changes based on MODIS time-series data in the Yangtze River Delta region

在快速城市化的长江三角洲地区,植被覆盖状况变化问题已经引起公众的关注并且成为科学研究的热点。植被指数作为监测植被和生态环境变化的有效指标而得到广泛应用。基于2000—2011年的Terra Modis NDVI植被指数数据集,从时间尺度和空间变化上分析长江三角洲地区植被变化特征,应用一元线性回归趋势分析方法和更适合研究非正态分布数据的Mann-Kendall非参数检验方法对长江三角洲地区的植被覆盖状况进行研究。研究结果表明,(1)利用Terra Modis NDVI数据可以很好地从宏观上监测长江三角洲地区植被覆盖的时空变化。(2)植被覆盖的时空变化是自然和人类活动共同作用的结果,特别是在快速城市化的长江三角洲地区,人类活动对植被覆盖的影响更为剧烈,并且植被覆盖变化对人类活动做出积极的响应。12年来,长江三角洲地区植被覆盖总体呈下降趋势,尤其是研究区中的城市及周边地区则以显著下降趋势为主,说明长江三角洲地区随着城市化进程的加速,植被覆盖状况正面临着恶化。(3)一元线性回归趋势分析方法和Mann-Kendall非参数检验方法对植被变化趋势的检测总体上趋于一致,两者可以相互印证,但局部有差异。因此,当进行局部区域植被覆盖变化的深入研究及驱动力分析时,需要较高分辨率的数据作为辅助数据。研究揭示了快速城市化背景下长江三角洲地区植被覆盖变化的规律,从而为长江三角洲地区生态保护和植被恢复提供有利的依据和借鉴。

As one of the most sensitive areas responding to the rapid urbanization, the Yangtze River Delta region has been recognized as a scientific research hotspot and attracted the attention of the public for coupled studies on vegetation coverage. Due to large area coverage and long-time span, vegetation index is widely used as an effective indicator for monitoring changes in vegetation and the ecological environment at regional scales. Using Terra Modis NDVI vegetation index data set from 2000 to 2011, we analyzed the spatio-temporal vegetation variations covering the Yangtze River Delta region in China. As to the methodology for analyzing vegetation dynamics, linear regression and the Mann-Kendall nonparametric test were mostly used to analyze time serial NDVI data. In these methods, vegetation dynamic trends were analyzed through calculating the regression slope and U statistic. The results show that： （1） Terra Modis NDVI dataset are able of monitoring the spatio-temporal variation of the vegetation cover in the entire Yangtze River Delta region; （2） The spatio-temporal variation of the vegetation coverage is a result of the natural and human activities. Especially in the rapid urbanization in the Yangtze River Delta region, vegetation coverage makes a positive response to intensive human activities. In the past 12 years, the vegetation cover of the Yangtze River Delta region is facing overall degradation trend and even deterioration, especially in the city and sub-city with the accelerated urbanization; （3） A linear regression trend analysis method for the detection of vegetation change trends was found to be in overall acceptable agreement with Mann-Kendall nonparametric test method. Therefore, it needs to analyze the driving force of the vegetation coverage change at local area, and higher resolution data would be required as an auxiliary. This study reveals the trends of vegetation coverage change in the Yangtze River Delta region in the context of rapid urbanization, which provides a favorable reference for the ecological protection and vegetation recovery.