【摘要】 利用SPOT VGT传感器的NDVI时序数据,采用时间序列谐波分析算法(HANTS)对ND-VI数据进行去云预处理,基于趋势分析、Hurst指数分析等方法,研究了1999~2008年新疆植被覆盖的时空变化,分析了Hurst指数的空间分布规律及其在不同土地覆盖类型下的差异。结果表明:近10 a来,新疆植被总体呈现不显著的增加趋势且具有明显的空间差异性。R/S分析表明,新疆大部分地区植被变化将保持现在的趋势,局部地区具有反持续性。各土地覆盖类型Hurst指数均在0.63以上,沙漠、戈壁、裸地的Hurst指数明显高于其它土地覆盖类型,受人类活动影响较大的土地覆盖类型其Hurst指数较低,可持续性低于自然或半自然状态下的植被覆盖类型。更多还原

【Abstract】 Vegetation change is a hot spot of global change research.Using the SPOT VGT NDVI time series data sets,comprehensive utilization of RS and GIS technology,based on trend analysis,HURST index analysis and other methods,the temporal and spatial vegetation cover changes from 1999 to 2007 in Xinjiang were analyzed and the spatial distribution of HURST index and HURST index differences in land-use types were explored in this paper.The results show that:an increasing trend of vegetation which is not significant in the past decade is detected in Xinjiang.And changes in vegetation cover in most parts of Xinjiang would maintain the current trend.And the time series in local area is anti-sustainability.While the mean of HURST indices of NDVI time series in various land-use types is above 0.65,and the HURST indices in desert,Gobi,and bare land are significantly higher than other land cover types,the HURST indices of NDVI time series in the land-use types affected by human activities are generally low,which sustainability were lower than that of natural or semi-natural vegetation cover types.更多还原