The NOAA/AVHRR NDVI data, as an important index, is universally used to analyze vegetation dynamic growth and global climate change. In order to explore the interannual variation in various vegetation types, this study analyzes the temporal dynamic of Normalized Difference Index (NDVI) for 10 vegetation types in Xinjiang and investigates the relationships between NDVI and climatic variables over the period of 1982-2003. The results demonstrate that: 1) Average NDVI of the study area increased significantly at rates of 0.58% each year in the growing season (April-October), 0.53% in the spring (March-May), 0.50% in the summer (June-August) and 0.65% in autumn (September-November). The interannual variation of NDVI in growing season, spring, summer, and autumn were increased significantly (P

使用归一化植被指数(Normalized Difference Vegetation Index, NDVI)和气象数据研究了全球气候变化背景下，1982—2003年新疆10种植被类型NDVI时间动态变化及其与气候因子的关系。研究区植被生长季NDVI以0.58%的年平均增长率显著增加，并且春、夏、秋三季的总体及各种植被类型NDVI都显著增加(P

The dynamic change of 23-years vegetation cover in Wushen County ofInner Monggulia was analyzed using NASA GIMMS(Global Inventory Modeling and Mapping Studies) Normalized Difference Vegetation Index(NDVI) data set. The average value of annual max NDVI in latest five years and the slope of annual max NDVI were used to analyze the vegetation distribution in Wushen County during the last 23 years. Mean absolute deviation of maximum NDVI was also used to study the change intensity of the vegetation cover. The highresolution images Landsat TM data of the year 1986 and 1994 and ETM+ data of 2002 were used to find the change patterns of vegetation in Wushen County. The influence of rainfall on vegetation cover in Wushen County was explored through correlation analysis between annual rainfall and annual maximum of NDVI. The result shows that, the change of the vegetation cover in the last 23 years is consistent with the change of rainfall. The annual maximum of NDVI is significantly related to the rainfall during August to July of second year. The significance of correlation in grassland and shrub forest is higher than in cropland and arboreal forest. The lowest rate of vegetation cover is in desert area of northwestern Wushen County; the highest is In cropland in the southwest of the county. Generally human activities have been promoting the increase of vegetation cover. But in some areas the vegetation cover presents a trend of decrease due to dry weather.

利用GIMMS（Global Inventory Modelling and Mapping Studies）数据集的8km的AVHRR NDVI（均一化植被指数）数据，分别用年最大NDVI的平均值和年最大NDVI的斜率对1981年～2003年23年间的内蒙古乌审旗植被覆盖度的空间分布及其动态变化进行了分析，并通过比较降水量与年最大NDVI的年际变化以及分析降水量与年最大NDVI之间的相关系数研究了降水对乌审旗植被覆盖的影响；利用1986年，1994年的TM和2002年的ETM+数据分析了乌审旗植被的主要变化类型。结果表明，这23年间植被覆盖度和降水量的年际变化趋势基本一致，年最大NDVI与前一年8月至当年7月的降水有最好的相关性，并且在地表覆盖为草地和灌木林地的地区，二者的相关性要高于地表覆盖为耕地和乔木林地的地区；植被覆盖度最低值分布在地表主要覆盖为沙地的乌审旗西北部，植被覆盖度最高值分布在地表覆盖为耕地的东南部；人类活动的积极影响使得植被覆盖度总的趋势是在增加的，但由于气候的干旱趋势使中部一些乡镇的植被覆盖度是在减少的，其余的地区都是在增大的，且变率很大。

In the present study, the normalized difference vegetation index (NDVI) estimated from MODIS data were employed to analyze the temporal and spatial dynamic of different grassland types during growth season (April to October) in the recent 10 years (from 2003 to 2012) in Xilingol. The NDVI decreased from northeast to southwest in Xilingol in spatial. The maximum spatial differences occurred in 2005 and the minimum spatial differences occurred in 2007. The spatial variation of the same grassland type was similar during the study period. Temporal series analysis showed that the whole regional NDVI in 2012 was the maximum which followed by 2003. However, the minimum regional NDVI variations occurred in different period for different grasslands. For example, sandy grassland and desert steppe had minimum regional NDVI in 2005 and major NDVI fluctuations from 2004 to 2006, however, meadow steppe and typical steppe had minimum regional NDVI in 2007 and major NDVI fluctuations from 2007 to 2009. Spatial variation analysis showed the coefficient of variation (CV) of sandy grassland from 2003 to 2012 was less than 0.1 which indicated this type grassland was very stabilization. The CV of meadow steppe, typical steppe and desert steppe gradually increase which also was stable at 0.1 to 0.2. These results indicated that the ecological vulnerability of Xilingol was stabilization. The correlation analysis between main meteorological elements and NDVI indicated that there was a significant positive correlation between NDVI and precipitation which has a coefficient of 0.83. However, there was negative correlation between sunshine period and NDVI and no significant relationship between average temperature and NDVI.

利用MODIS数据计算得到的归一化差值植被指数(NDVI), 对内蒙古自治区锡林郭勒草原不同类型草地近10年(2003－2012)生长季(4－10月)植被状况时空特征和变化情况进行研究。结果表明, 锡林郭勒草原NDVI总体表现为由东北向西南递减的规律, 空间差异2005年最大, 2007年最小, 且不同年度间相同类型草原空间变化基本一致。全区整体年际间变化表现为2012年最大, 2003年次之;沙地草原和荒漠草原在2005年变化最小, 波动主要在2004－2006年;草甸草原和典型草原则在2007年变化最小, 波动主要在2007－2009年。空间变异分析显示, 2003－2012年沙地草原变异系数(CV)小于0.1, 属非常稳定型, 草甸草原、典型草原和荒漠草原的CV值依次增大, 但都介于0.1～0.2, 属于稳定型。说明近10年来锡林郭勒草原状况较为稳定。通过对区域主要气象要素与NDVI间关系的分析, 得到NDVI主要受降水影响, 二者之间存在显著正相关关系, 相关系数高达0.83;日照时数与NDVI间存在相反的变化趋势;而平均气温与NDVI没有明显关系。

The spatial and temporal variation of NDVI in the watershed of Nujiang River in Tibetan Autonomous Region was studied by using each ten day period maximization NDVI data set (S10) product from December 1998 to November 2008. The result indicated that vegetation was good and little change occurred. With the data of monthly mean temperature, precipitation and sunshine from December 1998 to November 2008 at 9 meteorological stations in this region, the relation between NDVI and the climate factors was studied and it was found that the temperature in April might influence the regreening in terms of accumulated temperature, and the temperature in May, June and July affected the grass growth.

利用1998年12月-2008年11月西藏怒江流域每旬最大化NDVI数据集(S10)产品，研究了10年内流域NDVI的时空变化，发现研究区植被情况总体比较好且在研究时段内变化不大。结合流域内9个气象站点1998年12月-2008年11月的月平均气温、降水量、日照时数资料，研究了流域各类型NDVI与气候因子的关系，发现4月的气温可能以积温的形式影响牧草返青，而5、6、7月的气温影响牧草生长，这对草地放牧有一定的指导意义。

The vegetation coverage changes in Korla region have crucial impacts on local and regional climate, agriculture and economy. This study uses the Landsat TM data to investigate the vegetation coverage changes in this region from 1990 to 2011 (four periods in 1990, 1998, 2006, and 2011). The vegetation coverage is calculated using the NDVI data, derived from raw TM images, and a mixed pixel model. We further classify the vegetation coverage into four categories and analyze the main characteristics of its changes using transpose matrices of each vegetation coverage category and land use data. The results show that, overall, the vegetation coverage has improved from 1990 to 2011 in the Korla City region. The area with increased vegetation is 23.8% higher than that with decreased vegetation. The vegetation coverage in southeastern sector oasis plain greatly increased, while vegetation degraded in the north and south region.

以库尔勒市域为研究区，基于1990、1998、2006和2011共4期TM遥感影像提取归一化植被指数（NDVI），将NDVI结果输入到像元二分模型中计算得到研究区各时期植被覆盖度，然后根据研究需要将植被覆盖度划分为4个等级，最后计算覆盖度差值并结合各级覆盖度转移矩阵和土地利用情况分析了库尔勒市域植被覆盖度动态变化特征。结果表明，在1990、1998、2006和2011年间，库尔勒市域总体植被覆盖情况有所改善，植被恢复改善面积比退化面积多23.8%，其中东南部的扇形绿洲平原植被状况改善明显，北部和南部区域植被有所退化。

16-day composite data set of normalized difference vegetation index （ NDVI ） from an moderate-resolution imaging spectroradiometer（MODIS） was used to capture essential feature of temporal and spatial vegetation variability in Ili River valley of Xinjiang during the 2000-2010 period. Various statistical analyses involving vegetation anomaly index, linear regression analysis and Hurst index were applied. Examination shows as follows：（1）Vegetation cover of Ili River Valley increases firstly and then decreases as altitude increases，with highest vegetation cover located in the area with the altitude from 2 000 to 2 500 m. Generally，vegetation cover of different elevation zones of Ili River valley followed a decreasing trend during the period of 2000-2010，not including the areas located below 1 000 m above the sea level. （2）As vegetation anomaly index is subject to humidity of the air，so the highest spatially averaged vegetation anomaly index value was observed in the wettest year of 2000，and the lowest value in the driest year of 2008, and changes of vegetation anomaly index of different area exhibit varied patterns. （3）The areas with increasing or decreasing of vegetation cover account for 4.09% and 19.34% of the study region’s total area，respectively. The increasing area’s regions are mainly located at the plains of both banks of the Ili River. The decreasing area’s regions are mainly located at the both ends of the Wusun Mountains and the low mountain areas surrounding the study region with the altitude of 2 000 m. Rescaled range analysis shows that changes of vegetation cover in the Ili River valley exhibits strong sustainability，which means that the change of vegetation cover will follow the same variation trend in the future some period in the Ili River valley

利用2000-2010年16 d合成的MODIS/[NDVI]数据，结合植被异常指数、趋势线分析和Hurst指数等分析方法，对新疆伊犁河谷植被覆盖的时空变化特征进行了分析。结果表明：（1） 伊犁河谷内植被覆盖随海拔增高而先增加后减小，最高植被覆盖区位于2 000～2 500 m的高程带；2000-2010年伊犁河谷各高程带内植被覆盖整体下降趋势明显，但海拔低于1 000 m的高程带除外。（2）受干湿环境影响，伊犁河谷全区平均被植异常指数最高值出现在降水最多的2002年，最低值出现在降水最少的2008年，但不同区域植被异常指数的变化存在较大差异。（3）伊犁河谷内植被覆盖增加和减小的区域分别占总面积的4.09%和19.34%，增加区域主要位于伊犁河两岸的平原区，减小区域主要位于乌孙山两端以及伊犁河谷周围海拔2 000 m左右的低山区域；变标度极差分析结果表明，伊犁河谷内植被覆盖年际变化呈现很强的持续性，未来一定时间内将保持现有变化趋势不变。