南疆西部不同植被类型NDVI对风沙灾害的响应

Response of NDVI of different vegetation types to aeolian sand disaster in west of southern Xinjiang

不同植被类型对风沙灾害的响应研究有助于风沙灾害的遥感监测评估和防灾减灾工作。采用传统统计学和空间自相关方法，利用MODIS的每日反射率和土地类型产品，分析了风沙灾害后南疆西部不同植被类型NDVI的变化规律及其空间分布特征。结果表明：风沙灾害后，研究区不同植被类型NDVI的差异在增大，NDVI变化值在一0．3865~0．4148之间，NDVI减小的面积占整个研究区面积的54．98％，变异系数增大值在2．44％～36．75％之间；受到风沙灾害的影响，研究区植被NDVI全局Moran’SI系数从0．7982减小到0．6786，但在空间上仍存在显著的正相关性，具有显著的空间集聚特征；由NDVI差值的局部空间自相关指标集聚图以及不同植被类型NDVI差值的空间关联区域面积百分比，可以发现落叶针叶林和裸地或低植被覆盖地受风沙灾害的影响较小，郁闭灌丛和作物受风沙灾害的影响较大。上述关于风沙灾害后南疆西部不同植被类型的变化及其空间格局的研究结果，不仅证明了防护林的风沙阻挡作用，也可以为风沙灾害的防治工作提供参考依据。

The west of South Xinjiang lies in the aeolian sand disaster zone. Normalized difference vegetation index （NDVI） is widely used as an effective indicator for monitoring changes of vegetation and ecological environment. Research on response of NDVI of different vegetation types to aeolian sand disaster is helpful for monitoring and evaluating the effect of the aeolian sanddisaster using remote sensing as well as combating and reducing disaster. Base on MODIS land surface reflectance products and land cover type product, this paper analyzed the changes and spatial distribution characteristic of NDVI index of different vegetation types after aeolian sand disaster in the west of South Xinjiang. The results showed that the numerical value of NDVI varied remarkably in space, the NDVI difference ranged from 0.05 to 0.31, the area of NDVI decreases accouted for 54.98%, the variation coefficients of NDVI were from 2.44% to 36.75%. Under influence of the aeolian sand disasters, the global Moran＇s I coefficient of NDVI decreased from 0.798 2 to 0.678 6, but there were still a significant positive correlation and spatial agglomeration pattern. Based on local Moran＇s I indicators of NDVI difference of different vegetation types, NDVI difference among vegetation types has significant spatial dependence and spatial heterogeneity. Synthesizes the above analysis and the cluster type area percentage of NDVI difference of different vegetation types, the results also revealed that deciduous coniferous forests and barren were less affected, dense shrub and croplands were greatly affected by the aeolian sand disaster. All of these findings theoretically proved that protection forest around croplands was a natural barrier to prevent the wind and sand and provided a scientific basis for preventing the aeolian sand disaster.