Seasonal And Interannual Variations Of VegetationVertical Belt Within The Kaidu River Basin

The Kaidu River,the mother river of Xinjiang Bayheng Guo Leng Mongolia Autonomous Prefecture and the major influx of Bostern Lake,the largest inland freshwater lake in China,is the most important water source for meeting the life,industrial,and ecological demand in the Yanqi oasis.It is believed that the climate and the vegetation are the major factors affecting the flow capacity of the Kaidu River.To deep understand about the influence mechanism by vegetation,this paper has analyzed the differences of vegetation growth at different altitudes at the upstream basin of the Kaidu River,and expounded the seasonal and interannual variation of vegetation growth along vertical belt.In detail,the Landsat remote sensing images from 1977 to 2015 and the DEM had been corrected by the analytical tools from Arcgis and ENVI,firstly;secondly,drawn the "NDVI-cover" distribution curve at different heights by MATLAB programming,and determined the vegetation vertical zone based on its morphological and distribution eigenvalues;Finally,analyzed the seasonal and interannual variations of whole study area and each zone from the aspects of growth status and coverage.Through research and analysis,some new analytical approach have been estimated,as follows:(1)A method of weakening the joint of remote sensing image has been proposed: a MATLAB programming based approach to extract and fit the effective pixel values of the overlapping region in single band image.Then,according to the fitting relation,the correlation function between the corresponding pixel values is determined.Finally,used the correlation function to correct the pixel value of the whole image and then stitching.(2)A method of determining the vertical zone of vegetation based on remote sensing image has been proposed:a NDVI value of each pixel based approach to determine the vegetation growth status by the;at different altitudes,determined the vegetationcoverage by the ratio of NDVI effective pixel area to the total area.Then,determined the vegetation growth according to the "NDVI-cover" distribution curve and the NDVI distribution eigenvalues,and determined the vegetation vertical zone's boundary according to the difference of the growth.The vegetation vertical zone determined through the above method is precise compared with the vertical band divided by the community type.This method provides a way to dynamically monitor the movement of vegetation vertical zone and provides a more accurate method for determining the vegetation vertical zone boundaries in the future.In this paper,six vegetation vertical zones were identified based on the TM remote sensing images in 2006,and are represented by I-VI: I(1528-1848m),II(1848-2338m),III(2338-3308m),IV(3308-3828m),V(3828-4368m),VI(4368-4808m).The III zone area is largest and the vegetation growth is also the best.In order to determine the boundaries of vegetation vertical zone more accurately,this paper divides subzones of III zone.From I to III,there was a positive correlation between vegetation growth and elevation,and the vegetation growth was negatively correlated with altitude from III to VI.In addition,used the NDVI-coverage distribution curve from February to November 2015 to determine the vegetation growth zone(indicated by(1)-(6)),and analyzed the changes of vegetation growth and vegetation growth in seasons.(1)Annual change of vegetation growth: From the whole study area,between 1977 and2015,vegetation in 1977,2014 and 2015 was the best and in 1993 was the worst.From the vertical changes of vegetation zones,the growth of vegetation in zone III is better than that in other zones.From the vegetation growth of vertical zone in the interannual differences,from 1977 to 2006,the vegetation growth of each zone was reduced from the maximum to the minimum and then gradually recovered.From 2006 to 2011,vegetation growth of each zone was decreasing,and the fluctuation range of different vertical bands was different.From 2011 to 2015,vegetation growth of each zone is on the rise.(2)Seasonal variation of vegetation growth: From the entire study area,August planting was the best growth;from February to July,the vegetation coverage fluctuated greatly,and the growth condition was getting better;from September to November,the vegetation growth was rapidly declining and the coverage was also significantly decreased.The differences of every zone in vegetation growth: The vegetation growth of each zone was best in August and September,and was significant differencesbetween February and July,and was declining from September to November.Vertical migration of vegetation zones with seasons: February to March,the elevation range of each belt upward;March to June,the elevation range of each belt moved downward;from June to September,the elevation range of each belt was different;from September to November,the elevation range of each belt was not obvious.