The Analysis Of Relationship Between NDVI And Hydro-meteorological Factors In The Arid Areas

As an important element of the ecosystem,vegetation plays a vital role in water cycle,energy flow,and biochemical cycle.At the same time,vegetation growth is influenced by meteorological and hydrological process.In arid regions,the ecological environment is fragile due to the shortage of water resources.The healthy growth of vegetation plays a crucial role in maintaining the stable development of the ecological environment.Due to the lack of precipitation,vegetation in arid areas need to absorb groundwater to maintain normal life activities.Groundwater is a key factor affecting the evolution and distribution of vegetation.Therefore,research on the relationship between spatial-temporal changes of vegetation and the hydrological and meteorological elements in arid areas has important practical significance for water resources management and reasonable groundwater mining.In this paper,the spatial and temporal changing characteristics of NDVI,temperature,and precipitation in the middle reaches of the Heihe River basin during 2001-2011 were analyzed using remote sensing NDVI data,surface observation precipitation,temperature data.At the same time,the observed groundwater water level of 53 observation wells was used to compare 6 kinds of different interpolation types including Inverse distance weighting interpolation?IDW?,Radial basis function interpolation?RBF?,Simple Kriging interpolation?SK?,Original Kriging interpolation?OK?,Universal Kriging interpolation?UK?and Yamamoto Kriging interpolation?Yamamoto?,and using the best interpolation method analyzed the spatial-temporal variation of the groundwater level in the area from 1985 to 2011and its influencing factors.Taking into account the complexity of the relationship between NDVI and meteorological and hydrological factors,the joint entropy theory was applied to explore the relationship between NDVI and different influencing factors.Based on the Tsallis entropy,the functional relationship between groundwater depth and NDVI distribution was derived,and the formula was verified by collecting the data of grassland in shallow groundwater area in the middle reaches of Heihe River,Ejina Oasis in the lower reaches of Heihe River,Qaidam Basin map data and Hailiutu basin data.The main results are as follows:?1?During 2001-2011,the temperature and precipitation of the three stations of Gaotai,Linze and Zhangye in the middle reaches of the Heihe River show increasing trend;the average kriging method has the highest interpolation accuracy from 1985 to 2000,and the original Kriging interpolation has the highest precision from 2005 to 2011.The Kriging interpolation method modified by Yamamoto is more effective in extreme groundwater level better than other interpolation methods.The mean value of groundwater table showed a downward trend during the 27 years.The year with the most obvious downward trend was from 2000 to 2005,which was mainly due to the increase of population in the middle reaches of the Heihe River and the overexploitation of groundwater caused by the expansion of cultivated land.After 2005,due to the increase of upstream water and precipitation,and the reduction of groundwater extraction,the groundwater declining trend in the middle reaches has been somewhat alleviated,and the groundwater level along the coast of the Heihe River and the southern reaches of the middle reaches has picked up.?2?On the spatial pattern,bare land area accounted for 34%of total area,vegetation covered area accounted for 66%of total area,among them,the lower covered area accounted for 50.29%of total area,higher covered?NDVI>0.45?area accounted for 15.25%of the whole area,mainly distributed in the artificial oasis area including farmland,woodland and grassland,etc.From 2001 to 2011 year,the regional NDVI showed an overall trend,with an average annual growth rate of 0.006/yr and more than 90%of the areas in the study area showed significant or slightly increased NDVI status.?3?In desert regions??,??,temperature and precipitation are the main drivers of vegetation growth.In the oasis area?sub-area ? and sub-area ??,groundwater is an important factor influencing vegetation growth.The overall correlation coefficient between model-based simulated NDVI and measured NDVI is 0.97,and the higher simulation accuracy further proves the applicability of the model.Due to the influence of human activities,simulation accuracy is less effective in the artificial oasis?sub-area ? and sub-area V?,the shortage of data may also be a factor affecting the modeling accuracy.?4?The relationship model between NDVI distribution and groundwater depth in arid regions based on Tsallis entropy has only three parameters?m,?₁,?_*?and?₁ and?_*can be determined by measured data.The theory verified through collecting database from different regions,the optimal depth of vegetation in different regions was not the same,but the correlation between simulated NDVI and measured NDVI was relatively high.It shows that the relationship model between NDVI and groundwater depth based on Tsallis entropy is better and has wider applicability.However,in areas influenced by human activities,the simulation effect is not so ideal,because under the influence of human irrigation,water for mataining vegetation growth has a large part from the irrigation water,so this model is more applicable to that groundwater-dependent regions for vegetation growth in the arid area.