Quantitative Evaluation Of Vegetation Restore In The Lower Reach Of Tarim River Using Remote Sensing Data

The aim of this study is to evaluate and monitor vegetation restoration in the lower reach of Tarim Riverusing remote sensing data with different resolutions? For this purpose, we used restoration ecology,landscape ecology, remote sensing and other disciplines theory and method as a guide, and set the targetarea water transferring canal of the lower reaches of Tarim River on both sides of the river vegetation maindistribution area. Based on the ground survey data, the vegetation recovery sign was selected using thestatistical analysis and system analysis method. Based on different-resolution remote sensing data, theremote sensing parameter extraction and model inversion were studied using RS and GIS techniques. Themethod of extracting multilevel remote sensing quantitative evaluation factors of vegetation restoration wasdetermined by mining the ability to identify vegetation recovery factor of the different-resolution remotesensing data. A comprehensive evaluation index system was developed for vegetation restoration of the aridinland river basin of Northwest China. Finally, vegetation restoration in the lower reach of Tarim River wasquantitatively evaluated using remote sensing technique. The following conclusions were drawn.(1) A statistical analysis of recovery ground survey of the vegetation restoration was carried out in thelower reaches of Tarim River. The results show that the growth of tree ring of Populus euphratica Oliv., thebiomass, density and mean crown width of constructive species, the tree diameter, tree height, crown widthand community coverage of the regeneration species are the indicators of vegetation restoration. Theseindicators are the basis for the construction of evaluation the index system and multi order of remotesensing data recovery index selection.(2) Using the analytic hierarchy process and expert scoring method, the index of vegetation restorationcommunity biomass, average crown width, the right number of trees per unit area, the coverage, thenumber of patches per unit area weight coefficients and factors such as the overall recovery in the level ofvegetation score value was determined. A comprehensive evaluation index system for vegetation restorationwas then developed. A comprehensive evaluation criteria was determined for vegetation restoration, and provides a basis for quantitative evaluation.(3)To meet the needs of extracting vegetation restoration indicators using different-resolution remotesensing data, the parameters of remote sensing of vegetation recovery rate targets were calibrated. Modelinversion was carried out using multi-stage remote sensing data, the results show that:①Using multi-scalemethod and mark the spot area watershed transformation segmentation method of remote sensing data, wecan efficiently extract the constructive species of Populus euphratica Oliv. and Tamarix chinensis Lour. andnumber of trees of the study area using remote sensing data (QB, WV). The resulting accuracy was higherthan85%.②For the extraction of vegetation coverage, Hyperion data with near-infrared833.83nm, redband640.5nm composition parameters SARVI (soil adjusted vegetation index) vegetation coverconstructed extraction model (VC31.368382.086*SARVI), this highest accuracy (84.6%)outperform multispectral data model.③We analyzed the field survey data of the constructive species ofPopulus euphratica Oliv. and Tamarix chinensis Lour. and the biomass of Phragmites australis (Cav.) Trin.ex Steud and medium resolution remote sensing data. The optimal model for biomass estimation wasdetermined with accuracy higher than82%.(4) Based on the results of the vegetation restoration comprehensive evaluation index system andmulti-level remote sensing data, the response of vegetation water law in the study area was analyzed. Theresults showed that:①The average crown width, number of trees per unit area and study area itspopulation biomass of Populus euphratica Oliv. and Tamarix chinensis Lour. and groundwater depthgradient were significantly correlated. The growth recovery of Populus euphratica Oliv. and Tamarixchinensis Lour. varies from300m to800m to the river, its population biomass restore optimum area arelocated from150m to300m to the river.②During2000to2010, restore vegetation cover of waterdistrict was positively related to vegetation coverage from low to high coverage. Nature vegetation areaincreases, and the degree of aggregation of contiguous improve and the sandy saline area reduced.③Takingthe nature vegetation of34nearest groups as a reference, the water district of Kaerdayi sectional area wasset as the restoration evaluation area. Based on the results of the vegetation restoration comprehensiveevaluation index system and multi-level remote sensing data, the analysis of the results between reference zones and restoration areas comparative. The results showed that, in the coming11years, the vegetationwas recovered from the previous four to secondary, but water conditions are better compared with theregion without water transfer.In sum, the vegetation restoration visual indicators in the lower reach of Tarim River expressing indifferent resolution remote sensing data of downstream Tarim vegetation restoration on different visualindicators. After multi-level remote sensing parameter fixed and post-processed, recovery indicatorsextraction accuracy was significantly improved. Based on the results of the vegetation restorationcomprehensive evaluation index system and extraction of vegetation restoration parameters frommulti-level remote sensing data, we can carry out quantitative analysis for ecological conditions ofvegetation in the study area. This study could provide direct theoretical basis and technical support forrestoring a large area of desert vegetation in arid areas and ecological protection monitoring.