Study On Remote Sensing Retriveral Models About Ecological Water Information Index

Soil-Plant-Atmosphere Continuum(SPAC) contains atmosphere, soil, vegetation,water and many other factors, ecological water as an important part of SPAC, itcontains water related to vegetation layer, is an important part of water cycle andwater resources monitoring. Reasonable evaluation and measurement of ecologicalwater resources is very practical to water resources integrated management and regionsustainable development.Most of the researches at home and abroad pay attention to water quantitativeresearch of farmland water, vegetation water, a variety of biochemical parameters,drought and flood hazard, for the regional-scale forest water is limited tocomprehensive statistic of site and sample monitoring. It can not adapt to dynamicchanges in overall geographical conditions. Remote sensing technology candynamically monitor vegetation, soil, etc. access large-scale ground information inshort term, objectively reflect its water conservation. Because ground surface iscomplex and changing, any kind of remote sensing band, both in the visible, nearinfrared, or thermal infrared band, and even microwave band can not fully andaccurately monitor the ecology of water resources. Quantitative remote sensingtechnology can quickly collect data and do quantitative analysis, use of this techniqueto extract water information will be an important area of academic problems.In this paper, Maoergai area in the upper minjiang which contain typicalvegetation and soil is chosen as study area, its ecological water model index system isstudied, the uncertain parameters`relations are cleared. Vegetation water content (EWT and FMC), soil moisture, soil evaporation and vegetation transpiration iscalculated by using remote sensing images and measured spectrum information,ecological water coefficient module is established. It is simple and convenient inactual ecological monitoring and correct evaluation of water conservation.The mainly content and innovation of this paperis as follows:Combined with the previous research results of the project, improve theecological water information index parameters system further, discussed indexparameters inversion method of variety of vegetation index, vegetation biochemicalindex, surface albedo, the surface temperature, soil humidity index. The meaning ofthe parameters and its weight is determined and ecological water information index iscalculated.Based on spectroscopy and the basic principle of remote sensing, the visiblespectrum can reflect the vegetation coverage, and the near-infrared spectrum can beused to analyze the vertical structure of vegetation. Vegetation sensitive band to wateris analysed using spectroscopy and RS. Various spectrum index and model relation tovegetation water content is analysed, moisture content (EWT) and vegetation relativewater content (FMC) model is established using SR vegetation index, through theresidual analysis, precision is accords with requirement, and the model has certainpractical value.Considering the complexity of geology and geomorphology in the upperminjiang, three soil water content model was set up from different anglerespectively: analyze the relstionship between near infrared spectral reflectance andsoil water content, established hyperspectralsoil moisture content model combinedwith near infrared sensitive bands; select soil line using soil sensitive band,established soil moisture content model combined with TM remote sensing imagefeatures; inversed soil moisture content model which is suitable for high forestvegetation coverage using TS-NDVI space established moisture index. The threeresultsis contrasted and the applicability of the model is analyzed.It is difficult to compute the latent heat flux in surface water cycle.BuildNIR-RED spectrum characteristic space using ETM+remote sensing image of thenear infrared infrared bands, accord to the energy balance principle, choose SEBALmodel and the improved algorithm of latent heat flux, inverse soil evaporation and vegetation transpiration water model, and the results is accord with meteorologicalstations monitoring data and related literature material.This paper established ecological water-rich index model firstly, and dividedwater-rich district of the study area, the result is consistent with the results ofbackground investigation. On this base, analyzed changing characteristics of studyregional ecological water in recent10years and predicted its future, and proved themodel applicability for forest area and in the upper minjiang.