Remotely Sensing Model Research Of Eco-water Storage Index

Soil water plays an important part in Eco-water, it is the link between surfacewater and groundwater.At present,many researchers established numerous soilmoisture model by TM and ETM and ASAR remote sensing data, but the applicationsof various model have certain limitation.Such as apparent thermal inertia method islimited by region and season; Crop water shortage index method is complex incalculate and is rather poor in real-time anything. At the same time, it has achievedmuch success that many researchers estimate the soil water content, using spectraldata measured in countryside. Beside, the teams have done some research in theprophase of moisture content.This article chooses the upper MinJiang River as astudy area，on the basis of predecessors’ research, based on theories and methodssuch as ecology, feature spectrum, using high-tech means such as remote sensing.processing and analyzing a large amount of data real-time monitoring, Finally,establishing the ecological water information index in study area and inverting soilmoisture.In this paper, the main work done is as follows:(1) In this paper, on the basis of researching on soil spectral’ nature, the soilspectral reflectance not only was analysis, but also was made six mathematicaltransformations, such as reciprocal transformation and logarithm transformation and1st_derivative transformation. This process was programmed by spectral math ofENVI. The result shows that the soil spectral reflectance has a larger difference inspectrum after spectral mathematical transformations.Then,we can use this differenceto find the sensing spectrum.(2) In this article, the writer choose sensitive bands to establish the model ofstudy area, using the method that combined related-analysis with the multiplestepwise statistical methods. In the first, the author analysis the relationship betweensoil moisture and the soil spectral reflectance and mathematical transformation,calculate the coefficient related, and choose these bigger ones act as the initialsensitive bands. Besides, the writer choose the final sensitive bands using the multiple stepwise statistical methods.It is better to establish a regression equation using the twomethods all than using either method.(3) In this paper，the author invert the soil moisture content in study area by thesensitive bands of each spectral mathematical transformation，calculate the model’accuracy and test the model’significance.In the end, the writer choose the best modelact as the soil moisture content model of the study area.