Estimation On Water Regulation Of Ecosystem In Poyang Lake Basin

Grassland is a renewable resource, not only provides the necessary resources for human livestock and plants, but also with has a windbreak, sand fixation, soil and water conservation, maintaining the ecological balance and other important ecological functions, as well as important tourist function. Leaf area index (LAI) is an important parameter of vegetation physiological process simulation of the relationship between vegetation and climate, global climate change research has important significance. Inner Mongolia grassland belonging representative temperate grassland ecological type, representative in temperate grasslands.This study selected Landsat8-OLI data based on data of Inner Mongolia grassland LAI uses statistical models and PROSAIL radiative transfer model to inversion of grassland LAI. Choice and measurement LAI correlation best MSAVI vegetation index for ground truth LAI, constructed statistical models (exponential functions, logarithmic function, one-place linear, power function, etc.), LAI-MASVI a linear model can well simulate grassland LAI based on the correlation coefficient R2. The way of sensitivity analysis is fixing some parameters in the PROSAIL model, determine the values of the model parameters, the canopy reflectance-LAI lookup table was created, to inversion LAI of different type's grassland. Comparison precision of two methods inversion LAI, choose optimal inversion model for different grassland to obtain LAI spatial distribution. In this study, the main conclusions are as follows:1) To establish the correlation between vegetation indexes and ground measurement LAI, the best selection of MSAVI vegetation index is used to establish the statistical model. Depending on the type of grassland, build statistical model. One-place linear of LAI-MSAVI is the best statistical model to inverse LAI. Statistical model fit for desert steppe, followed by typical steppe.2) Combine Landsat 8-OLI satellite remote sensing data and ground synchronization observations LAI value using PROSAIL canopy reflectance model inversion reflectance, working a lookup table method to calculate LAI, inversion LAI image in study area. PROSAIL model has the best correlation of the meadow steppe LAI and measurement data, and PROSAIL model simulated LAI is the best accuracy in desert grassland.3) Comparative analysis of the accuracy for two kinds of grassland LAI inversion model. The use of statistical models in desert grassland (lower vegetation coverage) can accurately inversion grassland LAI. And at higher coverage of grassland (typical steppe and meadow steppe), using canopy radiative transfer model PROSAIL is more suitable for inversion LAI values.