Chlorophyll Content Of Vegetation In Qinghai Lake Basin Quantitative Remote Sensing Inversion

In this study, use the Qinghai Lake basin as a study area, combined with Landsat-8 OLI and MODIS remote sensing image data obtained by collecting a variety of biophysical parameters and spectral information in the basin plots data point, the use of model-based radiative transfer model PROSPECT and SAIL model for grassland in Qinghai Lake basin chlorophyll content from remote sensing data. The main research includes the following aspects:1. The data acquisition and processing: Includes collection point field plots biophysical parameters measured pretreatment and finishing as well as remote sensing data preprocessing.2. Applicability of the model: Use the measured data and model parameters section empirical data input model basin vegetation spectral inversion calculated simulation data and measured spectral difference spectral data, according to the theory of PROSPECT and SAIL model in Qinghai Lake basin inversion applicability and practical application of the applicability of the analysis;3. The model parameter sensitivity analysis: PROSPECT and SAIL model input parameters based on the measured sensitivity model biophysical parameter data. Sensitivity is calculated based on the model and calculate the relative sensitivity of the model parameters, identifying sensitive parameters;4.Optimization method to calculate: the sensitive parameters of the model in steps according to certain values, the combination of these parameters optimization algorithm in Mat Lab using complex nonlinear equations or algorithms optimal value, simulate the closest simulation of spectral data and biophysical parameters measured spectral data;5.Inversion of chlorophyll content: the correction treated by radiometric calibration and atmospheric Landsat8 OLI and MODIS250 meter resolution remote sensing image Reflectance cost function in accordance with the simulated spectral reflectance of a comparison, find the optimal inversion reflectance data to obtain the optimal value corresponding chlorophyll concentration data, and then use intensive field data points measured chlorophyll content and inversion of the data were analyzed to verify the accuracy of the inversion.Through these studies, get the following conclusions:(1) PROSPECT and SAIL model is suitable for vegetation biophysical parameters Qinghai Lake basin inversion, spectral reflectance curves and the measured spectral reflectance curves obtained by simulation using the model in addition to the peaks and valleys of the differences and larger, the other band at the absolute deviation less than 0.02;(2) After the input parameters PROSPECT and SAIL model sensitivity analysis can understand the major impact of the chlorophyll content of vegetation visible light, green light reflectance of the most influential bands; leaf area index in the red band, near-infrared bands and shortwave infrared reflectance effects on vegetation are larger; leaf water content in the shortwave infrared reflectance greater impact on vegetation; structural characteristics of leaves in the near-infrared reflectance significant impact on vegetation; leaf angle of each band on vegetation reflectance influence are relatively small;(3) The model input parameters PROSPECT and SAIL models for sensitivity in the visible band is Cab> LAI> Cm> N> Cw> LAD. While determining selection chlorophyll content and leaf area index as sensitive parameters of the model spectrum simulation;(4) Based on the model parameter sensitivity analysis to determine the content of chlorophyll a greater impact on visible light, thus Landsat8 OLI image selection green band, red band and near infrared bands for use inversion, MODIS image select 2-channel data;(5) Through the inversion results showed that the use of PROSPEAT and SAIL model can perform anti-vegetation leaf chlorophyll content, combined Landsat8 OLI data inversion of magnitude better than the results of MODIS data inversion.