Based On Remote Sensing Monitoring Model Of Winter Wheat Leaf Area Index Of The Environmental Ccd Data

The leaf area index (Leaf Area Index, LAI) regarded as one of the attribute vegetation tube level structure basic parameters, is an important structural parameter in a land process research. The LAI is a important agronomy parameter to reflect the crops growing trend and a crop yields forecast, the accurate crops leaf area index data can provides the effective data resource for the crops growing trend monitor, the crops recognition, the per unit area yield forecast that even grain yield's estimation.HJ-1 satellite is the independent research and circulator environment disaster monitors constellation in china 2009, which with two optical satellites HJ-1-A and HJ-1-B load two wide width multi-spectrum CCD cameras separately, satisfies the two days revisit period which with the significant meaning for the crops' growing trend monitor and the yield estimation, the vegetation and the river system condition investigation.This research take the Yucheng in Shandong province as the research area, proposed the HJ CCD data's atmospheric adjustment method based on the 6S model, used the Hyperion hyper spectrum resolution data to perform the HJ-1 A/B CCD spectrum simulation, validated the HJ vegetation index authenticity by taking the Hyperion data as the reference. Four vegetation indexes selected as estimate factors by the HJ CCD data including the ratio vegetation index (RVI), the normalization vegetation index (NDVI), the soil adjustment vegetation index (SAVI) and the enlargement mode vegetation index (EVI), integrated the ground survey LAI data to establish the top leaf area index inversion model and to examine its precision. The primary contents and the conclusion are of the research listed as follows:1) Atmospheric correction of HJ-1 CCD data was implemented based on look-up table which is created by using 6S model to acquire land surface reflectance. The result proved that the image after atmospheric correction was much clearer than the original image, and the attenuation of NDVI was rectified.2) spectral reflectance data from EO-1 Hyperion of close date was used to simulate the band reflectance of HJ-1 A/B CCD.4 vegetation indices (NDVI, RVI, SAVI and EVI) were computed from both simulated and actual HJ-1 A/B CCD band reflectance. Comparisons between simulated and actual HJ-1 A/B CCD band reflectance, as well as that between simulated and actual HJ-1 A/B CCD vegetation were made to validate the VI product of HJ-1 A/B CCD. The R2 between simulated and actual HJ-1 A/B CCD band reflectance were 0.841,0.895,0.914 and 0.926 for blue, green, red and near infra-red band, that between simulated and actual HJ-1 A/B CCD vegetation indices were 0.979,0.977,0.921 and 0.975 for SAVI, RVI, NDVI and EVI. The standard deviation of differential images between real and simulated HJ-1 CCD VI are 0.052, 0.527,0.073 and 0.133. The result showed that the VI product from HJ-1 A/B CCD is consistent with that from simulated VI from Hyperion; the VI variation could be effectively reflected in HJ-1 A/B CCD,with the potential to inverse the Leaf Area Index。3) four vegetation indices (NDVI, RVI, SAVI, EVI) were calculated to analyze the relationship with observed LAI. Compared among the four models, it proved that the relationships between LAI and the four vegetation indices are all acceptable, and RVI is the best parameter to estimate LAI (LAI=2.967Ln(RVI)-1.201). With the RVI-LAI model, LAI map of Yucheng had been retrieved in May,2009, and the average error is about 19%, which means the model achieves a high precision.