Inversion Of Leaf Area Index Of Vegetation In Temperature-corrected Radiation Transfer Model

The construction and development of high-grade highway have brought huge economic benefits to China and promoted the vigorous development of the transportation industry.At the same time,however,the construction of the highway has had a serious impact on the surrounding ecological environment,mainly including the vegetation damage,soil erosion,geological environment damage,etc.Therefore,it is especially important to monitor the ecological environment of the road area quickly and accurately.This thesis selected the leaf area index which can characterize vegetation growth status and biomass as the research target for monitoring and inversion.At present,the more methods in quantitative remote sensing research are to use satellite remote sensing data and radiation transmission model for inverting the vegetation physiological and biochemical parameters of the wide road area.In this study,the Louxin Expressway in Hunan Province was choosed as the research area.The Landsat80LI remote sensing image was selected as the data resource.Combined with data of the leaf area index mesaured synchronously in the field,the PR04SAIL radiation transmission model was used to realize the remote sensing monitoring of the leaf area index of the vegetation.The main research contents and results are as follows:(1)Temperature correction of leaf radiation transfer model.Calculate atmospheric refractive index by using atmospheric physical quantities.Make the temperature correction of leaf radiation transfer model by the theory of medium refractive index change through temperature change.The revised model further improves the accuracy of the model simulation of blade spectrum.(2)Study the inversion factor of the physical model.By extracting the single-band of remote sensing image and six vegetation indices of NDVI,TVI,SAVI,MSAVI,DVI,EVI to do the correlation analysis with leaf area index and to construct a linear regression model,Find that the correlation between single-band and leaf area index is low,and the correlation between vegetation index and leaf area index is higher,among which MSAVI and SAVI coefficient are the highest.At the same time,the coefficient of determination of MS AVI and SAVI is the largest in the linear regression model,so MSAVI and SAVI are choosed as the inversion factors of the physical model.(3)Construct a combined inversion model of locally weighted multiple regression.Construct a leaf area index lookup table with MSAVI and SAVI as inversion factors by using a modified radiation transfer model.The inverse factor constructed by spectrum of OLI remote sensing image is the prediction set.The weighting factor distance formula of the local weighted multiple regression is extended from the one-dimensional space to the two-dimensional space according to the number of inversion factors,and the wavelength coefficient k is 0.01.The combined inversion model of PR04SAIL and local weighted multiple regression was constructed by using 40 sets of data,which solved the problem of excessive data of the lookup table and successfully inverted the leaf area index of the vegetation.The combined model approach is more advantageous than the traditional lookup table method.(4)Analyze and draw the spatial distribution level map of the vegetation LAI in the experimental road area.According to the leaf area index obtained from the inversion,construct the spatial distribution level map of vegetation LAI in Louxin highway,which show that the construction of Louxin Expressway has caused some damage to the surrounding vegetation.The inversion results of image are consistent with the surveys in the field.The results show that the combined model of improved radiative transfer model and local weighted regression has good predictive ability and is suitable for large-area prediction of vegetation leaf area index.The results lay a foundation for quantitative inversion of vegetation parameters of high-grade highways in southern hilly areas by using multi-spectral data,and provide technical support for better monitoring vegetation environment in road.