Comparison of radiative transfer model inversions to estimate vegetation physiological status based on hyperspectral data

This study compares the performance of radiative transfer model inversion techniques to estimate leaf chlorophyll content (LCC) from summer barley based on hyperspectral data. The PROSAIL model was used to simulate vegetation reflectances. Model performance was tested against 168 ASD measurements taken under controlled conditions in an experimental station. An iterative optimization technique (IO), a look-up table (LUT) and a support vector regression (SVR) were applied to invert the PROSAIL model. A new and efficient method for LUT generation is presented which serves also as an extensive data basis to train the SVR. Highest accuracy for LCC estimation was achieved with the IO (R2=0.72). The performance dropped to R2=0.51 using the LUT but improved when the same dataset was used to apply the machine learning technique (R2 for SVR=0.67).

[1]  A. Skidmore,et al.  Inversion of a radiative transfer model for estimating vegetation LAI and chlorophyll in a heterogeneous grassland , 2008 .

[2]  W. Verhoef.  Light scattering by leaf layers with application to canopy reflectance modeling: The Scattering by Arbitrarily Inclined Leaves (SAIL) model , 1984 .

[3]  Michael E. Schaepman,et al.  A review on reflective remote sensing and data assimilation techniques for enhanced agroecosystem modeling , 2007, Int. J. Appl. Earth Obs. Geoinformation.

[4]  Yuri Knyazikhin,et al.  Retrieval of canopy biophysical variables from bidirectional reflectance Using prior information to solve the ill-posed inverse problem , 2003 .

[5]  Wouter Dorigo,et al.  Applying different inversion techniques to retrieve stand variables of summer barley with PROSPECT + SAIL , 2010, Int. J. Appl. Earth Obs. Geoinformation.

[6]  John R. Miller,et al.  Remote Estimation of Crop Chlorophyll Content Using Spectral Indices Derived From Hyperspectral Data , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[7]  Deyong Sun,et al.  A Unified Model for Remotely Estimating Chlorophyll a in Lake Taihu, China, Based on SVM and In Situ Hyperspectral Data , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[8]  C. Atzberger.  Object-based retrieval of biophysical canopy variables using artificial neural nets and radiative transfer models , 2004 .

[9]  L. Prior,et al.  Ecological Models and Data in R , 2011 .

[10]  Stuart R. Phinn,et al.  Efficient radiative transfer model inversion for remote sensing applications , 2009 .

[11]  F. Baret,et al.  PROSPECT: A model of leaf optical properties spectra , 1990 .

[12]  Gustavo Camps-Valls,et al.  Retrieval of oceanic chlorophyll concentration with relevance vector machines , 2006 .

[13]  W. Verhoef,et al.  PROSPECT+SAIL models: A review of use for vegetation characterization , 2009 .

[14]  C. Giardino,et al.  Estimation of leaf and canopy water content in poplar plantations by means of hyperspectral indices and inverse modeling , 2008 .

[15]  J. Markwell,et al.  Calibration of the Minolta SPAD-502 leaf chlorophyll meter , 2004, Photosynthesis Research.

[16]  Stephen J. Wright,et al.  Numerical Optimization , 2018, Fundamental Statistical Inference.

[17]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[18]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[19]  S. Durbha,et al.  Support vector machines regression for retrieval of leaf area index from multiangle imaging spectroradiometer , 2007 .

[20]  R. Brereton,et al.  Support vector machines for classification and regression. , 2010, The Analyst.