Assessing the influence of spectral band configuration on automated radiative transfer model inversion

The success of radiative transfer model (RTM) inversion strongly depends on various factors, including the choice of a suited radiative transfer model, the followed inversion strategy, and the band configuration of the remote sensing system. Current study aims at addressing the latter, by investigating the influence of band configuration on the automated CRASh RTM inversion approach (Dorigo et al., 2008) which is based on PROSPECT and SAILh. The tested band combinations included the configurations of two commonly used hyperspectral (HyMap, CHRIS) and three multispectral (Landsat ETM+, SPOT HRV, Quickbird) sensors which, apart from the number of bands, greatly differ in the covered spectral range. For the comparison study, reflectance data were taken with an ASD Fieldspec PRO FR field spectrometer at various intensively managed grasslands in southern Germany, and measured spectra were resampled to the five studied band configurations. Leaf area index, leaf water content, and leaf dry matter content were determined for validation purposes. Most accurate inversion results were obtained for the full-range, hyperspectral HyMap configuration, shortly followed by the multispectral Landsat ETM+ configuration and at some distance by the SPOT configuration. For the studied variables, CHRIS and Quickbird configurations provided clearly less accurate results. The obtained results indicate that an even distribution of nearly uncorrelated bands across the entire solar-reflective domain contributes more heavily to a robust inversion than a high absolute number of bands in strongly correlating waveband regions, such as provided by CHRIS. The inclusion of SWIR bands led to regularization of the leaf water retrievals and hence to stabilization of the complete inversion process. The results in this study obtained from measured data may provide an important contribution to sensor development studies, which are often based only on simulated data

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

[2]  F. Baret,et al.  Neural network estimation of LAI, fAPAR, fCover and LAI×Cab, from top of canopy MERIS reflectance data : Principles and validation , 2006 .

[3]  F. Baret,et al.  Leaf optical properties with explicit description of its biochemical composition: Direct and inverse problems , 1996 .

[4]  S. Running,et al.  MODIS Leaf Area Index (LAI) And Fraction Of Photosynthetically Active Radiation Absorbed By Vegetation (FPAR) Product , 1999 .

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

[6]  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.

[7]  W. Verhoef,et al.  A Bayesian optimisation approach for model inversion of hyperspectral - multidirectional observations : the balance with A Priori information , 2007 .

[8]  W. Verhoef.  Light scattering by leaf layers with application to canopy reflectance modelling: The SAIL model , 1984 .

[9]  S. Tarantola,et al.  Detecting vegetation leaf water content using reflectance in the optical domain , 2001 .

[10]  Frédéric Baret,et al.  RETRIEVING CANOPY VARIABLES BY RADIATIVE TRANSFER MODEL INVERSION - AN AUTOMATED REGIONAL APPROACH FOR IMAGING SPECTROMETER DATA , 2007 .

[11]  T. Faurtyot.  Vegetation water and dry matter contents estimated from top-of-the-atmosphere reflectance data: A simulation study , 1997 .