Model intercomparison for validating the 2003 DART model

DART (Discrete Anisotropic Radiative Transfer) model was designed in 1996 for simulating optical directional images and 3-D radiation budget of heterogeneous 3-D scenes with various landscape elements (i.e., trees, water, grass, soil, etc.). It was already used for many scientific works; e.g., impact of canopy structure on satellite images texture and 3-D canopy photosynthesis rate and primary production rate. It was successfully tested against reflectance measurements and also against radiative transfer (RT) models in the frame of the RAMI exercise. Recently, DART was greatly improved to make it more comprehensive and operational. The new DART 2003 model simulates directional images in the sensor plane, for any altitude, simultaneously in several spectral bands in the whole optical domain, for natural, agricultural and urban landscapes with topography, with/without the simulation of the atmospheric radiative transfer, with/without the use of spectral databases (0.3 /spl mu/m-15 /spl mu/m), etc. In order to validate these improvements, DART 2003 simulations were tested against red and NIR reflectance values that were simulated by some RT models used in the frame of RAMI exercise: (1) two 1-D RT models (ProSAIL, 1/2 Discrete) and (2) five 3-D RT models (Flight, DART, Sprint, RAYTRAN, RGM). Results stress that DART accuracy is compatible with that of other models. Work is being conducted for generalizing this first result. Unfortunately, a few major features of DART (i.e., simulation of directional images, atmospheric RT, thermal inferred) could not be tested because other RT models do not simulate them.

[1]  Jean-Philippe Gastellu-Etchegorry,et al.  Modeling BRF and Radiation Regime of Boreal and Tropical Forests: I. BRF , 1999 .

[2]  Narendra S. Goel,et al.  Two models for rapidly calculating bidirectional reflectance of complex vegetation scenes: Photon spread (PS) model and statistical photon spread (SPS) model , 1998 .

[3]  Ferran Gascon Modélisation physique d'images de télédétection optique , 2001 .

[4]  Jean-Philippe Gastellu-Etchegorry,et al.  Recovery of forest canopy characteristics through inversion of a complex 3D model , 2002 .

[5]  Jean-Philippe Gastellu-Etchegorry,et al.  DART: a 3D model for simulating satellite images and studying surface radiation budget , 2004 .

[6]  N. Gobron,et al.  A semidiscrete model for the scattering of light by vegetation , 1997 .

[7]  V. Demarez,et al.  Modeling radiative transfer in heterogeneous 3D vegetation canopies , 1995, Remote Sensing.

[8]  V. Arora MODELING VEGETATION AS A DYNAMIC COMPONENT IN SOIL‐VEGETATION‐ATMOSPHERE TRANSFER SCHEMES AND HYDROLOGICAL MODELS , 2002 .

[9]  Nadine Gobron,et al.  Radiation transfer model intercomparison (RAMI) exercise , 2001 .

[10]  Peter R. J. North,et al.  Three-dimensional forest light interaction model using a Monte Carlo method , 1996, IEEE Trans. Geosci. Remote. Sens..

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

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

[13]  Michel M. Verstraete,et al.  Raytran: a Monte Carlo ray-tracing model to compute light scattering in three-dimensional heterogeneous media , 1998, IEEE Trans. Geosci. Remote. Sens..

[14]  W. Qin,et al.  3-D Scene Modeling of Semidesert Vegetation Cover and its Radiation Regime , 2000 .