DART: 3-D model of optical satellite images and radiation budget Gastellu-Etchegorry J.P., Martin E., Gascon F., Belot A., Lefevre M. J., Boyat P., Gentine P., Ader G., Deschard J., Torruella P., Chourak K. Centre d'Etudes Spatiales de la BIOsphère: UPS - CNES - CNRS - IRD (UMR 5126) Toulouse - France - E-mail: gastellu@cesbio.cnes.fr Abstract—DART (Discrete Anisotropic Radiative Transfer) was developed in 1996 for simulating radiative transfer in 3D scenes. Since then, it was greatly improved to make it more accurate, comprehensive and operational (e.g., simulation of thermal infrared and atmospheric radiative transfer). Presently, a single DART simulation gives 2 major products. (1) 3-D radiation budget of the Earth-Atmosphere system. (2) Optical remote sensing images at any altitude from bottom up to top of the atmosphere, for many view directions, simultaneously in several spectral bands, from the visible up to thermal infrared. DART works with natural landscapes (i.e., forests, field mosaics, etc.) made of trees, grass, rivers, etc. and urban landscapes made of buildings, roads, etc. Topography is simulated with digital elevation models. Atmosphere (vertical profiles, etc.) and Earth surface (spectral reflectance, etc.) databases can be used, sensor characteristics can be accounted for, etc. Moreover, a Graphic User Interface (GUI) is used to input scene parameters and to display scene and DART simulations. Recent improvements of DART (patent (PCT/FR 02/01181) are presented here. Keywords-radiative transfer, image simulation, radiation budget I. INTRODUCTION Modeling radiative transfer (RT) with terrestrial surfaces is increasingly useful for scientific applications such as vegetation studies using remote sensing images. Typically, information retrieval (albedo, temperature, Leaf Area Index: LAI, etc.) on Earth's surfaces from remotely sensed images should benefit from 3D RT models that simulate accurately their spectral bi-directional reflectance (BRDF) and temperature (BTDF) distribution functions. For example, they should help in associating surface conditions with signal characteristics such as BRDF and/or BTDF anisotropy [1]. Usually, this anisotropy limits a lot the vegetation studies with satellite images acquired under different view - illumination conditions. Associated errors depend on sun - view conditions, and on target characteristics; e.g. the albedo of a canopy with an anisotropic BRDF may be underestimated by as much as 45% if it is computed with nadir reflectance only [2]. Quantification of vegetation functioning is another important domain of application of 3-D RT models when these are coupled with leaf physiological models. Indeed, vegetation development depends on within stand radiation regime and vegetation photosynthesis potential. 3-D RT models are expected to give accurate 3-D distributions of the radiative energy that vegetation intercepts and absorbs. The 3-D RT model DART [3] was developed with these two objectives in mind. Its 1996 initial version simulates BRDFs, remote sensing images and the radiation budget of 3D natural landscapes (e.g., trees, roads, grass, soil, water) in the visible and short wave infrared domains. It was successfully used in several works; e.g., impact of canopy structure on satellite images texture [4] and on 3D photosynthesis rate and primary production rate [5] [6]. DART simulates RT with the exact kernel and discrete ordinate approaches. Landscapes are simulated as juxtaposed rectangular matrices of parallelepipedic cells (Fig. 1). Radiation is restricted to propagate in a finite number of directions (Ω i ) with a finite angular width (ΔΩ i ). Any set of N discrete directions can be selected. Any radiation W(r,Ω i ) that propagates along direction (Ω i ) at a position r has three components: total radiation, radiation that never interacted with leaf mesophyll and first order polarization. Leaves, grass Crown Gap Wout Soil, wall, trunk, roof Cell i : LAIi , LADi, ρf,i, τf,i, Ti Lake Leaves Triangle i : ρs,i(Ω’,Ω), Ti, Ωn,i Wall Roof Win=Wout Soil y z x φ θ (Ω) Transmitted sun irradiance Wout Sun irradiance Atmospheric layer i : α e m (z), ωm(z), αe p (z), ωp(z), Pm(Ω’,Ω), Pp(Ω’,Ω) Atmospheric radiance Grass Win=Wout Figure 1: "Atmosphere + Earth" simulation used by DART. Mixed "built-up/natural Earth landscape + atmosphere". 0-7803-7929-2/03/$17.00 (C) 2003 IEEE 3242