A Surface Reflectance Database for ESA's Earth Observation Missions (ADAM).
Muller, Jan-Peter¹; Lewis, Philip²; Bréon, François-Marie³; Bacour, Cédric⁴; Price, Ivan⁴; Chaumat, Laure⁴; Prunet, Pascal⁴; Gonzales, Louis⁵; Schlundt, Cornelia⁶; Vountas, Marco⁶; Burrows, John⁶; von Hoyningen-Huene, Wolfgang⁶; Guanter, Luis⁷; Fischer, Jürgen⁷; North, Peter⁸; Heckel, Andreas⁸; Straume-Lindner, Anne Grete^9
1UCL MSSL, UNITED KINGDOM; ²UCL Geography, UNITED KINGDOM; ³CEA-CNRS-UVSQ, FRANCE; ⁴Noveltis, FRANCE; ⁵Laboratoire d’Optique Atmosphérique (LOA), FRANCE; ⁶University of Bremen, GERMANY; ⁷FUB, GERMANY; ⁸Swansea University, UNITED KINGDOM; ⁹ESA-ESTEC, NETHERLANDS

Current and future missions require the simulation of ToA (Top-of-Atmosphere) radiances based on surface models of spectral BRDF. These requirements will be outlined based on several key missions from ESA. To meet these requirements a global monthly climatology, ADAM, has been constructed on a monthly time-scale on a 0.1°x0.1° grid of the entire Earth's surface (land and ice-free ocean). A companion stand-alone calculation toolkit allows the simulation of spectral directional reflectances with a 1 nm spectral resolution within the range of 0.3-4µm given any observation or illumination geometry. For land and snow surfaces, the spectral BRDF simulations is derived from normalised reflectances derived from MODIS data in seven spectral bands. For ocean areas, the surface spectral BRDF is simulated using ocean colour (SeaWifs) and sea-surface wind (QuikScat) climatologies

Over land pixels (except for snow), the spectral interpolation/extrapolation of MODIS observations over the 300-4000nm spectral range is performed using EOF (Empirical Orthogonal Functions) derived from spectral reflectance databases collected by groups around the world. The Ross-Li-HotSpot BRDF model is used to compute the land surface reflectance in any viewing geometry [4,5]. For snow covered surfaces, a different processing relying on a spectro-directional model is employed [6]. Over ocean, the surface reflectance is generated by the combination of (i) foam reflectance, (ii) water column reflectance, and (iii) specular reflectance: The spectral variation of the water column reflectance is a function of the chlorophyll concentration while the directional variation of ocean reflectance is a function of wind speed [7].

The land component of the ADAM database consists of normalised reflectances and the corresponding spectral BRDF simulation tools. Inputs for these normalised reflectances include ADAM-FDS[1] and ADAM-GA [2,3]. These datasets are themselves based on different NASA MODIS derived products such as MOD09 surface reflectance (ADAM-FDS) and the 16-day composited and processed MOD43 BRDF (ADAM-GA). Each 0.1°x 0.1°km pixel has an associated variance, the ADAM-FDS based on the spatial variability from downscaling from 500m to 0.1° pixels and the ADAM-GA on the estimated uncertainty using optimal estimation [3].

Validation of the ADAM database will be shown using both direct comparisons between ADAM-FDS and ADAM-GA as well as triple collocation and correlation statistics with MISR DHR data and instantaneous MISR surface BRF over the same 2005 epoch. Assessments of the impact of ADAM-FDS on the retrieval of Nitrogen dioxide has also been performed and will be illustrated. An interactive website with large functionality has been developed to display ADAM spectral BRFs as well as download subsets of the datasets and this will be demonstrated.

References cited
[1] Louis Gonzalez, Francois-Marie Bréon, Xavier Briottet (2010) Construction of a Global Database of Surface Reflectance and Emissivity at a Sub km Resolution. PIERS Online 01/2010; 6(2):195

[2] Jan-Peter Muller, Gerardo López, Gill Watson, Neville Shane, Tom Kennedy,
P. Lewis, Jürgen Fischer, Luis Guanter, Carlos Domench, Réné Preusker, Peter North, Andreas Heckel, Olaf Danne, Uwe Krämer, Marco Zuhlke, Carsten Brockmann, 2012. The ESA GlobAlbedo Project For Mapping The Earth's Land Surface Albedo For 15 Years From European Sensors. IGARSS 2012, Munich

[3] Philip Lewis, Luis Guanter, Gerardo Lopez Saldana, Jan-Peter Muller, Gill Watson, Neville Shane, Tom Kennedy, Jürgen Fisher, Carlos Domenech, Réné Preusker, Peter North, Andreas Heckel, Olaf Danne, Uwe Kraämer, Marco Zühlke, Norman Fomferra, Carsten Brockmann, Crystal Schaaf, The ESA GlobAlbedo Project: Algorithm, IGARSS 2012, Munich

[4] Maignan, F., Bréon, F.-M., Lacaze, R. (2004). Bidirectional reflectance of Earth targets: Evaluation of analytical models using a large set of space-borne measurements with emphasis on the Hot Spot. Remote sensing of Environment, vol. 90 pp. 210-220.

[5] Bréon, F.M., Vermote, E. (2012), Evaluation of BRDF models for the correction of MODIS time series, Remote Sensing of Environment, 125: 1-9

[6] Kokhanovsky, A. A., Zege, E. P. (2004) Scattering optics of snow, Applied Optics, 43, 1589-1602.

[7] Bréon, F.M., Henriot, N. (2006) Space-borne observations of ocean glint reflectance and modeling of wave slope distributions, Journal of Geophysical Research, 111, C06005, doi:1029/2005JC003343.

* work supported under ESA study contract No. C4000102979