11th Conference on Satellite Meteorology and Oceanography


A synergistic surface BRDF/Albedo retrieval with MODIS and MISR observations 1.INTERCOMPARISION

Yufang Jin, Boston University, Boston, MA; and F. Gao, C. Schaaf, A. Strahler, C. Bruegge, J. Martonchik, and D. Diner

A synergistic approach is developed to utilize complimentary angular sampling characteristics of MODIS-TERRA and MISR for more accurate surface bidirectional reflectance (BRDF) and albedo retrieval. Inversions of POLDER BRDF datasets demonstrate that higher albedo accuracy can be obtained with observations close to principal plane than those close to cross principal plane, especially for white sky albedo in the red band. MISR observations, which are perpendicular to MODIS-TERRA samplings in the azimuthal plane, are expected to add more constraints and improve the stability of BRDF retrieval and hence the accuracy of derived albedos. Various intercomparisons are made to investigate possible effects of the different spectral specifications and atmospheric correction procedures on surface BRDF/albedo retrievals. Results show that both MISR and MODIS observations can be used to obtain independent models of the BRDF which in turn can predict each other with mean differences mostly less than 10 percent. The individually retrieved white sky albedos are also similar, as well as black sky albedos at the common solar zenith angle. The differences between individually derived black sky albedos at other solar zenith angles are significant and indicate their different extrapolation abilities. Direct combination, as well as 'a priori' synergism, are explored to improve MODIS albedo accuracy by adding MISR observations. With its high accuracy and spatial resolution, a MODIS/MISR combined global albedo product will benefit climate models in their albedo specifications.

extended abstract  Extended Abstract (948K)

Poster Session 1, Environmental Applications
Monday, 15 October 2001, 9:45 AM-11:15 AM

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page