Jan-Peter Muller - 613 Seminar Series

University College London

Within the ESA GlobAlbedo project we have developed an unique approach to the retrieval of broadband BRDF/albedo using data fusion of multiple satellite sensor data from European sensors at the top-of-canopy level and an optimal estimation approach employing a climatology derived from MODIS Collection 5 data. I will show methods and results as well as their detailed validation and visualisation.

Since January 2014, the 4-year EU-FP7 QA4ECV (Quality assurance for Essential Climate Variables) project has been extending these results to include a spectral MODIS prior and an evaluation of the error propagation through a traceability analysis of the algorithm. In parallel, another group at UCL in collaboration with JRC Ispra have been developing a radiative transfer modelling framework for investigating this error propagation through the retrieval algorithms. Together with FastOpt GmbH, fapar and an effective LAI have also been derived from the GlobAlbedo BRDF products using a novel method of upscaling using energy conservation.

I will show some of these recent results from QA4ECV and discuss the plans to include MISR & MODIS data in future as well as the fusion of AVHRR-LTDR and GEOstationary products to go back in our ECV record to 1982. In addition, I will show some initial results on the data fusion of MISR and MODIS for the retrieval of instantaneous BRDF/albedo over sea-ice floes.