Objective The objective is to investigate the feasibility of using genetic algorithms (GA) as an optimization procedure to retrieve quantitative information from satellite remote sensing data. By necessity, satellite instruments can only measure radiation levels. To convert these data into useful information on the state and evolution of the environment, it is necessary to invert physically-based mathematical models of the interaction of the radiation with the earth surface. So far, only classical optimization methods have been used to solve this problem. Raw satellite data will be simulated using an existing bidirectional reflectance model to generate data for a selection of illumination and viewing angles. Both classical optimization techniques and GA will be used to invert the physically-based model against these simulated data. The influence of instrument noise on the accuracy of the retrieval will be investigated. Classical optimization techniques and GA will be compared in terms of efficiency, accuracy and complementarity. Finally sensitivity studies will be performed to assess the optimal observing strategy. This information could be used to enhance the design of future earth observing satellite. Further work could follow either of two directions: improvement of procedures to analyze satellite data and retrieve useful information on the environment; monitoring of the evolution of terrestrial ecosystem and their interactions with the climate systems. Programme(s) JRC-EXPLORES 2C - Specific research programmes (EEC) to be implemented by the Joint Research Centre - Exploratory research -, 1992-1994 Topic(s) Data not available Call for proposal Data not available Funding Scheme Data not available Coordinator Joint Research Centre (JRC) EU contribution No data Address 21020 Ispra Italy See on map Total cost No data
