PLANT WATER-CONTENT AND TEMPERATURE OF THE AMAZON FOREST FROM SATELLITE MICROWAVE RADIOMETRY

In this study, an attempt is made to derive the evolution of the tempe rature and the water status of the Amazon forest canopy from satellite microwave radiometry. The Nimbus-7 Scanning Multichannel Microwave Ra diometer (SMMR) temperature-corrected tapes data are analyzed for the 6.6, 10.7, 18, and 37 GHz frequencies, at daytime and nighttime, over a zone near Manaus (3-degrees-S, 60-degrees-W), Brazil. Two periods ar e investigated: the wet (April-May) and dry (July-August) seasons of 1 985. After separating forest- from river-contaminated pixels, atmosphe ric corrections are performed for water vapor, clouds, and rain, using surface and satellite data. Algorithms are developed to model the mic rowave thermal emission of vegetation following a continuous approach and a discrete approach. A sensitivity study is performed in order to determine which frequencies are relevant to retrieve land surface para meters. The models are then used along with an optimization procedure so as to carry out the inversion of the canopy structure parameters. T he vegetation temperature and water content are retrieved through the continuous model.