Determination of the Water Volume Variation in the Rio Negro Sub-Basin By Combination of Remote Sensing and In-Situ Data

Our study is focused on one of the major new use of spatial data and especially radar altimetry on continental waters, i.e. the ability of quantifying the role of the inundation plains in the estimate of water and sediment fluxes of the great fluvial basins of the world. Our aim is to assert the type and precision of information that can be derived from radar altimetry temporal series comparing with the in-situ records of gauging stations. We will then present a new method to determine spatio-temporal variations of water volume in the Rio Negro basin, based on the results of a SAR images classification, Topex/Poseidon altimetry data and in-situ water level measurements. We choose as a test site the Negro River sub-basin whose area is only 12 percents (700,000 km2) of the whole Amazon basin but which represents the volumetrically largest supply to the Amazon river and is estimated as the fifth largest river in the world for its water discharge. The Negro River is a northern tributary as it joins the Solimoes River to form the Amazon, downstream from Manaus. It is characterized by the black color of its water, due to a high content in organic matter dissolved load and a low detritic particulate sediment load, and is considered as a low energy system, that partly explains (with the high amount of precipitation) the considerable extension of the inundation plain. A series of SAR images from the JERS satellite enables us to distinguish between neverflooded, occasionally or always flooded area, as well as between main course of the river and inundation system at high and low stage. We use 10 years of Topex/Poseidon altimetry (1993-2003) at about 70 sites combined with measurements from 30 in-situ hydrographic stations to determine water level time series over the basin. By combining area estimates from the radar images classification with the water levels, we infer water volume change over a seasonal cycle and compare the information given by both sources of data, i.e. altimetric measurements and in-situ gauges.