ESA     
       17-Jul-2014
envmail@esa.int










	
dcSearch
dc
dc

Surface Water Storage and Runoff: Modelling, In-Situ Data and Remote Sensing
Delclaux, F.1; LeCoz, M.1; Leblanc, M.2
1IRD; 2James Cook University

Lake Chad (Africa) experienced a drastic shrinkage during the 1970s and 1980s. To understand the links between climate, hydrology and water use, we used distributed models of the basin water cycle, one with a routing algorithm THMB forced by a LSM, the other derived from THMB with an internal rainfall-runoff production module.

SRTM30" DEM was integrated after some processes: calibration, which showed a +3 m shift of SRTM data with respect to local geoid ones; generation and integration of a Lake Chad bottom DEM; DEM aggregation from 30" to THMB resolution 5 with a set of operators: minimum, maximum, mode, median, mean, nearest neighbour. Bathymetry comparison revealed large differences only for minimum and maximum. Analysing water volume across the basin showed that only nearest neighbour depicted local depressions that play important role as they recycle water by evaporation. Drainage network generation with minimum gave the best results (48% overlapping ratio). But, as network from nearest neighbour was consistent enough, this method was finally retained.

Model comparison showed a better performance for the modified THMB than for LSM+THMB as Nash index increased from -50% to 62% (Logone) and from -28% to 54% (Chari). Introducing a function governing sub-surface drainage improved model by about 8%. Finally, simulations show that Lake Chad shrinkage is mainly due to a decrease by half of the Chari-Logone system inputs while irrigation intake is less than 5% lake input.

 

Legal Disclaimer - Glossary - FAQ - Site Map - Contact us Last updated: 09-Feb-2010