A GIS for hydrological modelling in the semi‐arid, HAPEX‐Sahel experiment area of Niger, Africa

Abstract

During the HAPEX-Sahel experiment ( 199 1-94), water redistribution processes were studied at the meso-scale (10 000 h2) near Niamey, Niger. A project now under way at ORSTOM aims at modelling the regional water balance through a spatial approach combining GIS data organization and distributed hydrological modelling. The main objective is to extend the surface water balance, by now available only on a few, small (around 1 km2) unconnected endoreic catchments, to a more significant part of the HAPEX-Sahel square degree, a 1500 h2 region called SSZ that includes most of the environmental and hydrology measurqnent sites. GIS architecture and model design consistently consider data and processes at the local, catchment scale, and at the regional scale. The GIS includes spatial and temporal hydrological data (rainfall, surface runoff, ground water), thematic maps (topography, soil, geomorphology, vegetation) and multi-temporal remote sensing data (SPOT, aerial pictures). The GIS supports the simulation of the composite effect at the regional scale of highly variable and discontinuous component hydrologic processes operating at the catchment scale, in order to simulate interannual aquifer recharge and response to climatic scenarios at the regional scale.

Impact on Sahelian runoff of stochastic and elevation‐induced spatial distributions of soil parameters

“…These values grow from Table I. Characteristics of land surface types encountered over the Wankama catchment; after D'Herbès and Valentin (1997) and Desconnets et al (1996) a Terminology by Casenave and Valentin (1992). This alternation explains the large variability of K s for elevations above 251 m. In actual fact, the overall plateau contribution to catchment runoff is very low due to quasi-total abstraction of bare soil runoff by the vegetation bands; it may therefore be ignored.…”

“…These estimates are consistent with direct measurements of saturated hydraulic conductivity (Valentin, 1991;Peugeot et al, 1997;Vandervaere et al, 1997 (Table I), a 20 m resolution K s map was derived (Figure 1). Desconnets et al (1996) added to this table estimates of the Manning roughness parameter n for the surface feature classes encountered in the Niamey area (Table I), and obtained a 20 m resolution hydraulic roughness map. They produced a DEM from a 1/5000 topographical field survey with a 20 m resolution in space and 10 2 m in elevation.…”

Simulated impacts of climate change and land‐clearing on runoff from a small Sahelian catchment

“…Neglecting runoff to ponds created since 1950 would lead to underestimating the groundwater recharge. Along the same idea, infiltration is perhaps not limited to ponds as Desconnets et al (1996) or Leduc et al (2001) suggested, but may also occur in spreading zones. Recent surveys confirm that growing spreading zones are a common feature of the landscape.…”

“…Based on Casenave and Valentin's table (infiltration versus surface feature class), a 20-m resolution K s map can be obtained for each of the three epochs considered here, by reclassifying the respective land surface maps according to Table I. The same can be done for the Manning roughness parameter (n), using estimates produced by Desconnets et al (1996) for the surface feature classes encountered in the Niamey area (Table I). More information on the hydraulic parameter values used for the Wankama catchment model can be found in Peugeot et al (2003).…”

DEM-based modelling of surface runoff using diffusion wave equation

“…Hydrological models with a spatial structure based on digital terrain models have been developed by Beven and Kirkby (1979), Beasley et al (1980), Abbott et al (1986), Moore et al (1988), Young et al (1989), Grayson et al (1992), Palacios-Velez and CuevasRenaud (1992), Wigmosta et al (1994), Vieux and Gaur (1994), Julien et al (1995), Bouraoui and Dillaha (1996), Wang and Hjelmfelt (1998), Johnson et al (2000), Fortin et al (2001), Jain (2002), and Karssenberg (2002), and others. New models compatible with remote sensing and GIS data have also been developed recently (Leavesley and Stannard, 1990;Schultz, 1990;Wigmosta et al, 1994;Julien et al, 1995;Desconnets et al, 1996;Dupont et al, 1996;Olivera and Maidment, 1999;Fortin et al, 2001;Karssenberg, 2002). In most of these models, the gridbased catchment discretization procedure was followed and the same has been adopted for present study.…”