Spatio-temporal variability of hydrological regimes around the boundaries between Sahelian and Sudanian areas of West Africa: A synthesis

Summary

Abundant information is available on West African drought and its hydrological and environmental impacts. Land-use and climatic changes have greatly modified the conditions of Sudanian and Sahelian hydrology, impacting the regime and discharge of the main rivers. Human pressure on the environment (significant increase in crops and disappearance of natural bushes and landscapes, for example) has led to severe soil crusting and desertification throughout Sahelian regions.

Despite recent increases in rainfall, the drought has not ended, resulting in two different hydrological evolutions. In the Sudanian areas, stream flows have been reduced, sometimes as much as twice the rainfall reduction rate. In the Sahelian regions, runoff coefficients have increased to such a degree that discharges are increasing, in spite of the reduced rainfall.

The main goal of this paper is to synthesize the recent advances in the Sahelian and Sudano-Sahelian West African hydrology. The other objectives are two fold: First, to discuss the “Sahelian Paradox” (the increase in runoff in most of the Sahel during the drought, at least during the 1968–1995 period, as described in the 1980s) and paradox of groundwater highlighted in the square degree of Niamey (the rise in water table levels in some endorheic areas during the same drought, evidenced in the 1990s), and second, to attempt to define the application of their respective geographical areas.

The land-use changes act as a general factor of hydrological evolution of soils and basins, while some spatial factors explain the great variability in the response to environmental evolution, such as endorheism, geological context, latitudinal climate gradient, and local hydrodynamic behaviour of environment.

This paper is literature-based, and incorporates current research advances in the field, as well as a prospective focused on resources and socio-economic impacts.

Introduction

The severe drought, and its hydrological impact, endured by West Africa over the past 40 years or so was the key motivation for setting up the AMMA-CATCH observing system (Lebel et al., 2009). The intent of this paper is to provide a general perspective on how the long lasting rainfall deficit impacted runoff and large river discharges at the regional scale. The decrease of the mean annual discharge of the largest rivers of the region, namely the Senegal and Niger rivers, in proportions almost twice as large as the decrease in rainfall is a well established fact for the period 1970–2000 (see e.g. Andersen et al., 2005, Lebel et al., 2003). Similar trends were observed on smaller river systems (Le Barbé et al., 1993, Mahé et al., 2000) while, at the opposite, other studies pointed to a runoff increase in some Sahelian catchments (Albergel, 1987, Mahé et al., 2005, Séguis et al., 2004). This paper is a factual update on these different reactions of the West African river systems to the drought with the aim of discussing the behind the scene factors involved in this variability, land surface changes being the main factor.

Land-cover changes, most often resulting from land-use changes, trigger alterations in hydrodynamic soil surface behaviour – for instance increase in bare soils areas causes erosion and soil crusting – and can impact the local and regional water budget.

Studying the relationship between the evolution of the vegetation cover and the evolution of discharges or runoff, a number of papers show that deforestation and soil compaction cause an increase in runoff. This seems to hold for very different climates, from humid and tropical (Fritsch, 1990, Calder et al., 1995, Scott Munro and Huang, 1997), to Mediterranean (Sorriso-Valvo et al., 1994, Croke et al., 1999) and temperate (Cosandey et al., 1990, Hudson and Gilman, 1993, Andreassian, 2004). Arid or semi-arid environments are especially sensitive in this respect as shown by Snelder and Bryan, 1995, Bergkamp, 1998, Casenave and Valentin, 1992.

Trees and litter constitute a stream flow regulator and a protective screen for the soil against the splash of rain drops (Croke et al., 1999, Descroix et al., 2002, D’Herbès and Valentin, 1997, Fritsch, 1990, Scott Munro and Huang, 1997, Sorriso-Valvo et al., 1994) and against evaporation (Braud et al., 1997). Inversely, soil crusting provokes an increase in runoff, conforming to observations made by Valentin and Casenave, 1992, Vandervaere et al., 1997 in the Sahel, and Janeau et al. (1999) in Mexico.

In conclusion, natural vegetation – and to a lesser extent fallow land – enhances the water holding capacity of watersheds, thus reducing runoff. Enhanced soil water holding capacity increases the buffering effect of the soil on rainfall and delays the runoff processes, hours or days after a rainfall event. At the basin scale, dense vegetation causes soil attenuating low flows and floods, generating higher base flows than bare soils or cultivated areas with overexploited (or crusted) soil characteristics do.