JLE

Science et changements planétaires / Sécheresse

MENU

Impact of the Manantali dam on the modifications of the Senegal river basin hydrological regime: A statistical approach Volume 20, issue 1, janvier-février-mars 2009

Figures

See all figures

Authors
Département de physique Faculté des sciences et techniques Université Cheikh-Anta-Diop Dakar Sénégal, Département de mathématiques Faculté des sciences et techniques Université Cheikh-Anta-Diop Dakar Sénégal, Département de géologie Faculté des sciences et techniques Université Cheikh-Anta-Diop Dakar Sénégal

It is well known that climate change has significantly impacted both rainfall and runoff in West and Central Africa since 1968. The Manantali dam is a multipurpose dam built on the Bafing river, a tributary of the Senegal river that controls a major part of the total flows. It is important to know the effect of this dam on the hydrological regime of the Senegal river for water resource management. To observe this effect, we first analysed annual time series of mean annual flows of Bafing-Makana (upstream of Manantali dam) and Kayes and Bakel (downstream of this dam) from 1961 to 2006. This period includes that of Manantali dam operations (1988). We did not note any significant changes. We then applied statistical tests for homogeneity using Khronostat software. The null hypothesis was H 0: “Time series is homogeneous.” H 0 was rejected by all tests for homogeneity for all gauge stations at the annual scale. Change appears for all series before the dam construction and shows a decrease in mean flow. Dam operations are not known at the annual scale. We then analysed monthly flows at the same stations, using the same tests for homogeneity. H 0 was once again rejected for almost all months. We did however note that for the Bafing-Makana station, a decrease in monthly flows appeared for all months before the beginning of the dam operations (1985). For the Kayes and Bakel stations, at the monthly scale, flows increased after dam construction during the low-flow months and decreased before dam construction during the high-flow months. The fact is that in the time series we used, there were two signals: one caused by climate change and another caused by dam operation. Statistical tests for homogeneity detected the climate change signal at the annual time scale. At the monthly scale, these tests detected the signal from the dam only during low flow when this signal is strong enough to be detected.