Impacts of watershed management practices on water resources and anti-salt dam design in Casamance (Senegal)

Barry Boubacar, Purdue University

Abstract

For many years, the Casamance was a self-sufficient region. Recently, this region has experienced food deficits because of the persistent drought. Given the shortage of rainfall over the last several years and the increasing accumulation of salt, the number of abandoned mangrove paddy fields is multiplying throughout the region. Because of the shortage of available fresh water for leaching the salt out of these fields and preventing the acidification of the acid sulfate soils, site development (polders, anti-salt dams, surface drainage) has become the only solution to the salt intrusion problem since the early 1970's. Today, the drought has led the farmers to move on the plateau where they practice the "slash and burn" or "bush fallow" agricultural system which is the simplest, fastest and least laborious way of preparing bush land agriculture. The alternative of clearing the land is deemed to be too laborious and impractical by many farmers relying now on cash crops production to solve their dependency for imported rice. As a consequence of these new agricultural practices, more erosion had occurred in most of the sites. The present study describes the agricultural droughts in Casamance using a method developed at ICRISAT. More especially it examines the length of dry spells, their probabilities and frequencies in the Lower Casamance region during the growing season. Three numerical models (HSPF10, CEQUEAU and SWAT) were investigated for potential use in the Djiguinoum watershed where the first anti-salt dam was built in 1984. Calibration and validation of the best suitable model were done using data collected by research teams of ISRA and ORSTOM. An evaluation and discussion of the results of the simulation model was made to evaluate existent farming systems practices on the watershed. Different scenarios involving changes in landuses and weather were analyzed and compared. Hydrologic design parameters were estimated using the selected simulation model and then, compared with those obtained by the Auvray and Rodier method widely used in West Africa for water resources assessment in small agricultural watersheds and with observed data.

Degree

Ph.D.

Advisors

Bralts, Purdue University.

Subject Area

Agricultural engineering|Soil sciences|Urban planning|Area planning & development

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS