Lansana, Philip Karimu2016-05-252020-01-072016-05-252020-01-071987Lansana, P. K. (1987) Peak flow prediction in Sierra Leone, Masters dissertation, University of Dar es Salaam. Available at (http://41.86.178.3/internetserver3.1.2/detail.aspx)http://localhost:8080/xmlui/handle/123456789/164This study carried out in Sierra Leone considers seven watersheds from four river basins. These watersheds are; one from moa, three from sewa, two from rokel and one from pampana. There are fourteen year records for both moa and the yokel river basins, thirteen year records for the pampana and two watersheds in the sewa river basin and, twelve year records for one watershed in the sewa river basin. The Thomas-F'iering model is selected for use in Sierra Leone for the purposes of generating synthetic monthly peak flow values and it is used in generating fifty years synthetic flow values. To the annual peak flora values of the generated series the t appropriate probability distributions are fitted. The two-parameter lognormal distribution is fitted to the annual peak flow values of two of the watersheds of the sewa river basin. The Pearson type III distribution is fitted to one of the watersheds of the sewa river basin while, the extreme value type one distribution is fitted to the pampana river basin, the moa river basin and one of the watersheds of the yokel river basin. The respective distributions are used in predicting peak flow values at given return periods for each watershed. A graph of growth factor against return period is obtained. From this graph, the peak flow value at a required return period can be determined, knowing the mean of the peak flow value. The most appropriate relationship between the mean value of the peak flow values and the physical characteristics of watersheds is establishedenRivers and river engineeringSierra LeoneThesis