Evaluation of Runoff Prediction Using Synthetic Rainfall and CREAMSby J. E. Peter Green, Univ of Natal, Pietermaritzburg, S Africa, Univ of Natal, Pietermaritzburg, South Africa,
C. T. Haan, Univ of Natal, Pietermaritzburg, S Africa, Univ of Natal, Pietermaritzburg, South Africa,
Document Type: Proceeding Paper
Part of: Watershed Management in the Eighties
Of all the climatological data, rainfall is the variable to which a watershed responds most significantly. The amount and sequence of occurrence of rainfall are most important in the prediction of runoff. This paper reports on research that was conducted in an attempt to evaluate the use of synthetic rainfall data to predict runoff volumes that can be expected from a watershed. The runoff predicted by the CREAMS model from synthetic rainfall (generated using a Markov chain-lognormal model) and the original historical rainfall was compared in terms of accumulated monthly and annual yield. The runoff from the historical rainfall exceeded the runoff from the synthetic rainfall by approximately 25 percent although no significant differences between the simulated and historical rainfall were noted using normal statistical tests.
Services: Buy this book/Buy this article
Return to search