Pages: 56 to 64
Ghumman A. R., Yousry Ghazaw and M.S. Abulohom
Monthly runoff from a catchment is required for water resources engineering and management. This paper deals with runoff modeling from catchments which have scarce data of low quality. Two models namely mathematical conceptual model and Multivariate AutoRegressive – Moving Average model (MARMA) were developed and applied to two catchments (Brandu River and Khost River catchments) in Pakistan. The precipitation and meteorological data used were collected from Meteorological Department Baluchistan and North West Frontier Province, Pakistan. The results of mathematical conceptual model and multivariate autoregressive – moving average model were compared with each other. It was observed that multivariate autoregressive – moving average model could not provide the satisfactory results due to presence of high nonlinearity in the Rainfall-Runoff (R-R) processes of these watersheds. The performance of the conceptual model was acceptable.