Seite - 601 - in Book of Full Papers - Symposium Hydro Engineering

ACKNOWLEGEMENTS Most of the data used for this study were obtained from World Bank funded Climate Resilience Improvement Project (CRIP) under the Ministry of Irrigation and Water Resources Management. Therefore, the author would like to convey his sincere gratitude to Eng. (Mrs) P.A.A.P.K. Pannala, Deputy Project Director of component I of CRIP for giving the permission to use the data. REFERENCES [1] Long Term Generation Expansion Plan 2015-2034, Ceylon Electricity Board, Sri Lanka, September, 2016 [2] NEITCH S.L. et al. Soil and water assessment tool, Theoretical documentation, version 2009. Texas water resources institute, 2011 [3] DAHMEN E.R., HALL M.J. Screening of hydrological data: Tests for stationarity and relative consistency, ILRI, The Netherlands, 1990 [4] http://swat.tamu.edu SUMMARY The Mahaweli River Upper Basin (MRUB) in Sri Lanka consists of four major hydro reservoirs namely, Kothmale, Victoria, Randenigala and Rantambe situated in a cascade system. The total storage capacity and power generation capacity of these reservoirs are 1,756.3 million m3 and 582 MW respectively. The catchment above Victoria reservoir was the area selected for this study (1,865.52 km2). As 39.9% of this area consists of slopes above 30% aforementioned reservoirs are facing the risk of sedimentation due to catchment soil erosion. Eeventual loss of energy production potential when the storage capacities are reduced would be detrimental as the country heavily depends on hydropower. In addition, the reduction in reservoir capacity may create water scarcity for other users when the demand is high. Further, the loss in capacity of these reservoirs increases the probability of floods as Mahaweli river basin downstream areas are prone to flooding during rainy seasons. Soil Water Assessment Tool (SWAT) developed by USDA Agricultural Research Service was used as the modelling software for this study. It is GIS- based and a model was developed for the study area at a daily time-step for the period 1960 to 2016. Main inputs to the model were digital elevation model, land- use map, soil map, rainfall and temperature. Unavailable climatic data; solar radiation, wind speed and relative humidity were generated by the model itself. 601