Simulating hydrological alterations in the conditions of limited data: lessons from the Walawe River Basin in Sri Lanka

The paper examines the performance of two modeling methods of varying complexity in the conditions of limited observed data, using the example of Rakwana Oya catchment a tributary of Walawe river basin in the Southern Sri Lanka. The first method is a distributed physically based SWAT model. The second is the non-linear spatial interpolation method which deals exclusively with observed records. The SWAT model was calibrated and subsequently applied to assess the impacts of land use change on river flows with different land use scenarios for years 1956, 1985 and 1999. While the model is useful for assessing such impacts on flow, its intensive data requirements suggest that its application to bigger catchment (e.g. the entire Walawe basin) is not feasible. It is shown that less data intensive, spatial interpolation method can also perform satisfactorily and may be used to simulate non-regulated river flows in small and big basins.