SMA Based Continuous Hydrologic Simulation Of The Blue Nile

Continuous hydrologic models account for the soil moisture balance in the catchment over a long-term period. Different hydrologic physical processes such as: interception, surface depression storage, infiltration, soil storage, percolation, and groundwater storage should be considered in continuous hydrologic modeling. In this paper, the Hydrologic Modeling System (HMS) model was used for continuous hydrologic simulation of the Blue Nile. The model encompasses Soil Moisture Accounting (SMA) algorithm to simulate the long-term relationship between rainfall, runoff, storage, evapotranspiration, and soil losses in the Blue Nile. The objective of this paper is to evaluate the performance and potentiality of the HMS with the SMA algorithm on the Blue Nile as a case study in the Nile basin. Daily observed rainfall and runoff records of seven years (1990-1996) and the Digital Elevation Model (DEM) of the Blue Nile were made available for the calibration and verification stages. The obtained results were reasonable and acceptable taking into consideration the lumped time-invariant parameters used in the calibration and verification of the model. Despite the lumping of the input, the output and the parameters accounting for the physical processes in the catchment, the model showed satisfactory performance by accounting for more than 90% of the initial variance. More rainfall-runoff data, seasonal parametrization and modeling the Blue Nile watershed in sub-basins of smaller areas may improve the results. Introduction Long-term rainfall-runoff simulation requires modeling of different hydrological processes in the watershed based on water balances during a long-term period. Continuous hydrologic computer modeling may be a tool to study the hydrologic regime of a study area. However, the seasonality of the hydrological processes may be one of the main challenges facing continuous hydrologic simulation and is a main source of error in the simulation. Average dvalue (lumped) parameters during the simulated period may be unrealistic. A seasonal parameterization approach, where separate parameters sets are established for each season, may enhance the results, yet, more data and initial conditio ns settings will be required.