Hydrological models are used for a variety of issues related to real time flood forecasting to water structures design and policies to prevent from damages of extreme hydrological events (i.e. floods and droughts). In these models, the values of some parameters can be estimated by observation and direct measurements; while the other
should be estimated through indirect techniques of model fitting using historical data. The main objective of this study was assess Continuous rainfall-runoff modeling using Soil Moisture Accounting (SMA) to the performance of different calibration time scales (annual, semi-annual, seasonal and monthly) in Zolachay watershed as part of the Urmia Lake basin in Iran. Continuous daily values (2003-2006 for calibration and 2007-2009 for validation phases) of rainfallrunoff data relevant to Zolachay watershed were used. Comparison between observed flow data and results provided by simulation was in good agreement. The model showed satisfied performance by accounting initial moisture condition by SMA model with Nash_Sutc_ liffe (NS) coefficient of 0.51 for calibration in monthly calibration scale. Performances of monthly validation scale with Nash-Sutcliffe values of 0.66 were generally superior to other scales. Optimized values of parameters related to monthly calibration time scale for estimating annual flow volume showed satisfactory results than the other scales; while for peak flows prediction, the seasonal calibration scale had accurate results than the others scales.
Key Words: Continuous Rainfall-Runoff, Calibration Scales, HMS-SMA, Modeling, Iran