*3.5. Model Calibration and Validation*

The water balance of the basin was estimated based on climatic data analysis from 1980–2016. The base year is considered from January to December 2016, which is taken as a normal hydrological year. The results showed that the total water balance of the basin during the study period was found to be 427 MCM. The values of parameters for the calibration results in the WEAP model is shown (Table 4).


**Table 4.** The calibration parameters' possible ranges used in the WEAP model.

The WEAP model was calibrated and validated with monthly observed streamflow from 1980–1999 and 2000–2014, respectively (Figure 7). In order to simulate the streamflow values in the model, the crop coefficient of (kc) of different land covers, soil water capacity, root zone conductivity, and preferred flow direction were manually calibrated using the default values of the WEAP model (Table 4). Hence, during the calibration period of this study, the values of the coefficient of determination (R2) ranged from 0.73 to 0.91, with an average value of 0.82. On the other hand, the value of the Nash–Sutcliffe efficiency (NSE) was 0.65 and 0.83, the minimum and maximum values, respectively, with an average value of 0.74. However, during validation periods, 0.54 and 0.91 are the minimum and maximum values of the coefficient of determination (R2), with an average value of 0.73. As far as the values of the Nash–Sutcliffe efficiency (NSE), 0.50 and 0.84 were the minimum and maximum values respectively, with an average value of 0.67. From these values, we have observed that in both calibration and validation periods, the values correspond to a perfect match between the observed and the modeled streamflow values. Thus, the performance of the model is acceptable between the trends of observed and simulated streamflow in both calibration and validation of the study basin. This also helps to accurately project the prediction of future discharges based on the future scenario set. The statistical values of Akaki (R2 = 0.73; NSE = 0.65) were less than other gauging stations; Kuntre (R2 = 0.85; NSE = 0.73), Hombole (R2 = 0.89; NSE = 0.83), and Modjo (R2 = 0.91; NSE = 0.85) (Table 5). This may be due to urbanization.


**Table 5.** Statistical values of stations for calibration and validation.

**Figure 7.** Calibration and validation of the four gauging stations and the entire Awash River Basin.
