The evaluation of hydrologic model behavior is made through comparison of simulated and observed variables, and efficiency criteria (EC) are used to provide an objective assessment of such closeness. In this study, the utility of two different EC, NSE (Nash-Sutcliffe Efficiency) and KGE (Kling Gupta Efficiency), is investigated. In the study area of Celone (Italy) an updating of the Krause et al. (2005) method is proposed by adding the KGE and by modifying the length of the dry season. In the study area of Oued Chiba (Tunisia) the model SWAT (Soil and Water Assessment Tools) is applied to assess the water balance, and SWAT Cup to calibrate the outputs using both NSE and KGE. Results show that KGE is the best EC in evaluating the performance of the model and NSEmod can be considered as an intermediate solution. Moreover, there is no impact of the variation of the basin‘s characteristics on the behavior of the EC.
