The physically based hydrological models require the estimation of various model parameters through calibration. Several past studies that focused on parameter estimation of hydrological models have found that no single objective performance criterion is adequate for matching different essential characteristics of the observation data. Since physically based hydrological models simulate many of the catchment hydrological processes, it needs to define multiple performance criteria to effectively use the information from various datasets and application of multiobjective optimization for attaining Pareto optimal solutions. In the present study, a Multiobjective Self-adaptive Differential Evolution algorithm (MOSaDE) is applied to perform multiobjective calibration of hydrological models. MOSaDE is an advancement of well-known Differential Evolution (DE) algorithm, using the notion of Pareto dominance, fast nondominated sorting approach, diversity preservation using crowding distance and elitist strategy of joining parent and offspring population. The parameter self-adaptation strategy in the MOSaDE also increases the robustness of the algorithm and alleviate the needs of computationally demanding sensitivity analysis of the algorithm parameters. The methodology is verified for calibration of Variable Infiltration Capacity (VIC) model, which is a popular physically based hydrological model, for a case study in Krishna basin, in India and the results are found to be promising.