In today's world, natural phenomena can be used to analyze and predict the events under study using the statistical modeling process. Many hydrological modeling methods do not make the best use of available information because hydrological models represent a wide range of environmental processes, which increases the complexity of the model. When making predictions, parameters clearly affect the performance of statistical models. The presence of uncertainty in many risk assessment problems in parameters leads to uncertainty in model predictions. General sensitivity analysis is a tool used to represent uncertainty and is useful in decision making, risk assessment, model simplification, etc. Two methods, Minkowski distance sensitivity analysis and regional sensitivity analysis, are methods that can work with a given sample set of input-output pairs of a model. A notable difference between these two methods is that Minkowski distance sensitivity analysis analyzes output distributions conditional on input values ​​(forward) while regional sensitivity analysis analyzes input distributions conditional on output values ​​(inverse). This paper will determine the relationships between general sensitivity methods (Minkowski distance and regional) and show that regional sensitivity analysis converges to Minkowski distance sensitivity analysis when focused on probability density functions. Similarly to forward sensitivity indices, inverse sensitivity indices can also be obtained. Finally, the sensitivity analysis of a water storage scheme is addressed using high-dimensional models with high-dimensional model outputs.