Sensitivity Analysis in Hydrological Modeling

Sensitivity analysis has become instrumental in refining hydrological models by quantifying how uncertainty in input parameters propagates to model outputs. Both local and global approaches allow model developers to prioritise parameters for calibration, identify interactions and guide data collection. Local methods probe the response around a nominal parameter set, while global methods explore high-dimensional parameter spaces to capture non-linear and interaction effects. By integrating sensitivity analysis with model calibration workflows, practitioners can streamline parameter estimation, reduce uncertainty and improve predictive performance under diverse hydrological conditions. Emerging techniques for time-varying and spatially distributed sensitivity enable insights into dynamic processes, supporting flood forecasting, groundwater risk assessment and water-resource management at multiple scales. The adoption of surrogate models and advanced sampling schemes has further enhanced the feasibility of global sensitivity analyses for complex, computationally demanding hydrological models.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Sensitivity Analysis in Hydrological Modeling publication trend

The graph below shows the total number of articles in sensitivity analysis in hydrological modeling across all publications each year (not limited to Nature Index journals).

Technical terms

Sensitivity analysis: A suite of techniques to quantify how uncertainty in model inputs affects outputs.

Global sensitivity analysis (GSA): An approach evaluating parameter effects over their entire plausible ranges to capture non-linearities and interactions.

Morris method: A screening technique that estimates elementary effects to identify influential parameters with reduced computational cost.

Sobol’ indices: Variance-based metrics that decompose output variance into contributions from individual parameters and their interactions.

Surrogate model: A computationally efficient approximation of a complex model used to accelerate sensitivity and uncertainty analyses.

References

1. Ten strategies towards successful calibration of environmental models. Journal of Hydrology (2023).
2. Identification of parameter importance for benzene transport in the unsaturated zone using global sensitivity analysis. Hydrology and Earth System Sciences (2024).
3. Global Sensitivity Analysis of environmental models: Convergence and validation. Environmental Modelling & Software (2016).
4. From maps to movies: high-resolution time-varying sensitivity analysis for spatially distributed watershed models. Hydrology and Earth System Sciences (2013).