A Pareto-Based Sensitivity Analysis and Multiobjective Calibration Approach for Integrating Streamflow and Evaporation Data

TL;DR

This paper introduces a Pareto-based framework for sensitivity analysis and multiobjective calibration of hydrologic models, integrating streamflow and evaporation data to improve model realism and understand trade-offs.

Contribution

It presents a novel Pareto optimization approach that simultaneously considers streamflow and evaporation in sensitivity analysis and calibration, enhancing hydrologic model accuracy.

Findings

Vegetation parameters significantly influence evaporation performance.

Soil and routing parameters mainly affect streamflow sensitivity.

Joint calibration improves model performance for both variables.

Abstract

Evaporation is gaining increasing attention as a calibration and evaluation variable in hydrologic studies that seek to improve the physical realism of hydrologic models and go beyond the long-established streamflow-only calibration. However, this trend is not yet reflected in sensitivity analyses aimed at determining the relevant parameters to calibrate, where streamflow has traditionally played a leading role. On the basis of a Pareto optimization approach, we propose a framework to integrate the temporal dynamics of streamflow and evaporation into the sensitivity analysis and calibration stages of the hydrologic modeling exercise, here referred to as Pareto-based sensitivity analysis and multiobjective calibration. The framework is successfully applied to a case study using the Variable Infiltration Capacity (VIC) model in three catchments located in Spain as representative of the different hydroclimatic conditions within the Iberian Peninsula. Several VIC vegetation parameters were identified as important to the performance estimates for evaporation during sensitivity analysis, and therefore were suitable candidates to improve the model representation of evaporative fluxes. Sensitivities for the streamflow performance, in turn, were mostly driven by the soil and routing parameters, with little contribution from the vegetation parameters. The multiobjective calibration experiments were carried out for the most parsimonious parameterization after a comparative analysis of the performance gains and losses for streamflow and evaporation, and yielded optimal adjustments for both hydrologic variables simultaneously. Results from this study will help to develop a better understanding of the trade-offs resulting from the joint integration of streamflow and evaporation data into modeling frameworks.