Estimating River Conductance from Prior Information to Improve Surface-Subsurface Model Calibration

Area of Science:

• Hydrogeology
• Environmental Modeling
• Computational Science

Background:

• Groundwater models simulate stream-aquifer interactions using river conductance (CRIV).
• CRIV calibration is challenging due to ill-posed inverse problems and lack of direct measurement.
• Existing CRIV estimation methods are often too simplistic or complex for practical use.

Purpose of the Study:

• To develop a flexible and operational tool for estimating river conductance (CRIV).
• To improve the prior estimation of CRIV by incorporating physical parameters, model grid size, and aquifer properties.
• To enhance the calibration and uncertainty analysis of surface-subsurface hydrological models.

Main Methods:

• A 2D numerical model in a local vertical cross-section is used to compute CRIV.
• The computation incorporates geometric and hydrodynamic parameters, regional model grid size, and aquifer hydraulic conductivity anisotropy.
• Global sensitivity analysis was performed to identify key parameters influencing CRIV.

Main Results:

• A novel method for calculating river conductance (CRIV) from physical parameters is presented.
• The tool accounts for regional model grid size and aquifer anisotropy, unlike previous methods.
• Sensitivity analysis confirmed the significant influence of selected parameters on CRIV.

Conclusions:

• The developed tool offers a more reliable and practical approach for estimating CRIV.
• This advancement supports better calibration and uncertainty quantification in hydrological models.
• Accurate CRIV estimation is crucial for objectives like conjunctive surface water-groundwater management.