Integrated hydrogeophysical modelling and data assimilation for geoelectrical leak detection

Area of Science:

• Geophysics
• Environmental Science
• Hydrology

Background:

• Time-lapse electrical resistivity tomography (ERT) offers high-resolution subsurface hydrological insights.
• ERT is utilized for detecting leaks and monitoring contaminant plumes but lacks direct leak parameter estimation.
• Current methods visualize plume evolution but do not quantify leak characteristics or uncertainties.

Purpose of the Study:

• To develop and demonstrate an ensemble-based data assimilation framework for estimating leak parameters from ERT data.
• To enable direct estimation of contaminant leak rate, location, and associated uncertainties.
• To integrate hydrological models with geophysical measurements for improved subsurface characterization.

Main Methods:

• Utilized an ensemble-based data assimilation framework evaluating hydrological models against time-lapse ERT data.
• Employed the PFLOTRAN-E4D code for parallel coupled hydro-geophysical simulations to generate synthetic ERT measurements.
• Applied an iterative ensemble smoother to update model proposals based on observed ERT data.

Main Results:

• Successfully identified contaminant source location, initial release time, and solute loading using synthetic and field ERT data.
• Quantified uncertainties associated with the estimated leak parameters.
• Demonstrated a reduction in site-wide uncertainty by comparing prior and posterior plume mass discharges.

Conclusions:

• The proposed framework effectively estimates contaminant leak parameters and their uncertainties from ERT data.
• This approach complements existing ERT imaging techniques, particularly for sites with prior geological investigations.
• The method provides a robust tool for subsurface contaminant source zone characterization and risk assessment.