Estimating spatially-variable first-order rate constants in groundwater reactive transport systems

Area of Science:

• Environmental science
• Hydrogeology
• Geochemistry

Background:

• Numerical reactive transport models are crucial for assessing contaminated aquifers and evaluating mitigation strategies.
• Accurate simulation of solute fate and transport is often limited by incomplete knowledge of chemical reaction parameters.
• Parameter uncertainty, particularly for reaction rates, significantly impacts model reliability.

Purpose of the Study:

• To improve estimates of spatially variable first-order rate constants (λ) in reactive transport models.
• To assess the effectiveness of the Ensemble Kalman Filter (EnKF) for parameter estimation using concentration data.
• To investigate the impact of various uncertainty sources on parameter estimation.

Main Methods:

• Employed a steady-state Ensemble Kalman Filter (EnKF), a data assimilation algorithm.
• Integrated solute concentration measurements into reactive transport simulations to update parameter estimates.
• Utilized a synthetic aquifer system with first-order decay and investigated uncertainties in hydraulic conductivity and spatial parameter structure.

Main Results:

• The EnKF successfully conditioned the ensemble of λ values to match the reference λ field.
• Parameter estimation accuracy was sensitive to the number and location of assimilated concentration measurements.
• Measurement error and the correlation length of the λ fields also influenced the reliability of the estimated parameters.

Conclusions:

• Data assimilation using EnKF is an effective method for improving estimates of reaction rate parameters in reactive transport models.
• The study highlights the importance of measurement strategy (quantity, location) and understanding parameter field characteristics for accurate modeling.
• This approach enhances the predictive capability of models for contaminated aquifer assessment and remediation planning.