The Ensemble Kalman Filter for Groundwater Plume Characterization: A Case Study.

Area of Science:

• Environmental Science
• Hydrogeology
• Data Assimilation

Background:

• The Kalman filter is a powerful tool for refining state variable estimates using measured data and spatiotemporal correlations.
• The ensemble Kalman filter (EnKF) utilizes Monte Carlo analysis to define correlations for state refinement, with successful applications in engineering fields like reservoir modeling and weather forecasting.
• EnKF application in hydrology remains limited, despite its potential for complex environmental modeling.

Purpose of the Study:

• To apply the ensemble Kalman filter (EnKF) methodology to refine a simulated groundwater tetrachloroethylene (TCE) plume at the Tooele Army Depot-North (TEAD-N) site in Utah.
• To simulate the EnKF-assimilated plume forward in time to predict future plume migration.
• To compare the efficacy of EnKF with ordinary kriging for water quality data assimilation in groundwater modeling.

Main Methods:

• Employed the ensemble Kalman filter (EnKF) to assimilate observed TCE concentrations into a groundwater model.
• Utilized Monte Carlo analysis within EnKF to define spatial and temporal correlations based on a calibrated groundwater flow and transport model.
• Performed forward simulations of the assimilated plume to predict future migration patterns.
• Compared EnKF performance against an ordinary kriging-based assimilation method.

Main Results:

• The EnKF successfully refined the simulated groundwater TCE plume using observed aquifer data.
• The EnKF-based assimilation implicitly captured complex site hydrology and variable source history through its correlation structure.
• The study provides a basis for evaluating the relative accuracy of EnKF versus ordinary kriging in representing plume concentrations.

Conclusions:

• The ensemble Kalman filter (EnKF) demonstrates potential for effective data assimilation in groundwater contamination studies.
• EnKF's ability to incorporate spatiotemporal correlations offers advantages for predicting contaminant plume migration.
• Further investigation into EnKF's efficacy compared to other methods is warranted for optimizing water quality data assimilation in hydrology.