Groundwater contaminant boundary input flux estimation in a two-dimensional aquifer

The present study considers the estimation of groundwater input contaminant flux and its transient distribution in a hypothetical two-dimensional aquifer. The paper proposes Kalman filter with recursive least square algorithm (RLSA) technique for the inverse estimation problem. Numerical simulation results reveal that the proposed estimator has outstanding estimation performance both for smooth and abruptly varying input flux. The study further takes into account the dependence of the RLSA technique on the process noise. Selected input scenarios are examined to count for the effect of measurement sensor arrangement and RLSA shows more improved results when sensors number is increased and they are placed closer to the supposed input location.