| Contaminants in subsurface environment have been modeled using conventional methods to predict their behavior and transport in 3-dimensional (3-D) space. Typically, numerical models have been used in simulating the contaminant movement in groundwater. A 3-D contaminant transport model is numerically solved by approximation which plagued the model with truncation and round-off errors. In this research, to improve the accuracy of contaminant prediction spatially and temporally; and to assess the impact of first-order decay rate parameter estimation, three simulation filters, Kalman filter without parameter estimation, Kalman filter with parameter estimation and Kalman filter embedded with neural network were used in a specified 3-D domain space. The data assimilation filters are perturbed with random Gaussian noise to reflect real life case of contaminant movement. The filters are also guided by a set of sparse observation points at each time step to improve the accuracy of the prediction. The algorithms to generate the simulation results were run in Matlab 7.1. The effectiveness of the data assimilation filters and the numerical method were tested using Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Maximum Absolute Error (Emax) equations. Sensitivity analysis was also carried out to ascertain the effectiveness of the filters. The results show that the data assimilation filters perform better than the numerical method. Also, the filters are capable of reducing the error in the numerical solution by approximately 75%. |
