ANALYSIS AND PARAMETER ESTIMATION OF VIRUS TRANSPORT THROUGH SUBSURFACE MEDIA

Abstract

In recent years there has been a tremendous increase in the contamination of groundwater due to rapid industrial growth and use of fertilizers and pesticides in agriculture. This contaminated water passes through the soil and may produce hazardous chemicals, which are risk to public health. The definition of contaminant is defined as the presence of any objectionable substance in water which make unsafe for drinking. The substance may be physical, chemical, biological or radiological. The biological contaminants are bacteria and virus. It is viruses in drinking water that are an important source of human enteric diseases. Pathogenic microorganisms from * sewage sludges, septic tanks and other sources can transport with subsurface water to drinking water wells. Production wells for drinking water must be at an adequate distance from source of contamination. Thus, there is a need to predict the contaminant distribution in ground water once these are released from the source. Understanding of the movementof contaminants in subsurface is necessary for taking up proper remedial measures. Numerical models are very important tools for studying the movement of contaminants in subsurface. The present study is concerned with the modeling of conservative as well as nonconservative virus transport in subsurface. The model is based on an operator split approach which employs a globally second order accurate explicit finite volume method for the advective transport and an implicit finite difference method for the dispersive transport. The performance of the numerical model in predicting solute/virus movement for both advection dominated and dispersion dominated flow scenario is studied by comparing the model prediction with the corresponding analytical solutions for a wide range of ii Peclet numbers and Courant numbers. The comparison is made for various cases of movement of conservative, reactive and virus transport in subsurface. In addition the virus transport numerical model is coupled with Richards equation governing moisture flow through the unsaturated zone. The numerical model simulating moisture flow through unsaturated zone is based on a mass conservative fully implicit finite difference numerical scheme. The application of the flow and transport models on virus movement through unsaturated zone is demonstrated through an example. The present study is also concerned with the estimation of transport parameters of virus movement in subsurface. The parameter estimation is formulated as a least square minimization problem in which the parameters are estimated by minimizing the deviation between the model predicted and observed virus concentrations. For this purpose, a hybrid finite volume numerical model simulating one dimensional virus transport in subsurface is coupled with Levenberg-Marquadart optimization algorithm. The efficacy and robustness of the optimization procedure is evaluated by estimating the parameter from hypothetically generated virus concentration data in both saturated and unsaturated zones. The present study also investigates the performance of the objective function while estimating transport parameters using inverse procedures in the presence of data errors. In this study the Gaussian noise is added to the hypothetical data generated at discrete times and at discrete distances from the source. A detailed statistical analysis is carried out to study the effect of bias induced by the objective function on the estimated parameters when the data contains the errors. The optimization algorithm is also applied to estimate the transport parameters from the virus concentration data of two column experiments involving MS2 and OX174 virus transport in saturated and unsaturated zones. in