Simulation of NAPL Remediation Process Using Parallel Simulator PARSSIM

by
Xiaohong Wang, MSE

University of Texas at Austin, 1999
Supervisors: Mary F. Wheeler
Steven Bryant

The objective of the work is to investigate the kinetic NAPL dissolution process using parallel simulator Parssim which is developed by the Center for Subsurface Modeling (CSM) of the Texas Institute for Computational and Applied Mathematics (TICAM) at The University of Texas at Austin. The effects of the number of gridblock in the given field, dispersion of the NAPL, well patterns, well injected rate, field heterogeneity, the kinetic reaction rate and biodegradation on the kinetic NAPL dissolution process have been investigated. According to the simulation results, for the given contaminant site, the grid-refinement does effect the simulation results, especially the concentration tail. The computational time increase significantly along with the increase in the number of gridblocks in the given field. The increase in the transverse coefficients of NAPL can speed up the NAPL remediation efficiency and the change in the longitudinal coefficient of NAPL has little effect on the process. Well pattern for the given problem impacts considerably on the remediation process. A good well pattern can prevent the area nearby the contaminant source from contaminated during the field remediation. A good injecting rate for a remediation process can save both remediation time and water quantity. From simulation results, the different realizations with the same statistical parameters give similar results. When we increase the standard deviation log K or the correlation length of the field, the NAPL concentration at the producer for both cases increases at the early stage of the remediation process. The NAPL concentration at the producer when considering the kinetic dissolution will shift away the concentration for the equilibrium dissolution process if the NAPL kinetic dissolution rate constant in the field decreases. Small NAPL concentration at the producer comparing with its solubility at the very early stage of the process may be the signal that the NAPL has very slow dissolution rate. The NAPL dissolution rate may be one of the key factors for cleaning up the contaminant site successfully. A very long tail would be expected for a slow NAPL dissolution rate. During the toluene bioremediation process, the injecting water saturated with oxygen and nitrogen can reduced the toluene concentration at the producer through biodegradation. Also the biodegradation rate does not seem to impose some significant effects on the process efficiency for the cases studied in this work, because the kinetic toluene dissolution process controls the whole remediation process. This implies if we turn off the biodegradation reaction when the toluene concentration at the producer is below the government standard during the remediation process, the toluene concentration at the producer may go up later since there is still some substantial residual toluene existed in the field. This result may warn us that during the bioremediation process, we have to monitor the NAPL concentration at the producer for a period of time after turning off the bioremediation to make sure that the NAPL concentration is still below that required by the government.

Back to theses index