Methyl tertiary-butyl ether (MTBE) and its degraded form tertiary-butyl alcohol (TBA) are both known carcinogens that have contaminated groundwater aquifers across the United States. MTBE is a synthesized compound, once widely used as an additive in gasoline to increase oxygenation. Because of its popularity, MTBE was released into the environment primarily through fuel combustion and leaking underground storage tanks. These two compounds are known to be recalcitrant to most conventional physico-chemical treatment methods. Previous studies have suggested that bioremediation is effective at degrading MTBE and TBA in contaminated groundwater. Bioremediation involves the injection of oxygen, nutrients, and pre-adapted bacterial cultures into contaminated groundwater to increase the rate of natural biodegradation. In this study, a historically documented spill in Cambria, CA was modeled employing the Groundwater Modeling System software (GMS) to compare the effectiveness of the baseline treatment approach to that of in-situ bioremediation. MODFLOW was used to simulate groundwater flow, while MT3DMS was used to simulate dispersal and biodegradation of MTBE. Well data from public records was used as comparative values for hydraulic head and MTBE concentrations. Additional information from cleanup reports provided data for the physical properties of the aquifer. This included bedrock elevation, soil types, and storativity. Conductance, recharge rate, and hydraulic conductivity were calibrated using Parameter Estimation Software (PEST). The constants applied in MT3DMS simulations, such as dispersivity values, molecular diffusion coefficients, and retardation factors, were calculated manually using available, semi-empirical approaches. The model was first run emulating bioremediation using a high first order biodegradation rate estimated to be 8.6 day-1. This was compared to an instance of natural attenuation, with a first order biodegradation rate of 0.0074 day-1.
The case study investigated herein primarily implemented a pump and treat system relying on granular activated carbon and a series of trickling filters and clarifiers. Pump and treat operations began in 2000 and officially ended by the start of 2015. Even though treatment was terminated, the preliminary remedial goal for MTBE was not achieved. In the model created for this project, the bioremediation simulation predicted attainment of this treatment goal by 2010 after starting treatment in 2002. This increase in predicted removal rate over conventional approaches suggests bioremediation may be a viable and effective treatment technique when removing MTBE from groundwater. This predicted rate of removal suggests that bioremediation is more effective than the techniques used during the Cambria cleanup. It is important to note, there were many assumptions and simplifications made during the creation of the model. This includes the calibrated parameter values obtained from PEST iterations along with calculated parameter estimates regarding MTBE fate and transport. During set up, it was assumed that soil type consisted solely of silty clay and the bedrock layer was at a constant 45 ft below ground level. Additionally, the modeled in-situ bioremediation scenario assumes a best-case scenario, with the high first order biodegradation rate. For future modeling improvements, it is recommended to conduct onsite field testing to obtain degradation rates that more closely reflect rates found in the modeled region. A more complete mapping of the aquifer would also provide the model with increased reliability. Future models should also evaluate additional MTBE spill events and how differing terrains impact the effectiveness of in-situ bioremediation of MTBE.
| Identifer | oai:union.ndltd.org:CALPOLY/oai:digitalcommons.calpoly.edu:theses-4504 |
| Date | 01 June 2024 |
| Creators | Bundy, Logan |
| Publisher | DigitalCommons@CalPoly |
| Source Sets | California Polytechnic State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Master's Theses |
Page generated in 0.0029 seconds