The high solubility and low volatility of MTBE makes it difficult to treat when present at low levels (as in most groundwater supplies) using established technologies such as air stripping and activated carbon. Unlike the established transfer technologies (e.g., air stripping, activated carbon adsorption), advanced oxidation by Ultraviolet (UV) light destroys MTBE and its byproducts. As part of this study, extensive pilot and bench scale evaluation was performed to understand the effectiveness of MTBE and its byproducts [e.g., tertiary butyl alcohol (TBA) and tertiary butyl formate (TBF)] removal using UV/peroxide in ground and surface waters. The water quality matrix evaluated the impact of MTBE oxidation in the presence of other UV light absorbing compounds (e.g., nitrate) and radical scavengers (e.g., organics). These UV oxidation tests were conducted using both the medium-pressure (MP) and low-pressure, high- output (LPHO) lamps. The electrical energy inputted for MP and LPHO lamp operation during testing was measured. To minimize the impact of reactor hydraulics on the oxidation process, the pilot tests were conducted using closed-loop reactors at recirculation rates in excess of 650 gallons per minute (or 2,500 liters per minute). The effectiveness of MP versus LPHO lamps was compared using electrical energy per order (EEO), defined as the kiloWatt-hours (kWh) of electrical energy required to reduce the concentration of MTBE by an order of magnitude in 1,000 liters (or 1 m3) of water. Water quality impacts were also summarized using the energy requirement calculations. Based on the information gleaned from pilot testing, optimal design and operational criteria were developed. Installation and operational costs for full-scale systems of varying sizes were developed using the optimal design and operational criteria. Includes 6 references, tables, figures.