Computer simulations tested the ability of gravel interceptor trenches to capture a plume of contaminated groundwater. The plume had a maximum length and width of 87 and 19 m, respectively. In alternate simulations, one-meter wide trenches were located 5, 10, 20, and 50 m downgradient of the plume. A minimum trench length and time required to capture the plume was determined for each location (setback). The plume was considered captured if it passed entirely through a trench. A 21 m-long trench captured the plume at setbacks of 5 and 10 m. Minimum trench length increased to 23 and 25 m at setbacks of 20 and 50 m, respectively. Increased contaminant spreading with distance traveled dictated longer trenches at larger setbacks. Results of this study suggest that, at settings where contaminant plumes are carefully monitored and spatially defined, passive interceptor trenches should be close to a plume's leading tip and slightly longer than the maximum width of the plume.