In this study, an interval parameter fuzzy relation analysis (IPFRA) model is proposed for environmental risk assessment of petroleum-contaminated aquifers due to leakage from underground storage tanks. The model can effectively incorporate effects of different pollutants and different remediation techniques within a general framework. Also, it can directly reflect uncertainties presented as inexact intervals for a number of modeling inputs. Results of a case study indicate that reasonable solutions for risk assessment under different system conditions have been generated. When no remediation action is undertaken, environmental conditions at the study site may be ''highly risky'' in terms of their relation to drinking water safety. Four potential site remediation strategies are analyzed. They have different environmental and economic characteristics, with lower risks generally correspond ing to higher costs. Trade-offs between environmental and economic objectives are then analyzed. In general, the IPFRA approach is useful for comprehensively evaluating risks within a system containing many factors with complicated interrela tionships. The modeling results have provided bases for determining desirable site remediation actions.