Abstract A comprehensive risk-based monitoring assessment methodology is developed to (i) incorporate background data for thresholds and monitoring design, (ii) use stochastic leakage simulations and monitoring modeling for risk scenarios given uncertainty, (iii) apply statistical methods to combine multiple monitoring techniques and provide decision support for evaluating proposed monitoring plans, and (iv) facilitate implementation in the National Risk Assessment Partnership (NRAP) integrated assessment model (IAM) framework. This methodology is illustrated using a case representative of the High Plains aquifer, considering the stochastic leakage events simulated using reactive transport simulations. The resulting groundwater quality changes were reflected in three groundwater monitoring parameters (pH, TDS and benzene concentrations), which were used to calculate the corresponding detection probability for each simulation, based on the background distributions and the selected thresholds. The consequent detection probability was then used to evaluate the monitoring well density and the proposed network designs. Beyond the detectability, the earliest response time and spatial coverage constraints of a monitoring network design are considered, and the role of adaptive monitoring strategies for compliance monitoring and incorporation of expert judgment on likely leakage locations is considered.