<p>The relationship between environmental and economic drivers in mine dewatering is complex, with many competing interests. Traditionally, numerical groundwater models have been used to assess the efficacy of dewatering designs, both with respect to the expected dewatering rates and to estimate the potential effects on sensitive groundwater receptors. However, the outputs of interest from the groundwater models depend on uncertain model inputs, which means these outputs are also subject to uncertainty. Herein, we quantitatively explore the tradeoff between dewatering costs and environmental effects within the framework of reliability-based multi-objective optimization using a synthetic case study designed to mimic many facets of real-world mine dewatering. The framework explicitly considers model input uncertainty and seeks to map the trade-offs between cost, environmental impact, and reliability. We also explore the cost implications of uncertainty and hydrogeologic data collection. The results demonstrate that formal management optimization outperforms a standard dewatering strategy in both environmental and economic outcomes, allowing for simultaneous improvement in both objectives. When widely recognized hydrogeologic uncertainties are explicitly included in the optimization, the economic and/or environmental ``cost'' of reliability is quantifiable, and, through calibration to pre-mining hydrogeologic data, the tradeoffs between reliability, economic costs and environmental outcomes are improved compared to the uncalibrated analysis. However, high levels of reliability come with substantial economic and/or environmental costs, and a highly reliable zero-impact environmental outcome is not feasible. Ultimately, we show that formal reliability-based management optimization enables decision makers to choose their level of acceptable risk and to understand the economic and environmental costs associated with this choice.</p>