Since 2000, a historic drought has threatened water supply for the Colorado River Basin (CRB), without a regulatory framework that could adequately protect supply. Interim policies for using delivery reductions and conservation to help protect water supply were enacted starting in 2007, but they are set to expire in 2026 in the context of continued dry conditions. Recently, high profile negotiations have been held that seek new policies, implemented by the Bureau of Reclamation. These new policies will have to exist under deeply uncertain conditions of hydrology and demand, where deep uncertainty refers to conditions where stakeholders do not know or cannot agree on the likelihood of particular scenarios. Existing methods address deep uncertainty by using large ensembles of hydrologic scenarios, calculating policies' robustness with statistical metrics such as the percentage of scenarios where performance degradation occurs. This presentation demonstrates a method where a self-organizing map (SOM) is used to systematically organize and visually represent the range of plausible future scenarios. The SOM arranges complex scenarios of hydrology, demand, and initial system conditions into a two-dimensional visual exploration of the uncertainty space. Representative potential post-2026 CRB policies are generated using simulation-based optimization, and policies' performance is overlaid on the SOM, providing novel visual insights into how policy performance fluctuates under varying scenarios. The method provides an intuitive linkage between hydrologic conditions and projected policy performance, better enabling conversations with stakeholders on how the new policies will perform in a changing future.