<p>Storm transposition is a cornerstone of probable maximum precipitation (PMP) estimation. Yet, according to a recent report by the National Academies of Sciences, Engineering, and Medicine, “a solid scientific foundation for storm transposition is not available”. In this article, a new approach to storm transposition is presented and applied to a mesoscale convective system (MCS) in the Midwestern U.S. Steering of the storm toward the target basin is achieved by leveraging internal variability—that is, the intrinsic spread in the outcomes of a numerical weather prediction model, arising from the nonlinear nature of the atmospheric system and its sensitivity to initial and boundary conditions. More precisely, an ensemble of realizations of the MCS was generated by incrementally advancing the simulation start time and expanding the domain boundaries, allowing the WRF model to explore a broad range of plausible storm evolutions and trajectories. This method not only provides a physically consistent basis for determining the transposition region—offering an objective alternative to traditional approaches that rely heavily on subjective meteorological judgment—but also substantially reduces the magnitude of initial and boundary condition shifts required to position the storm over the target (here, the Raccoon River Watershed in Iowa), thereby limiting the impact of such shifts on the physical consistency of the simulated fields.</p>