Electricity grids throughout the world face the dual challenge of decarbonizing while also withstanding increasing severe weather due to climate change. Grid operators must manage the integration of variable renewable energy sources and multi-sector electrification (i.e., increased demand) while also preparing for more frequent and severe droughts, heatwaves and cold snaps, which can adversely impact grid reliability. Consequently, simulation models of power systems, scalable for multiple hydrometeorological stresses, are critically needed to manage risks while reaching net zero emission goals. A persistent challenge for power system modelers has been the need to achieve adequate model spatiotemporal fidelity (e.g., time resolution and spatial scale) and computational cost (run time). This problem has been made more difficult to solve due to insufficient open-source data, often leading to oversimplification of necessary system components, which can bias the assessment of power system operations during extreme events. In this study, we are introducing a solution to this problem in the form of an open-source software that allows users to easily customize the temporal resolution with the desired spatial scale and track the accuracy of grid simulation models. We demonstrate our scalable approach for the U.S. Western Interconnection using existing synthetic grid datasets maintained by Texas A&M University. Our approach allows users to search over a wide range of model parameters (e.g., mathematical formulation, network topology, transmission line scalars, and economic hurdle rates) to find a model instantiation that accommodates their experimental needs. In this study, we demonstrate how this scalable approach supports the user in finding the best version of the model to examine specific natural hazards (e.g., heat waves, droughts, etc.). The grid models we developed are easily interoperable with observed balancing authority level data, as well as a range of multi-sector modeling platforms.