The increasing occurrences of global flood events, amidst climate change, highlight the need for hydrological data availability over large geographical domains for robust decision-making. Hydrological rating curves translate fluvial stage to streamflow and play a pivotal role in various applications, including flood inundation modeling and river geomorphology. Power law is an appropriate proxy for the nonlinear relationship between stage and discharge in natural systems. This study aims to develop a hierarchical data-driven approach to compute the power-law rating curve parameters (α, β) across the stream networks of CONtiguous United States (CONUS). The development of rating curve models is motivated by our interest in exploring a unifying solution linking rating curve parameters with hydro-climatological and geomorphological characteristics across CONUS. These can be applied to operational hydrological forecastins, such as the NOAA Office of Water Prediction NextGen framework, to enhance river routing and flood inundation mapping efforts. We used HYDRoacoustics in support of the Surface Water Oceanographic Topography (HYDRoSWOT), National Hydrography (NHDPlus v2.1), and STREAM-CATCHMENT (STREAMCAT) datasets for model development. Four empirical models—Multivariate regression, eXtreme Gradient boosting (XGBoost), Random Forest, Support Vector regression are compared. The first tier of models offers high accuracy but is limited to gauges, while the second-tier models offer a good compromise between accuracy and applicability across CONUS. We found XGBoost yielded R² of 0.67 and 0.55 for α, and 0.74 and 0.70 for β in the first and second-tier models. The spatial distribution of predicted α and β indicates sensitivity to elevation, aridity, and rainfall patterns.