Integrated Riverine Flood Warning System Coupling River Stage Predictions with Hydraulic Modeling

This dataset provides time series files of potential covariates and daily river stage during the monsoon season (June–September), which were used to implement a Bayesian Hierarchical Network Model (BHNM) for ensemble forecasting of daily stage in the Brahmaputra River Basin, India. The potential covariates for 1- to 5-day lead time forecasts include: (i) daily stage from an upstream (feeder) gauge with 1-day (L₁) and 2-day (L₂) lags, and (ii) 1-day accumulated spatially averaged precipitation (P₁) from the contributing area between station gauges.The dataset includes a file with basic metadata (longitude, latitude, flood thresholds) for the gauges considered. Observed stage and warning thresholds were obtained from the Central Water Commission (CWC) and Bangladesh Water Development Board (BWDB), while gridded precipitation was sourced from the India Meteorology Department (IMD). These data were processed to generate the dataset presented here.