Implication of Uncertainty in River Gauged Data and the Rating Curve Representations to Flood Quantiles: Case Studies from Stations Across Peninsular Malaysia

River discharge data are important in hydrological modeling and flood frequency analysis (FFA), e.g., in model tuning and determination of flood quantiles. A time series of river discharge is commonly generated from the application of a rating curve of gauged water level and discharge at a specific river cross-section. The curve fit using the data points inevitably comes with errors because of the imperfect knowledge on the stage and velocity measurements during individual gauge, for example, due to scour/fill, vegetation, and backwater effects. The errors  challenge the reliability of a single rating curve over the governing dataset. This study aims to analyze how uncertainty in the rating curve generated from a complete record of gauged data can provide additional information in flood frequency studies when consider error bounds. A methodology was developed to tailor the aim of the study to the available data from stations across the Peninsular Malaysia. All river gauging stations in Peninsular Malaysia under the Department of Irrigation and Drainage (DID) databank were first examined and finally selected based on a set criteria (e.g., length of data availability and missing data). Uncertainty in the rating curve was represented by error distribution with upper and lower limit curves. A  time series of discharges was then generated using the rating curves. Results show that there are 15 gauging stations from the overall 103 across Peninsular Malaysia complies with the conditions set for this study. The findings suggest that explicit consideration of rating curve uncertainty in streamflow data is potentially useful in hydrological modeling and flood frequency studies to manage long-term water security.