Construction of rating curve at high water level considering rainfall effect in a tidal river

•Water level is decomposed into four components (tide, wave, runoff, and noise).•Stage-discharge rating curve is developed to estimate the runoff component.•Water level is estimated by integrating all components in a tidal river. South Korea (KR), Ulsan tidal river. Climate change and urbanization pose a serious threat to rising water levels and flooding in the estuary region. In particular, tidal river, which is located in the estuary area, is impacted by many hydrological factors such as tide, wave height, and rainfall. This paper develops a new rating curve that can predict the water level of the tidal river. It also assesses the accuracy of the developed rating curve through the verification event. Tidal rivers show non-linear hydrological characteristics that are influenced by various hydrological factors. Thus, no rating curve has been developed for them. This study proposes a method to develop a rating curve that can estimate the water level by using the runoff from rainfall in a tidal river. To develop the rating curve, the water level at Ulsan station, Korea was decomposed into four components, which are tide, wave, noise and runoff component by using wavelet analysis, curve fitting, and a high pass filter. A distributed grid-based rainfall-runoff model (GRM) was applied to simulate the runoff. The simulated runoff and the water level components influenced by rainfall were used to develop a rating curve and estimate the water level from the simulated runoff. The results were then added to the other three components (tide, wave, noise) to calculate an integrated water level, which was compared with the observed water level at Ulsan station. With less than 10 % error during validation, the results are expected to contribute to the accurate prediction of water level of tidal river during flood season.