One-day offset in daily hydrologic modeling: An exploration of the issue in automatic model calibration

Abstract

Summary Hydrologic modeling literature illustrates that daily simulation models are incapable of accurately representing hydrograph timing due to relationships between precipitation and watershed hydrologic response that happen with a sub-daily time step in the real world. For watersheds with a time of concentration less than 24 h and a late day precipitation event, the observed hydrographic response frequently occurs one day after the precipitation peak while the model simulates a same day event. The analysis of sub-daily precipitation and runoff in this study suggests that, this one-day offset is inevitable in daily analysis of the precipitation–runoff relationship when the same 24-h time interval, e.g. the calendar day, is used to prepare daily precipitation and runoff datasets. Under these conditions, daily simulation models will fail to emulate this one-day offset issue (1dOI) and result in significant daily residuals between simulated and measured hydrographs. Results of this study show that the automatic calibration of such daily models will be misled by model performance metrics that are based on the aggregation of daily residuals to a solution that systematically underestimate the peak flow rates while trying to emulate the one-day lags. In this study, a novel algorithm called S hifting H ydrograph I n order to F ix T iming (SHIFT) is developed to reduce the impact of this one-day offset issue (1dOI) on the parameter estimation of daily simulation models. Results show that with SHIFT the aforementioned automatic calibration finds a solution that accurately estimates the magnitude of daily peak flow rates and the shape of the rising and falling limbs of the daily hydrograph. Moreover, it is shown that this daily calibrated model performs quite well with an alternative daily precipitation dataset that has a minimal number of 1dOIs, concluding that SHIFT can minimize the impact of 1dOI on parameter estimation of daily simulation models.