How good are hydrological models for gap-filling streamflow data?
Streamflow data underpin hydrological and climate change studies. Without such data, it is hard to understand catchment hydrological processes under climate change and non-stationarity. Unfortunately, continuous streamflow data are not always available and most gauges suffer from missing streamflow data. Often, the missing rate is important when selecting streamflow gauges, especially when the data is used for annual trend analysis.
Gap-filling streamflow data is a critical step for most hydrological studies, such as streamflow trend, flood and drought analysis and hydrological response variable estimates and predictions. Hydrological modelling is a useful method used in Australia for predicting daily streamflow in ungauged catchments. It has been used operationally by the Australian Bureau of Meteorology for filling daily streamflow data gaps. However, it is not clear how accurate the hydrological model gap-filled streamflow is.
For the first time, we used 217 unregulated Australian catchments to comprehensively evaluate the reliability and accuracy of the hydrological model gap-filled data that are influenced by different thresholds and by data missing patterns. We conducted three groups of experiments using two hydrological models (GR4J and SIMHYD) to test how the missing rates at 5%, 10% and 20% impact on streamflow trends. We found that when the missing rate is less than 10%, the gap-filled streamflow data obtained using calibrated hydrological models perform almost the same as the benchmark data (less than 1% missing) for estimating annual trends for the 217 catchments. Furthermore, the relative streamflow trend bias caused by the gap-filling is not very large in very dry catchments where the hydrological model calibration is normally poor. Our results clearly demonstrate that the gap-filling using hydrological modelling has little impact on the estimation of annual streamflow and its trends.
Zhang, Y., & Post, D. (2018). How good are hydrological models for gap-filling streamflow data? Hydrol. Earth Syst. Sci., 22, 4593-4604, https://doi.org/10.5194/hess-22-4593-2018.