The MuTHRE Model for High Quality Sub-seasonal Streamflow Forecasts

Abstract

Sub-seasonal streamflow forecasts, with lead times up to 30 days, can provide valuable information for water management, including reservoir operation to meet environmental flow, irrigation demands, and managing flood protection storage. A key aim is to produce “seamless” probabilistic forecasts, with high quality performance across the full range of lead times (1-30 days) and time scales (daily to monthly). This paper demonstrates that the Multi-Temporal Hydrological Residual Error (MuTHRE) model can address the challenge of “seamless” sub-seasonal forecasting. The MuTHRE model is designed to capture key features of hydrological errors, namely seasonality, dynamic biases due to hydrological non-stationarity, and extreme errors poorly represented by the common Gaussian distribution. The MuTHRE model is evaluated comprehensively over 11 catchments in the MurrayDarling Basin using multiple performance metrics, across a range of lead times, months and years, and at daily and monthly time scales. It is shown to provide “high” improvements, in terms of reliability for short lead times (up to 10 days), in dry months, and dry years. Forecast performance also improved in terms of sharpness. Importantly, improvements are consistent across multiple time scales (daily and monthly). This study highlights the benefits of modelling multiple temporal characteristics of hydrological errors, and demonstrates the power of the MuTHRE model for producing seamless sub-seasonal streamflow forecasts that can be utilized for a wide range of applications.