A Stochastic Weather Generator Based Framework for Generating Ensemble Sub-Monthly Precipitation for Streamflow Prediction

Abstract

Reliable sub-monthly streamflow prediction provides valuable information for disaster warning. This study proposes a sub-monthly streamflow prediction framework for organically combining stochastic weather generator (SWG) and monthly precipitation prediction to generate ensemble sub-monthly precipitation for streamflow prediction. To perform the framework, parameters that characterize precipitation occurrence and amounts are optionally adjusted based on the monthly precipitation prediction through three alternative schemes for illustration. The SWG-based schemes are then compared with the common numerical hydrometeorology ensemble streamflow prediction over two river basins in China (Xiangjiang and Hanjiang river basins). Results show that the numerical streamflow predictions exhibit a less accurate deterministic performance than the SWG-based framework with the climatology scheme over sub-monthly horizon, with leadtime-averaged mean absolute relative error dropping from 19.9% to 8.3%, and 21.8% to 11.1% for Xiangjiang and Hanjiang river basins, respectively.  In terms of the probabilistic prediction performance, the SWG-based methods yield approximately equivalent results with the numerical hydrometeorology streamflow prediction, with leadtime-averaged Continuous Ranked Probability Skill Score of the scheme with modified parameters of precipitation amounts being 0.51 and 0.56 for Xiangjiang and Hanjiang River basins, respectively. Furthermore, the SWG-based schemes are more suitable for predicting high-flow events during the flood season, with a significantly smaller Brier Score.  Overall, the proposed SWG-based framework offers a comparatively inexpensive and accessible way to achieve promising predictive skills for sub-monthly streamflow.

Keywords: stochastic weather generator, sub-monthly precipitation prediction, ensemble streamflow prediction, prediction framework