Enhanced Early Warning of Extreme River Discharge Events in the Yangtze River Basin Using Atmospheric Circulation Signals

Published: 9 March 2026| Version 1 | DOI: 10.17632/s78pj87p29.1
Contributor:
xiaoke xu

Description

We develop DetRF, a machine learning model that integrates anomaly detection with a Balanced Random Forest ensemble. The model was trained and validated using ERA5 reanalysis data from 2000–2019 and independently tested with ERA5 data from 2020–2024. Its early warning performance was assessed using precipitation and runoff observations.

Files

Categories

Hydrology, Extreme Event, Atmospheric Circulation

Licence