Data-intelligence-driven flood forecasting and early warning in a small river basin: A BiLSTM model integrated with the “631” response mechanism
Peisheng Yang, Xiaohua Xu, Meilan Shao, Yewei Liu, Longhui Zhu · 2026
Read Data-intelligence-driven flood forecasting and early warning in a small river basin: A BiLSTM model integrated with the “631” response mechanism by Peisheng Yang, Xiaohua Xu, Meilan Shao, Yewei Liu, Longhui Zhu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0357040.txt.
Description
Peisheng Yang and co-authors paired a deep learning flood model with Jiangxi Province's staged warning system. The work focused on the Xiangshui River Basin in China. Floods gather fast in this small mountainous catchment. Their 2026 study trained a bidirectional memory network on twenty past flood events. It tested the network on six held-back events, against three established approaches.
The new model forecast flood flows more accurately than all three rivals. It was strongest one hour ahead, though its skill faded by six hours out. Rainfall from upstream stations in the previous two hours mattered most. The authors map the model's output onto the province's six-three-one hour response stages. They use weather forecasts for the longer leads. That gives local officials a clearer chain from prediction to evacuation orders.
Catalog record
| CREATOR | Peisheng Yang, Xiaohua Xu, Meilan Shao, Yewei Liu, Longhui Zhu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37834 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |