Realistic daily discharge modelling in data-deficient regions using DL-assisted, parametrically-optimized hydrological model
Imee V. Necesito, Junhyeong Lee, Seonuk Baek, Yujin Kang, Duke Yu-Min Kim, Kyunghun Kim · 2025
Read Realistic daily discharge modelling in data-deficient regions using DL-assisted, parametrically-optimized hydrological model by Imee V. Necesito, Junhyeong Lee, Seonuk Baek, Yujin Kang, Duke Yu-Min Kim, Kyunghun Kim (2025) 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/pmc-12783632.txt.
Description
Imee Necesito and co-authors compare ways to model daily river flow where gauge records are thin. They took four sub-catchments in Samar, in the Philippines. Four models faced off: a classic event model, a deep learning network, a lean conceptual model called GR4J, and a hybrid that lets deep learning tune GR4J's parameters.
The plain GR4J ran low, while the deep network overshot peaks. The hybrid scored best, staying realistic on both floods and low flows, with strong fit scores across the basins. Pure deep learning followed trends but missed some peaks. Neither camp alone suffices, the authors argue; the blend wins where data are scarce.
Catalog record
| CREATOR | Imee V. Necesito, Junhyeong Lee, Seonuk Baek, Yujin Kang, Duke Yu-Min Kim, Kyunghun Kim |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |