Panda, C and Panda, K C and Singh, R M and Singh, R and Singh, V P (2025) A generalised hydrological model for streamflow prediction using wavelet Ensembling. Journal of Hydrology (TSI), 655. ISSN 0022-1694
Full text not available from this repository. (Request a copy)Abstract
Machine learning (ML) models have recently been employed for precise streamflow prediction. These ML models, however, suffer from overfitting, non-scalability, non-transferability, and low predictive accuracy when used for unseen data that have high spatiotemporal variability and heterogeneity. To overcome these problems, a generalised streamflow model was, therefore, developed using novel Wavelet Ensembling (WE). The WE models were tested for several predictor combinations encompassing various terrain and flow conditions and were compared with the standalone and ensemble ML models. The generalisability of the models employed in the study was tested using transfer learning-based leave-one-out cross-validation (LOOCV). It was found that the WE models demonstrated a significant improvement over the standalone and ensemble ML models. The genetic algorithm (GA) optimised RF-based WE model (WE-RF-GA) was found to be the most efficient and generalised model, having the highest efficiency (NSE = 0.94, R2 = 0.95) and the lowest error (RMSE = 0.038, MAE = 0.028). The standalone and ensemble models showed higher predictive accuracy for tributaries, plain topography, and low-flow conditions compared to the mainstream flow, hilly terrain, and high-flow values. The WE models significantly reduced the performance gap of these models for the aforementioned conditions. This study would further enhance streamflow models and help predict streamflow effectively for data-scarce and ungauged basins.
| Item Type: | Article |
|---|---|
| Divisions: | Global Research Program - Resilient Farm and Food Systems |
| CRP: | UNSPECIFIED |
| Uncontrolled Keywords: | wavelet Ensembling, hydrological model, streamflow prediction |
| Subjects: | Others > Village Level Studies Others > Water Resources |
| Depositing User: | Mr Nagaraju T |
| Date Deposited: | 19 May 2025 04:29 |
| Last Modified: | 19 May 2025 04:29 |
| URI: | http://oar.icrisat.org/id/eprint/13077 |
| Official URL: | https://www.sciencedirect.com/science/article/abs/... |
| Projects: | UNSPECIFIED |
| Funders: | UNSPECIFIED |
| Acknowledgement: | UNSPECIFIED |
| Links: |
Actions (login required)
 |
View Item |
Altmetric
Altmetric