A Dual-Branch TCN–TE Model for Multi-Horizon Runoff Forecasting Using Multi-Station Hydrological Observations
Sustainability · 25.05.2026 · 18 (11) · 5289
DOI: 10.3390/su18115289 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sustainability, online erschienen am 25.05.2026. Verfasst von Zhenzhu Meng und 6 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Zhejiang Provincial Natural Science Foundation, Nanxun Scholars Program for Young Scholars of ZJWEU, Huzhou Science and Technology Plan Project.
Abstract (Originalsprache)
Accurate runoff forecasting plays a critical role in sustainable flood risk management and climate-resilient water resources planning, particularly in plain river-network basins, where runoff processes are influenced by strong temporal variability and intensive human regulation. To address the limitations of existing data-driven models in representing short-term temporal variability and long-range dependencies, this study develops a dual-branch temporal convolutional network–transformer encoder (TCN–TE) forecasting framework for multi-station runoff prediction. The proposed model integrates a temporal convolutional network (TCN) with channel attention (CA) to extract local temporal patterns and adaptively reweight multivariate hydrological features, and a gated recurrent unit (GRU)-enhanced transformer encoder (TE) to improve long-range temporal dependency modeling. In addition, an autocorrelation-based analysis is conducted to quantitatively determine the effective memory length of the runoff system, providing a statistically grounded basis for input window selection. The model is evaluated using daily runoff and rainfall data from the Dongtiao River basin in eastern China, including seven runoff stations and two rainfall stations over the period 2012–2024. Forecasting results under multiple horizons (1, 3, 7, and 14 days) demonstrate that the proposed TCN-TE model consistently outperforms representative deep learning baselines in terms of R2, RMSE, and MAE, with particularly significant improvements for medium- and long-term forecasts. The results suggest that the proposed model provides a useful data-driven multivariate forecasting framework for runoff prediction using multi-station hydrological observations.
Förderung
- Zhejiang Provincial Natural Science Foundation
- Nanxun Scholars Program for Young Scholars of ZJWEU
- Huzhou Science and Technology Plan Project
- Open fund of Key Laboratory of Ecological and Environmental Monitoring, Forewarning and Quality Control of Zhejiang Province
- Central Guidance Funds for Science and Technology Local Development Projects
Zitieren
APA
Meng, Z., Ji, D., Lu, Y., Xu, J., Zhou, Y., Zheng, S., & Zhao, Y. (2026). A Dual-Branch TCN–TE Model for Multi-Horizon Runoff Forecasting Using Multi-Station Hydrological Observations. Sustainability, 18(11), 5289. https://doi.org/10.3390/su18115289
Vancouver
Meng Z, Ji D, Lu Y, Xu J, Zhou Y, Zheng S, et al. A Dual-Branch TCN–TE Model for Multi-Horizon Runoff Forecasting Using Multi-Station Hydrological Observations. Sustainability. 2026;18(11):5289. doi:10.3390/su18115289
BibTeX
@article{Meng2026,
author = {Meng, Zhenzhu and Ji, Dongwei and Lu, Yiqi and Xu, Jiajun and Zhou, Yuyue and Zheng, Sen and Zhao, Yinghui},
title = {A Dual-Branch TCN–TE Model for Multi-Horizon Runoff Forecasting Using Multi-Station Hydrological Observations},
journal = {Sustainability},
year = {2026},
volume = {18},
number = {11},
pages = {5289},
doi = {10.3390/su18115289}
}
EUVIMED