UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Improving LSTM hydrological modeling with spatiotemporal deep learning and multi-task learning: a case study of three mountainous areas on the Tibetan Plateau

Li, Bu; Li, Ruidong; Sun, Ting; Gong, Aofan; Tian, Fuqiang; Yawar Ali Khan, Mohd; Ni, Guangheng; (2023) Improving LSTM hydrological modeling with spatiotemporal deep learning and multi-task learning: a case study of three mountainous areas on the Tibetan Plateau. Journal of Hydrology , 620 (A) , Article 129401. 10.1016/j.jhydrol.2023.129401. Green open access

[thumbnail of 1-s2.0-S0022169423003438-main.pdf]
Preview
Text
1-s2.0-S0022169423003438-main.pdf - Accepted Version

Download (2MB) | Preview

Abstract

Long short-term memory (LSTM) networks have demonstrated their excellent capability in processing long-length temporal dynamics and have proven to be effective in precipitation-runoff modeling. However, the current LSTM hydrological models lack the incorporation of multi-task learning and spatial information, which limits their ability to make full use of meteorological and hydrological data. To address this issue, this study proposes a spatiotemporal deep-learning (DL)-based hydrological model that couples the 2-Dimension convolutional neural network (CNN) and LSTM and introduces actual evaporation ( ) as an additional training target. The proposed CNN-LSTM model is tested on three large mountainous basins on the Tibetan Plateau, and the results are compared to those obtained from the LSTM-only model. Additionally, a probe method is used to decipher the internal embedding layers of the proposed DL models. The results indicate that both LSTM and CNN-LSTM hydrological models perform well in simulating runoff ( ) and , with Nash-Sutcliffe efficiency coefficients ( ) higher than 0.82 and 0.95, respectively. The higher suggest that introducing spatial information into LSTM-only models can improve the overall and peak model performance. Moreover, multi-task simulation with LSTM-only models shows better accuracy in the estimation of volume and performance, with increasing by approximately 0.02. The probe method also reveals that CNN can capture the basin-averaged meteorological values in CNN-LSTM models, while LSTM ( ) models contain the information about the known ( ) process. Overall, this study demonstrates the value of spatial information and multi-task learning in LSTM hydrological modeling and provides a perspective for interpreting the internal embedding layers of DL models.

Type: Article
Title: Improving LSTM hydrological modeling with spatiotemporal deep learning and multi-task learning: a case study of three mountainous areas on the Tibetan Plateau
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.jhydrol.2023.129401
Publisher version: https://doi.org/10.1016/j.jhydrol.2023.129401
Language: English
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords: CNN-LSTM, spatiotemporal, multi-task, actual evaporation, Tibetan Plateau
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Inst for Risk and Disaster Reduction
URI: https://discovery.ucl.ac.uk/id/eprint/10166707
Downloads since deposit
76Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item