FRANÇAIS
A showcase of ÉTS researchers’ publications and other contributions
SEARCH

Continuous streamflow prediction in ungauged basins: long short-term memory neural networks clearly outperform traditional hydrological models

Downloads

Downloads per month over past year

Arsenault, Richard, Martel, Jean-Luc, Brunet, Frédéric, Brissette, François and Mai, Juliane. 2023. « Continuous streamflow prediction in ungauged basins: long short-term memory neural networks clearly outperform traditional hydrological models ». Hydrology and Earth System Sciences, vol. 27, nº 1. pp. 139-157.
Compte des citations dans Scopus : 27.

[thumbnail of Arsenault-R-2023-26238.pdf]
Preview
PDF
Arsenault-R-2023-26238.pdf - Published Version
Use licence: Creative Commons CC BY.

Download (4MB) | Preview

Abstract

This study investigates the ability of long short-term memory (LSTM) neural networks to perform streamflow prediction at ungauged basins. A set of state-of-the-art, hydrological model-dependent regionalization methods are applied to 148 catchments in northeast North America and compared to an LSTM model that uses the exact same available data as the hydrological models. While conceptual model-based methods attempt to derive parameterizations at ungauged sites from other similar or nearby catchments, the LSTM model uses all available data in the region to maximize the information content and increase its robustness. Furthermore, by design, the LSTM does not require explicit definition of hydrological processes and derives its own structure from the provided data. The LSTM networks were able to clearly outperform the hydrological models in a leave-one-out cross-validation regionalization setting on most catchments in the study area, with the LSTM model outperforming the hydrological models in 93 % to 97 % of catchments depending on the hydrological model. Furthermore, for up to 78 % of the catchments, the LSTM model was able to predict streamflow more accurately on pseudo-ungauged catchments than hydrological models calibrated on the target data, showing that the LSTM model's structure was better suited to convert the meteorological data and geophysical descriptors into streamflow than the hydrological models even calibrated to those sites in these cases. Furthermore, the LSTM model robustness was tested by varying its hyperparameters, and still outperformed hydrological models in regionalization in almost all cases. Overall, LSTM networks have the potential to change the regionalization research landscape by providing clear improvement pathways over traditional methods in the field of streamflow prediction in ungauged catchments.

Item Type: Peer reviewed article published in a journal
Professor:
Professor
Arsenault, Richard
Martel, Jean-Luc
Brissette, François
Affiliation: Génie de la construction, Génie de la construction, Génie de la construction
Date Deposited: 10 Mar 2023 19:11
Last Modified: 13 Mar 2023 14:17
URI: https://espace2.etsmtl.ca/id/eprint/26238

Actions (login required)

View Item View Item