Articles | Volume 26, issue 22
https://doi.org/10.5194/hess-26-5793-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/hess-26-5793-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
17 Nov 2022
Research article |
| 17 Nov 2022
| 17 Nov 2022
How can we benefit from regime information to make more effective use of long short-term memory (LSTM) runoff models?
Reyhaneh Hashemi
CORRESPONDING AUTHOR
INRAE, Aix-Marseille University, RECOVER Research Unit, Aix-en-Provence, France
Pierre Brigode
Université Côte d'Azur, OCA, CNRS, IRD, GEOAZUR, France
INRAE, Paris-Saclay University, HYCAR Research Unit, Antony, France
Pierre-André Garambois
INRAE, Aix-Marseille University, RECOVER Research Unit, Aix-en-Provence, France
Pierre Javelle
INRAE, Aix-Marseille University, RECOVER Research Unit, Aix-en-Provence, France
Related authors
Reyhaneh Hashemi, Pierre Javelle, Olivier Delestre, and Saman Razavi
Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2023-282, https://doi.org/10.5194/hess-2023-282, 2023
Manuscript not accepted for further review
Short summary
Short summary
Here, we have tackled the challenge of estimating water flow in areas without direct measurements, a crucial task in hydrology. We have applied deep learning techniques to a large sample of French catchments with various hydrological regimes. We have also compared our approach with traditional methods. We found that incorporating more data improves the accuracy of our deep learning predictions. Notably, our method outperforms traditional approaches in certain regimes, though not universally.
Ngo Nghi Truyen Huynh, Pierre-André Garambois, Benjamin Renard, François Colleoni, Jérôme Monnier, and Hélène Roux
Hydrol. Earth Syst. Sci., 29, 3589–3613, https://doi.org/10.5194/hess-29-3589-2025, https://doi.org/10.5194/hess-29-3589-2025, 2025
Short summary
Short summary
Understanding and modeling flash-flood-prone areas remains challenging due to limited data and scale-relevant hydrological theory. While machine learning shows promise, its integration with process-based models is difficult. We present an approach incorporating machine learning into a high-resolution hydrological model to correct internal fluxes and transfer parameters between watersheds. Results show improved accuracy, advancing the development of learnable and interpretable process-based models.
Ngo Nghi Truyen Huynh, Pierre-André Garambois, François Colleoni, and Jérôme Monnier
EGUsphere, https://doi.org/10.5194/egusphere-2025-2797, https://doi.org/10.5194/egusphere-2025-2797, 2025
This preprint is open for discussion and under review for Geoscientific Model Development (GMD).
Short summary
Short summary
To better understand hydrological processes and improve flood simulation, combining artificial intelligence (AI) with process-based models is a promising direction. We introduce a hybrid physics-AI approach that seamlessly integrates neural networks into a distributed hydrological model to refine water flow dynamics within an implicit numerical scheme. The hybrid models demonstrate strong performance and interpretable results, leading to reliable streamflow simulations for flood modeling.
Juliette Godet, Pierre Nicolle, Nabil Hocini, Eric Gaume, Philippe Davy, Frederic Pons, Pierre Javelle, Pierre-André Garambois, Dimitri Lague, and Olivier Payrastre
Earth Syst. Sci. Data, 17, 2963–2983, https://doi.org/10.5194/essd-17-2963-2025, https://doi.org/10.5194/essd-17-2963-2025, 2025
Short summary
Short summary
This paper describes a dataset that includes input, output, and validation data for the simulation of flash flood hazards and three specific flash flood events in the French Mediterranean region. This dataset is particularly valuable as flood mapping methods often lack sufficient benchmark data. Additionally, we demonstrate how the hydraulic method we used, named Floodos, produces highly satisfactory results.
Olivier Delaigue, Guilherme Mendoza Guimarães, Pierre Brigode, Benoît Génot, Charles Perrin, Jean-Michel Soubeyroux, Bruno Janet, Nans Addor, and Vazken Andréassian
Earth Syst. Sci. Data, 17, 1461–1479, https://doi.org/10.5194/essd-17-1461-2025, https://doi.org/10.5194/essd-17-1461-2025, 2025
Short summary
Short summary
This dataset covers 654 rivers all flowing in France. The provided time series and catchment attributes will be of interest to those modelers wishing to analyze hydrological behavior and perform model assessments.
Juliette Godet, Eric Gaume, Pierre Javelle, Thomas Dias, Pierre Nicolle, and Olivier Payrastre
Abstr. Int. Cartogr. Assoc., 9, 16, https://doi.org/10.5194/ica-abs-9-16-2025, https://doi.org/10.5194/ica-abs-9-16-2025, 2025
François Colleoni, Ngo Nghi Truyen Huynh, Pierre-André Garambois, Maxime Jay-Allemand, Didier Organde, Benjamin Renard, Thomas De Fournas, Apolline El Baz, Julie Demargne, and Pierre Javelle
EGUsphere, https://doi.org/10.5194/egusphere-2025-690, https://doi.org/10.5194/egusphere-2025-690, 2025
Short summary
Short summary
We present smash, an open-source framework for high-resolution hydrological modeling and data assimilation. It combines process-based models with neural networks for regionalization, enabling accurate simulations from catchment to country scale. With an efficient, differentiable solver, smash supports large-scale calibration and parallel computing. Tested on open datasets, it shows strong performance in river flow prediction, making it a valuable tool for research and operational use.
Pierre Brigode and Ludovic Oudin
Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2024-336, https://doi.org/10.5194/hess-2024-336, 2024
Preprint under review for HESS
Short summary
Short summary
We analyzed how well two global climate datasets can simulate river flows across Europe over the last 150 years. Our results show good performance overall, revealing important long-term changes in water availability and extreme events, like floods, in different regions. This research helps us better understand past and future water trends, providing insights to manage resources and address the challenges posed by climate change.
Carlo Mologni, Marie Revel, Eric Chaumillon, Emmanuel Malet, Thibault Coulombier, Pierre Sabatier, Pierre Brigode, Gwenael Hervé, Anne-Lise Develle, Laure Schenini, Medhi Messous, Gourguen Davtian, Alain Carré, Delphine Bosch, Natacha Volto, Clément Ménard, Lamya Khalidi, and Fabien Arnaud
Clim. Past, 20, 1837–1860, https://doi.org/10.5194/cp-20-1837-2024, https://doi.org/10.5194/cp-20-1837-2024, 2024
Short summary
Short summary
The reactivity of local to regional hydrosystems to global changes remains understated in East African climate models. By reconstructing a chronicle of seasonal floods and droughts from a lacustrine sedimentary core, this paper highlights the impact of El Niño anomalies in the Awash River valley (Ethiopia). Studying regional hydrosystem feedbacks to global atmospheric anomalies is essential for better comprehending and mitigating the effects of global warming in extreme environments.
Ralph Bathelemy, Pierre Brigode, Vazken Andréassian, Charles Perrin, Vincent Moron, Cédric Gaucherel, Emmanuel Tric, and Dominique Boisson
Earth Syst. Sci. Data, 16, 2073–2098, https://doi.org/10.5194/essd-16-2073-2024, https://doi.org/10.5194/essd-16-2073-2024, 2024
Short summary
Short summary
The aim of this work is to provide the first hydroclimatic database for Haiti, a Caribbean country particularly vulnerable to meteorological and hydrological hazards. The resulting database, named Simbi, provides hydroclimatic time series for around 150 stations and 24 catchment areas.
Maxime Jay-Allemand, Julie Demargne, Pierre-André Garambois, Pierre Javelle, Igor Gejadze, François Colleoni, Didier Organde, Patrick Arnaud, and Catherine Fouchier
Proc. IAHS, 385, 281–290, https://doi.org/10.5194/piahs-385-281-2024, https://doi.org/10.5194/piahs-385-281-2024, 2024
Short summary
Short summary
This work targets the improvement of a hydrologic model used for flash flood warnings. A gridded model is used to spatially describe the hydrological processes. We develop a method to estimate the best model setup based on scarce river flow observations. It uses a complex algorithm combined with geographical descriptors to generate gridded parameters that better capture catchment characteristics. Results are promising, improving the discharge estimations where no observations are available.
Juliette Godet, Eric Gaume, Pierre Javelle, Pierre Nicolle, and Olivier Payrastre
Hydrol. Earth Syst. Sci., 28, 1403–1413, https://doi.org/10.5194/hess-28-1403-2024, https://doi.org/10.5194/hess-28-1403-2024, 2024
Short summary
Short summary
This work was performed in order to precisely address a point that is often neglected by hydrologists: the allocation of points located on a river network to grid cells, which is often a mandatory step for hydrological modelling.
Guillaume Evin, Matthieu Le Lay, Catherine Fouchier, David Penot, Francois Colleoni, Alexandre Mas, Pierre-André Garambois, and Olivier Laurantin
Hydrol. Earth Syst. Sci., 28, 261–281, https://doi.org/10.5194/hess-28-261-2024, https://doi.org/10.5194/hess-28-261-2024, 2024
Short summary
Short summary
Hydrological modelling of mountainous catchments is challenging for many reasons, the main one being the temporal and spatial representation of precipitation forcings. This study presents an evaluation of the hydrological modelling of 55 small mountainous catchments of the northern French Alps, focusing on the influence of the type of precipitation reanalyses used as inputs. These evaluations emphasize the added value of radar measurements, in particular for the reproduction of flood events.
Reyhaneh Hashemi, Pierre Javelle, Olivier Delestre, and Saman Razavi
Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2023-282, https://doi.org/10.5194/hess-2023-282, 2023
Manuscript not accepted for further review
Short summary
Short summary
Here, we have tackled the challenge of estimating water flow in areas without direct measurements, a crucial task in hydrology. We have applied deep learning techniques to a large sample of French catchments with various hydrological regimes. We have also compared our approach with traditional methods. We found that incorporating more data improves the accuracy of our deep learning predictions. Notably, our method outperforms traditional approaches in certain regimes, though not universally.
Juliette Godet, Olivier Payrastre, Pierre Javelle, and François Bouttier
Nat. Hazards Earth Syst. Sci., 23, 3355–3377, https://doi.org/10.5194/nhess-23-3355-2023, https://doi.org/10.5194/nhess-23-3355-2023, 2023
Short summary
Short summary
This article results from a master's research project which was part of a natural hazards programme developed by the French Ministry of Ecological Transition. The objective of this work was to investigate a possible way to improve the operational flash flood warning service by adding rainfall forecasts upstream of the forecasting chain. The results showed that the tested forecast product, which is new and experimental, has a real added value compared to other classical forecast products.
Laurent Strohmenger, Eric Sauquet, Claire Bernard, Jérémie Bonneau, Flora Branger, Amélie Bresson, Pierre Brigode, Rémy Buzier, Olivier Delaigue, Alexandre Devers, Guillaume Evin, Maïté Fournier, Shu-Chen Hsu, Sandra Lanini, Alban de Lavenne, Thibault Lemaitre-Basset, Claire Magand, Guilherme Mendoza Guimarães, Max Mentha, Simon Munier, Charles Perrin, Tristan Podechard, Léo Rouchy, Malak Sadki, Myriam Soutif-Bellenger, François Tilmant, Yves Tramblay, Anne-Lise Véron, Jean-Philippe Vidal, and Guillaume Thirel
Hydrol. Earth Syst. Sci., 27, 3375–3391, https://doi.org/10.5194/hess-27-3375-2023, https://doi.org/10.5194/hess-27-3375-2023, 2023
Short summary
Short summary
We present the results of a large visual inspection campaign of 674 streamflow time series in France. The objective was to detect non-natural records resulting from instrument failure or anthropogenic influences, such as hydroelectric power generation or reservoir management. We conclude that the identification of flaws in flow time series is highly dependent on the objectives and skills of individual evaluators, and we raise the need for better practices for data cleaning.
Olivier Delaigue, Pierre Brigode, Guillaume Thirel, and Laurent Coron
Hydrol. Earth Syst. Sci., 27, 3293–3327, https://doi.org/10.5194/hess-27-3293-2023, https://doi.org/10.5194/hess-27-3293-2023, 2023
Short summary
Short summary
Teaching hydrological modeling is an important, but difficult, matter. It requires appropriate tools and teaching material. In this article, we present the airGRteaching package, which is an open-source software tool relying on widely used hydrological models. This tool proposes an interface and numerous hydrological modeling exercises representing a wide range of hydrological applications. We show how this tool can be applied to simple but real-life cases.
Léo Pujol, Pierre-André Garambois, and Jérôme Monnier
Geosci. Model Dev., 15, 6085–6113, https://doi.org/10.5194/gmd-15-6085-2022, https://doi.org/10.5194/gmd-15-6085-2022, 2022
Short summary
Short summary
This contribution presents a new numerical model for representing hydraulic–hydrological quantities at the basin scale. It allows modeling large areas at a low computational cost, with fine zooms where needed. It allows the integration of local and satellite measurements, via data assimilation methods, to improve the model's match to observations. Using this capability, good matches to in situ observations are obtained on a model of the complex Adour river network with fine zooms on floodplains.
François Colleoni, Pierre-André Garambois, Pierre Javelle, Maxime Jay-Allemand, and Patrick Arnaud
EGUsphere, https://doi.org/10.5194/egusphere-2022-506, https://doi.org/10.5194/egusphere-2022-506, 2022
Preprint archived
Short summary
Short summary
This contribution presents the first evaluation of Variational Data Assimilation successfully applied over a large sample to the spatially distributed calibration of a newly taylored grid-based parsimonious model structure and corresponding adjoint. High performances are obtained in spatio-temporal validation and at flood time scales, especially for mediterranenan and oceanic catchments. Regional sensitivity analysis revealed the importance of the non conservative and production components.
Małgorzata Chmiel, Maxime Godano, Marco Piantini, Pierre Brigode, Florent Gimbert, Maarten Bakker, Françoise Courboulex, Jean-Paul Ampuero, Diane Rivet, Anthony Sladen, David Ambrois, and Margot Chapuis
Nat. Hazards Earth Syst. Sci., 22, 1541–1558, https://doi.org/10.5194/nhess-22-1541-2022, https://doi.org/10.5194/nhess-22-1541-2022, 2022
Short summary
Short summary
On 2 October 2020, the French Maritime Alps were struck by an extreme rainfall event caused by Storm Alex. Here, we show that seismic data provide the timing and velocity of the propagation of flash-flood waves along the Vésubie River. We also detect 114 small local earthquakes triggered by the rainwater weight and/or its infiltration into the ground. This study paves the way for future works that can reveal further details of the impact of Storm Alex on the Earth’s surface and subsurface.
Abubakar Haruna, Pierre-Andre Garambois, Helene Roux, Pierre Javelle, and Maxime Jay-Allemand
Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2021-414, https://doi.org/10.5194/hess-2021-414, 2021
Manuscript not accepted for further review
Short summary
Short summary
We compared three hydrological models in a flash flood modelling framework. We first identified the sensitive parameters of each model, then compared their performances in terms of outlet discharge and soil moisture simulation. We found out that resulting from the differences in their complexities/process representation, performance depends on the aspect/measure used. The study then highlights and proposed some future investigations/modifications to improve the models.
Maxime Jay-Allemand, Pierre Javelle, Igor Gejadze, Patrick Arnaud, Pierre-Olivier Malaterre, Jean-Alain Fine, and Didier Organde
Hydrol. Earth Syst. Sci., 24, 5519–5538, https://doi.org/10.5194/hess-24-5519-2020, https://doi.org/10.5194/hess-24-5519-2020, 2020
Short summary
Short summary
This study contributes to flash flood prediction using a hydrological model. The model describes the spatial properties of the watersheds with hundreds of unknown parameters. The Gardon d'Anduze watershed is chosen as the study benchmark. A sophisticated numerical algorithm and the downstream discharge measurements make the identification of the model parameters possible. Results provide better model predictions and relevant spatial variability of some parameters inside this watershed.
Cited articles
Beck, C., Jentzen, A., and Kuckuck, B.: Full error analysis for the training of deep neural networks, Infin. Dimens. Anal. Qu., 25, 2150020, https://doi.org/10.1142/S021902572150020X, 2022. a
Bracken, L. J. and Croke, J.: The concept of hydrological connectivity and its contribution to understanding runoff-dominated geomorphic systems,
Hydrol. Process., 21, 1749–1763, https://doi.org/10.1002/hyp.6313, 2007. a
Burnash, R. J. C., Ferral, R. L., and McGuire, R. A.: A generalized streamflow simulation system: Conceptual modeling for digital computers,
Cooperatively developed by the Joint Federal-State River Forecast Center, United States Department of Commerce, National Weather Service, State of California, Department of Water Resources, https://books.google.fr/books?hl=en&lr=&id=aQJDAAAAIAAJ&oi=fnd&pg=PR2&dq=A+generalised+streamflow+simulation+system+conceptual+modelling+for+digital+computers.,+Tech.+rep.,+US+Department+of+Commerce+National+Weather+Service+and+State+of+California+Department+of+Water+Resources&ots=4tUeYd75bu&sig=9E64OzUeZxuyF4ULMgxbQyr9ktI&redir_esc=y#v=onepage&q&f=false) (last access: 16 November 2022), 1973. a
Chiverton, A., Hannaford, J., Holman, I., Corstanje, R., Prudhomme, C., Bloomfield, J., and Hess, T. M.:
Which catchment characteristics control the temporal dependence structure of daily river flows?, Hydrol. Process., 29, 1353–1369, https://doi.org/10.1002/hyp.10252, 2015. a
Chollet, F. et al.: Keras, GitHub, https://github.com/fchollet/keras (last access: 2 November 2022), 2015. a
Clark, M. P., Slater, A. G., Rupp, D. E., Woods, R. A., Vrugt, J. A., Gupta, H. V., Wagener, T., and Hay, L. E.: Framework for Understanding Structural Errors (FUSE): A modular framework to diagnose differences between hydrological models, Water Resour. Res., 44, W00B02,
https://doi.org/10.1029/2007WR006735, 2008. a
Coron, L., Thirel, G., Delaigue, O., Perrin, C., and Andréassian, V.:
The Suite of Lumped GR Hydrological Models in an R package, Environ. Modell. Softw., 94, 166–171, https://doi.org/10.1016/j.envsoft.2017.05.002, 2017. a
Coron, L., Delaigue, O., Thirel, G., Perrin, C., and Michel, C.:
airGR: Suite of GR Hydrological Models for Precipitation-Runoff Modelling, R package version 1.4.3.65, https://CRAN.R-project.org/package=airGR (last access: 2 November 2022), 2020. a
Delaigue, O., Génot, B., Lebecherel, L., Brigode, P., and Bourgin, P.-Y.:
Database of watershed-scale hydroclimatic observations in France, Recherche Data Gouv [data set], https://doi.org/10.15454/UV01P1, 2020. a, b, c
de Lavenne, A., Andréassian, V., Crochemore, L., Lindström, G., and Arheimer, B.: Quantifying multi-year hydrological memory with Catchment Forgetting Curves, Hydrol. Earth Syst. Sci., 26, 2715–2732, https://doi.org/10.5194/hess-26-2715-2022, 2022. a, b, c, d
Feng, D., Fang, K., and Shen, C.:
Enhancing streamflow forecast and extracting insights using long-short term memory networks with data integration at continental scales, Water Resour. Res., 56, e2019WR026793, https://doi.org/10.1029/2019WR026793, 2020. a
Frame, J. M., Kratzert, F., Klotz, D., Gauch, M., Shalev, G., Gilon, O., Qualls, L. M., Gupta, H. V., and Nearing, G. S.:
Deep learning rainfall–runoff predictions of extreme events, Hydrol. Earth Syst. Sci., 26, 3377–3392, https://doi.org/10.5194/hess-26-3377-2022, 2022. a
French Ministry of Environment: Hydro Portail, https://hydro.eaufrance.fr/edito/a-propos-de-lhydroportail, last access: 9 November 2022. a
Gao, S., Huang, Y., Zhang, S., Han, J., Wang, G., Zhang, M., and Lin, Q.:
Short-term runoff prediction with GRU and LSTM networks without requiring time step optimization during sample generation, J. Hydrol., 589, 125188, https://doi.org/10.1016/j.jhydrol.2020.125188, 2020. a
Gauch, M., Kratzert, F., Klotz, D., Nearing, G., Lin, J., and Hochreiter, S.:
Rainfall–runoff prediction at multiple timescales with a single Long Short-Term Memory network, Hydrol. Earth Syst. Sci., 25, 2045–2062, https://doi.org/10.5194/hess-25-2045-2021, 2021a. a, b
Gauch, M., Mai, J., and Lin, J.:
The proper care and feeding of CAMELS: How limited training data affects streamflow prediction, Environ. Modell. Softw., 135, 104926, https://doi.org/10.1016/j.envsoft.2020.104926, 2021b. a, b, c
Gupta, H. V., Kling, H., Yilmaz, K. K., and Martinez, G. F.:
Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling, J. Hydrol., 377, 80–91, https://doi.org/10.1016/j.jhydrol.2009.08.003, 2009. a
Haines, A., Finlayson, B., and McMahon, T.:
A global classification of river regimes, Appl. Geogr., 8, 255–272, https://doi.org/10.1016/0143-6228(88)90035-5, 1988. a
Hochreiter, S.:
The vanishing gradient problem during learning recurrent neural nets and problem solutions, Int. J. Uncertain. Fuzz., 6, 107–116, https://doi.org/10.1142/S0218488598000094, 1998. a
Hochreiter, S. and Schmidhuber, J.: Long Short-Term Memory, Neural Comput., 9, 1735–1780, https://doi.org/10.1162/neco.1997.9.8.1735, 1997. a
Kachroo, R. and Natale, L.:
Non-linear modelling of the rainfall-runoff transformation, J. Hydrol., 135, 341–369, https://doi.org/10.1016/0022-1694(92)90095-D, 1992. a
Kingma, D. P. and Ba, J.:
Adam: A Method for Stochastic Optimization, https://doi.org/10.48550/arXiv.1412.6980, 2017. a
Kratzert, F., Klotz, D., Herrnegger, M., Sampson, A. K., Hochreiter, S., and Nearing, G. S.:
Toward Improved Predictions in Ungauged Basins: Exploiting the Power of Machine Learning, Water Resour. Res., 55, 11344–11354, https://doi.org/10.1029/2019WR026065, 2019a. a, b, c
Kratzert, F., Klotz, D., Shalev, G., Klambauer, G., Hochreiter, S., and Nearing, G.: Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets, Hydrol. Earth Syst. Sci., 23, 5089–5110, https://doi.org/10.5194/hess-23-5089-2019, 2019b. a, b, c, d, e, f, g, h, i
Leavesley, G. H., Lichty, R. W., Troutman, B. M., and Saindon, L. G.:
US Geological Survey Water Resources Investigations, Water-Resources Investigations Report 83-4238, USGS, https://pubs.usgs.gov/wri/1983/4238/report.pdf (last access: 8 November 2022), 1983. a
LeCun, Y.-A., Bottou, L., Orr, G.-B., and Müller, K.-R.: Efficient backprop, in: Neural networks: Tricks of the trade, edited by: Montavon, G.,
Orr, G. B., and Müller, K.-R., Springer, 9–48, https://doi.org/10.1007/978-3-642-35289-8_3, 2012. a
Lees, T., Buechel, M., Anderson, B., Slater, L., Reece, S., Coxon, G., and Dadson, S. J.:
Benchmarking data-driven rainfall–runoff models in Great Britain: a comparison of long short-term memory (LSTM)-based models with four lumped conceptual models, Hydrol. Earth Syst. Sci., 25, 5517–5534, https://doi.org/10.5194/hess-25-5517-2021, 2021. a, b, c, d, e, f, g, h, i
Météo France: Données publiques,
https://donneespubliques.meteofrance.fr/, last access: 9 November 2022. a
Michel, C.: Hydrologie appliquée aux petits bassins versants ruraux (Applied hydrology for small catchments), internal Report, Cemagref, Antony, France, https://side.developpement-durable.gouv.fr/Default/doc/SYRACUSE/162685/hydrologie-appliquee-aux-petits-bassins-ruraux?_lg=fr-FR
(last access: 1 November 2022), 1989. a
Nearing, G. S., Klotz, D., Sampson, A. K., Kratzert, F., Gauch, M., Frame, J. M., Shalev, G., and Nevo, S.:
Technical Note: Data assimilation and autoregression for using near-real-time streamflow observations in long short-term memory networks, Hydrol. Earth Syst. Sci. Discuss. [preprint], https://doi.org/10.5194/hess-2021-515, in review, 2021. a
O, S., Dutra, E., and Orth, R.:
Robustness of Process-Based versus Data-Driven Modeling in Changing Climatic Conditions, J. Hydrometeorol., 21, 1929–1944, https://doi.org/10.1175/JHM-D-20-0072.1, 2020. a
Olah, C.: Understanding LSTM Networks, colah's blog, https://colah.github.io/posts/2015-08-Understanding-LSTMs/ (last access: 1 November 2022), 2015. a
Omernik, J. M. and Griffith, G. E.:
Ecoregions of the conterminous United States: evolution of a hierarchical spatial framework, Environ. Manage., 54, 1249–1266, https://doi.org/10.1007/s00267-014-0364-1, 2014. a
Oudin, L., Michel, C., and Anctil, F.:
Which potential evapotranspiration input for a lumped rainfall-runoff model?: Part 1–can rainfall-runoff models effectively handle detailed potential evapotranspiration inputs?, J. Hydrol., 303, 275–289, https://doi.org/10.1016/j.jhydrol.2004.08.026, 2005. a, b
Oudin, L., Andréassian, V., Perrin, C., Michel, C., and Le Moine, N.:
Spatial proximity, physical similarity, regression and ungaged catchments: A comparison of regionalization approaches based on 913 French catchments, Water Resour. Res., 44, https://doi.org/10.1029/2007WR006240, 2008. a, b
Pardé, M.: Fleuves et rivières, Collection Armand Colin; Section de Géographie (France) fre no. 155, Librairie Armand Colin, https://www.sorbonne-librairie.fr/listeliv.php?form_recherche_avancee=ok&tag=185605&base=ebook&select_tri_recherche=pertinence&page=1&formats=pdf
(last access: 2 November 2022), 1933. a
Perrin, C. and Littlewood, I.: A comparative assessment of two rainfall-runoff modelling approaches: GR4J and IHACRES, in: Proceedings of the Liblice Conference (22–24 September 1998), edited by: Elias, V. and Littlewood, I. G,, IHP-V, Technical Documents in Hydrology, vol. 37, 191–201, 2000. a
Perrin, C., Michel, C., and Andréassian, V.:
Improvement of a parsimonious model for streamflow simulation, J. Hydrol,, 279, 275–289, https://doi.org/10.1016/S0022-1694(03)00225-7, 2003. a, b
Phillips, J. D.: Sources of nonlinearity and complexity in geomorphic systems, Progress in Physical Geography: Earth and Environment, 27, 1–23, https://doi.org/10.1191/0309133303pp340ra, 2003. a
Quintana-Segui, P., Moigne, P. L., Durand, Y., Martin, E., Habets, F., Baillon, M., Canellas, C., Franchisteguy, L., and Morel, S.:
Analysis of near-surface atmospheric variables: Validation of the SAFRAN analysis over France, J. Appl. Meteorol. Clim., 47, 92–107, https://doi.org/10.1175/2007JAMC1636.1, 2008. a, b
R Core Team: R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/ (last access: 1 November 2022), 2019. a
Sauquet, E.:
Mapping mean annual river discharges: geostatistical developments for incorporating river network dependencies, J. Hydrol., 331, 300–314, https://doi.org/10.1016/j.jhydrol.2006.05.018, 2006. a
Valéry, A., Andréassian, V., and Perrin, C.:
'As simple as possible but not simpler': What is useful in a temperature-based snow-accounting routine? Part 2 – Sensitivity analysis of the Cemaneige snow accounting routine on 380 catchments, J. Hydrol., 517, 1176–1187, https://doi.org/10.1016/j.jhydrol.2014.04.058, 2014.
a, b, c
Van Rossum, G. and Drake, F. L.:
Python 3 Reference Manual, CreateSpace, Scotts Valley, CA, 2009. a
Vidal, J.-P., Martin, E., Franchistéguy, L., Baillon, M., and Soubeyroux, J.-M.:
A 50-year high-resolution atmospheric reanalysis over France with the Safran system, Int. J. Climatol., 30, 1627–1644, https://doi.org/10.1002/joc.2003, 2010. a, b
Short summary
Hydrologists have long dreamed of a tool that could adequately predict runoff in catchments. Data-driven long short-term memory (LSTM) models appear very promising to the hydrology community in this respect. Here, we have sought to benefit from traditional practices in hydrology to improve the effectiveness of LSTM models. We discovered that one LSTM parameter has a hydrologic interpretation and that there is a need to increase the data and to tune two parameters, thereby improving predictions.
Hydrologists have long dreamed of a tool that could adequately predict runoff in catchments....
