Skip to main content

Advertisement

SpringerLink
Log in

Towards a better consideration of rainfall and hydrological spatial features by a deep neural network model to improve flash floods forecasting: case study on the Gardon basin, France

  • Original Article
  • Published:
Modeling Earth Systems and Environment Aims and scope Submit manuscript

Abstract

Flash floods frequently hit the Mediterranean regions and cause numerous fatalities and heavy damage. Their forecast is still a challenge because of the poor knowledge of the processes involved and because of the difficulty to forecast heavy convective rainfall. In any case, early warning remains a strong need. In this study, the authors propose to build a deep artificial neural network for flash flood forecasting, allowing, by its specific architecture, to take better account of the spatial variability and the scales of the rainfall as well as the hydrological responses. The outcomes of the deep model are then compared to a classical global multilayer perceptron previously published. For this purpose, a database of 58 heavy rainfall events extracted from 16 years of hydrometeorological observations on a well-studied basin in Southern France is applied to train a deep recurrent neural network. The results are of twofold: first, the deep model improves the lead time from two hours to three hours providing then suitable forecast for an operational use. Second, the model selection process converged towards an architecture that explicitly considers spatial scales of the basin. More generally, this study shows that the implementation of a rigorous selection process mobilizing several well-known regularization methods has enabled the deep model to converge towards a parsimonious model highlighting some of the known physical processes of the basin: the roles of elevation and distance to the outlet. This work provides, thus, a very interesting piece of evidence to fuel the controversy on the interpretability of modern AI.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Data availability statement

Data will be made available on reasonable request.

References

  • Akil N, Artigue G, Savary M, Johannet A, Vinches M (2021) Uncertainty estimation in hydrogeological forecasting with neural networks: impact of spatial distribution of rainfalls and random initialization of the model. Water 13(12):1690. https://doi.org/10.3390/w13121690

    Article  Google Scholar 

  • Anctil F, Lauzon N, Filion M (2008) Added gains of soil moisture content observations for streamflow predictions using neural networks. J Hydrol 359(3–4):225–234. https://doi.org/10.1016/j.jhydrol.2008.07.003

    Article  Google Scholar 

  • Araujo P, Astray G, Ferrerio-Lage JA, Mejuto JC, Rodriguez-Suarez JA, Soto B (2011) Multilayer perceptron neural network for flow prediction. J Environ Monit 13(1):35–41. https://doi.org/10.1039/C0EM00478B

    Article  Google Scholar 

  • Artigue G, Johannet A, Borrell V, Pistre S (2012) Flash flood forecasting in poorly gauged basins using neural networks: case study of the Gardon de Mialet basin (southern France). NHESS 12:3307–3324. https://doi.org/10.5194/nhess-12-3307-2012

    Article  Google Scholar 

  • Atmaja BT, Akagi M (2020) Deep multilayer perceptrons for dimensional speech emotion recognition, Asia-Pacific signal and information processing association annual summit and conference, APSIPA ASC 2020 – Proceedings

  • Ayral P-A (2005) Contribution à la spatialisation du modèle opérationnel de prévision des crues éclair ALHTAÏR, Thèse de l’Université de Provence Aix-Marseille, 2005. Ayral P-A: Contribution to the spatialization of the operational flash flood forecasting model ALHTAÏR, PhD, Université de Provence Aix-Marseille.

  • Barron AR (1993) Approximation bounds for superpositions of a sigmoidal function. In Proceedings of the IEEE International Symposium on Information Theory—Proceedings, San Antonio, TX, USA, 17–22 January 1993. 930–945.

  • Bengio Y, Simard P, Frasconi P (1994) Learning long-term dependencies with gradient descent is difficult. IEEE Trans Neural Networks 5(2):157–166. https://doi.org/10.1109/72.279181

    Article  Google Scholar 

  • Bertin D, Lambert J (1993) Contribution à la modélisation du système karstique : application à la prédiction des débits d’eau souterrains, Rapport de stage d’initiation à la recherche de l’Ecole des Mines d’Alès. Alès, France

  • Borga M, Anagnostou EN, Blöschl G, Creutin J-D (2011) Flash flood forecasting, warning and risk management: the HYDRATE project. Environ Sci Policy 14(7):834–844. https://doi.org/10.1016/j.envsci.2011.05.017

    Article  Google Scholar 

  • Chandar APS, Sankar C, Vorontsov E, Kahou S, Bengio Y (2019) Towards non-saturating recurrent units for modelling long-term dependencies, 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Innovative Applications of Artificial Intelligence Conference, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019, 33(01), pp. 3280–3287. https://doi.org/10.1609/aaai.v33i01.33013280.

  • Darras T, Johannet A, Vayssade B, Kong-a-Siou L, Pistre S (2014) Influence of the initialization of multilayer perceptron for flash floods forecasting: How designing a robust model. In International Work-Conference on Time Series 2014, Springer: Granada, Spain, p. 13.

  • Dawson CW, Wilby R (1998) An artificial neural network approach to rainfall runoff modelling. Hydrol Sci 43(1):47–66

    Article  Google Scholar 

  • Dreyfus G (2005) Neural networks, methodology and applications, Springer: Berlin, Germany, p. 509.

  • Fang Z, Wang Y, Peng L, Hong H (2021) Predicting flood susceptibility using LSTM neural networks. J Hydrol 594:125734. https://doi.org/10.1016/j.jhydrol.2020.125734

    Article  Google Scholar 

  • Garambois PA, Larnier K, Roux H, Labat D, Dartus D (2014) Analysis of flash flood-triggering rainfall for a process-oriented hydrological model. Atmospheric Res 137:14–24. https://doi.org/10.1016/j.atmosres.2013.09.016

    Article  Google Scholar 

  • Gaume E, Bouvier C (2004) Analyse hydro-pluviométrique des crues du Gard et du Vidourle des 8 et 9 septembre 2002, Houille Blanche, (6), 99–106, 2004. Hydro-rainfall analysis of the Gard and Vidourle floods of September 8 and 9, 2002. Houille Blanche 6:99–106

    Article  Google Scholar 

  • Gaume E, Bain V, Bernardara P, Newinger O, Barbuc M, Bateman A, Blaškovičová L, Blöschl G, Borga M, Dumitrescu A, Daliakopoulos I, Garcia J, Irimescu A, Kohnova S, Koutroulis A, Marchi L, Matreata S, Medina V, Preciso E, Sempere-Torres D, Stancalie G, Szolgay J, Tsanis I, Velasco D, Viglione A (2009) A compilation of data on European flash floods. J Hydrol 367(1–2):70–78. https://doi.org/10.1016/j.jhydrol.2008.12.028

    Article  Google Scholar 

  • Gaume E (2002) Eléments d’analyse sur les crues éclair. Université du Québec.

  • Geman S, Bienenstock E, Doursat R (1992) (1992) Neural Networks and the Bias/Variance dilemma. Neural Comput 4:1–58

    Article  Google Scholar 

  • Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735

    Article  Google Scholar 

  • Hornik K, Stinchombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2:359–366

    Article  Google Scholar 

  • Jeannin PY, Artigue G, Butscher C, Chang Y, Charlier JB, Duran L, Gill L, Hartmann A, Johannet A, Jourde H, Kavousi A, Liesch T, Liu Y, Lüthi M, Malard A, Mazzilli N, Pardo-Igúzquiza E, Thiéry D, Reimann T, Schuler P, Wöhling T, Wunsch A (2021) Karst modelling challenge 1: Results of hydrological modelling. J Hydrol 600:126508. https://doi.org/10.1016/j.jhydrol.2021.126508

    Article  Google Scholar 

  • Johannet A, Mangin A, Hulst DD (1994) Subterranean water infiltration modelling by neural networks: use of water source flow, in international conference on artificial neural networks ICANN 94. Vol II. Sorrento, Italy, pp. 1033–1036

  • Kanuparthi B, Arpit D, Kerg G, Ke NR, Mitliagkas I, Bengio Y. H-Detach (2019) Modifying the LSTM gradient towards better optimization, 7th International Conference on Learning Representations, ICLR 2019

  • Katul GG, Porporato A, Daly E, Oishi AC, Kim HS, Stoy PC, Juang JY, Siqueira MB (2007) On the spectrum of soil moisture from hourly to interannual scales. Water Resour Res. https://doi.org/10.1029/2006WR005356

    Article  Google Scholar 

  • Kitanidis PK, Bras RL (1980) Real-time forecasting with a conceptual hydrologic model: 2. Applications and results. Water Resour Res 16(6):1034–1044. https://doi.org/10.1029/WR016i006p01034

    Article  Google Scholar 

  • Kong-A-Siou L, Johannet A, Estupina V, Pistre S (2015) Neural networks for karst groundwater management: case of the Lez spring (Southern France). Environm Earth Sci 74(12):7617–7632

    Article  Google Scholar 

  • Kratzert F, Klotz D, Brenner C, Schulz and K, Herrnegger: M. (2018) Rainfall-runoff modelling using Long Short-Term Memory (LSTM) networks. Hydrol Earth Syst Sci 22(11):6005–6022. https://doi.org/10.5194/hess-22-6005-2018

    Article  Google Scholar 

  • Lauzon N, Anctil F, Petrinovic J (2004) Characterization of soil moisture conditions at temporal scales from a few days to annual. Hydrol Process 18(17):3235–3254. https://doi.org/10.1002/hyp.5656

    Article  Google Scholar 

  • Levenberg K (1944) A method for the solution of certain non-linear problems in least squares. Q Appl Math 2:164–168

    Article  Google Scholar 

  • Li W, Kiaghadi A, Dawson CN (2021) High temporal resolution rainfall runoff modelling using Long-Short-Term-Memory (LSTM) networks. Neural Comput Appl 33(4):1261–1278. https://doi.org/10.1007/s00521-020-05010-6

    Article  Google Scholar 

  • Llasat MC, Llasat-Botija M, Prat MA, Porcú F, Price C, Mugnai A, Lagouvardos K, Kotroni V, Katsanos D, Michaelides S et al (2010) High-impact floods and flash floods in Mediterranean countries: the FLASH preliminary database. Adv Geosci 23(23):47–55

    Article  Google Scholar 

  • Llasat MC, Marcos R, Llasat-Botija M, Gilabert J, Turco M, Quintana-Seguí P (2014) Flash flood evolution in North-Western Mediterranean. Atmospheric Res 149:230–243. https://doi.org/10.1016/j.atmosres.2014.05.024

    Article  Google Scholar 

  • Marchandise A (2007) Modélisation hydrologique distribuée sur le Gardon d’Anduze, étude comparative de différents modèles pluie-débit, extrapolation de la normale à l’extrême et tests d’hypothèses sur les processus hydrologiques, Distributed hydrological modeling on the Gardon d'Anduze, comparative study of different rainfall-flow models, ex-trapolation from normal to extreme and hypothesis testing on hydrological processes Université Montpellier II-Sciences et Techniques du Languedoc. [online] Available from: http://www.ohmcv.fr/Documents/theses/these_marchandise-old.pdf (Accessed 8 December 2014).

  • Marquardt DW (1963) An algorithm for least-squares estimation of nonlinear parameters. J Soc Ind Appl Math 11:431–441

    Article  Google Scholar 

  • Montz BE, Gruntfest E (2002) Flash flood mitigation: recommendations for research and applications. Glob Environ Change Part B Environ Hazards 4(1):15–22. https://doi.org/10.1016/S1464-2867(02)00011-6

    Article  Google Scholar 

  • Müller J, Park J, Sahu R, Varadharajan C, Arora B, Faybishenko B, Agarwal D (2021) Surrogate optimization of deep neural networks for groundwater predictions. J Global Optim 81(1):203–231. https://doi.org/10.1007/s10898-020-00912-0

    Article  Google Scholar 

  • Nash JE, Sutcliffe JV (1970) River flow forecasting through conceptual models part I – A discussion of principles. J Hydrol 10:282–290

    Article  Google Scholar 

  • Nerrand O, Roussel-Ragot P, Personnaz L, Dreyfus G, Marcos S (1993) Neural networks and nonlinear adaptive filtering: unifying concepts and new algorithms. Neural Comput 5:165–199

    Article  Google Scholar 

  • Nikolopoulos EI, Anagnostou EN, Borga M, Vivoni ER, Papadopoulos A (2011) Sensitivity of a mountain basin flash flood to initial wetness condition and rainfall variability. J Hydrol 402(3–4):165–178. https://doi.org/10.1016/j.jhydrol.2010.12.020

    Article  Google Scholar 

  • Obled C, Zin I, Hingray B (2008) « Choix des pas de temps et d’espace pour des modélisations parcimonieuses en hydrologie des crues »."Choice of time and space steps for parsimonious modeling in flood hydrology", Colloque SHF - Prévisions hydrométéorologiques – pp. 127–134

  • Parajka J, Naeimi V, Blöschl G, Wagner W, Merz R, Scipal K (2006) Assimilating scatterometer soil moisture data into conceptual hydrologic models at the regional scale. Hydrol Earth Syst Sci 10(3):353–368. https://doi.org/10.5194/hess-10-353-2006

    Article  Google Scholar 

  • Pascanu R, Mikolov T, Bengio Y (2013) On the difficulty of training recurrent neural networks, 30th International Conference on Machine Learning, ICML

  • Price C, Yair Y, Mugnai A, Lagouvardos K, Llasat MC, Michaelides S, Dayan U, Dietrich S, Galanti E, Garrote L, Harats N, Katsanos D, Kohn M, Kotroni V, Llasat-Botija M, Lynn B, Mediero L, Morin E, Nicolaides K, Rozalis S, Savvidou K, Ziv B (2011) The FLASH Project: using lightning data to better understand and predict flash floods. Environ Sci Policy 14(7):898–911. https://doi.org/10.1016/j.envsci.2011.03.004

    Article  Google Scholar 

  • Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors’. Nature 323(6088):533–536. https://doi.org/10.1038/323533a0

    Article  Google Scholar 

  • Sahu RK, Muller J, Park J, Varadharajan C, Arora B, Faybishenko B, Agarwal D (2020) Impact of input feature selection on groundwater level prediction from a multi-layer perceptron neural network. Front Water 2:46. https://doi.org/10.3389/frwa.2020.573034

    Article  Google Scholar 

  • Schmidhuber J (1992) Learning to control fast-weight memories: an alternative to dynamic recurrent networks. Neural Comput 4(1):131–139. https://doi.org/10.1162/neco.1992.4.1.131

    Article  Google Scholar 

  • Schmitz GH, Cullmann J (2008) PAI-OFF: a new proposal for online flood forecasting in flash flood prone catchments. J Hydrol 360:1–14

    Article  Google Scholar 

  • Shen C (2018) A transdisciplinary review of deep learning research and its relevance for water resources scientists. Water Resour Res 54(11):8558–8593. https://doi.org/10.1029/2018WR022643

    Article  Google Scholar 

  • Siou KAL, Johannet A, Borrell Estupina V, Pistre S (2011) Complexity selection of a neural network model for karst flood forecasting: The case of the Lez Basin (southern France). J Hydrol 403(3–4):367–380. https://doi.org/10.1016/j.jhydrol.2011.04.015

    Article  Google Scholar 

  • Siou KAL, Johannet A, Borrell Estupina V, Pistre S (2011) Optimization of the generalization capability for rainfall-runoff modeling by neural networks: The case of the Lez aquifer (southern France). Environm Earth Sci 65(8):2365–2375. https://doi.org/10.1007/s12665-011-1450-9

    Article  Google Scholar 

  • Sjöberg J, Zhang Q, Ljung L, Benveniste A, Delyon B, Glorennec P-Y, Hjalmarsson H, Juditskys A (1995) Nonlinear black-box modeling in system identification: a unified overview. Automatica 31:1691–1724

    Article  Google Scholar 

  • Stone M (1974) Cross-validatory choice and assessment of statistical predictions (with discussion). J R Stat Soc Ser B 38:111–147

    Google Scholar 

  • Tabari H, Talaee PH (2013) Moisture index for Iran: spatial and temporal analyses. Global Planet Change 100:11–19

    Article  Google Scholar 

  • Toukourou M, Johannet A, Dreyfus G, Ayral PA (2011) Rainfall-runoff modelling of flash floods in the absence of rainfall forecasts: the case of “Cévenol flash floods.” J Appl Intell 35:1078–1189

    Article  Google Scholar 

  • Tripathy SS, Karmakar S, Ghosh S (2021) Hazard weather scale for extreme rainfall forecast reduces uncertainty. J Hydrol 14:100106. https://doi.org/10.1016/j.wasec.2021.100106

    Article  Google Scholar 

  • Wooldridge SA, Kalma JD, Walker JP (2003) Importance of soil moisture measurements for inferring parameters in hydrologic models of low-yielding ephemeral catchments. Environ Model Softw 18(1):35–48. https://doi.org/10.1016/S1364-8152(02)00038-5

    Article  Google Scholar 

  • Wunsch A, Liesch T, Broda S (2021) Groundwater Level Forecasting with Artificial Neural Networks: A Comparison of Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNNs), and Non-Linear Autoregressive Networks with Exogenous Input (NARX)”. Hydrol Earth Syst Sci 25(3):1671–1687. https://doi.org/10.5194/hess-25-1671-2021

    Article  Google Scholar 

  • Zhang J, Zhu Y, Zhanga X, Ye M, Yang J (2018) Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas. J Hydrol 561:918–929

    Article  Google Scholar 

  • Zulifqar Ali Z, Hussain I, Faisal M, Nazir HM, Hussain T, Shad MY, Shoukry AM, Gani SH (2017) Forecasting drought using multilayer perceptron artificial neural network model. Adv Meteorol. https://doi.org/10.1155/2017/5681308

    Article  Google Scholar 

Download references

Funding

IMT Mines Alès

Author information

Author notes

  1. Bob E. Saint-Fleur

    Present address: HydroSciences Montpellier, Univ. Montpellier, CNRS, IRD, Montpellier, France

Authors and Affiliations

  1. HydroSciences Montpellier, Univ. Montpellier, IMT Mines Alès, CNRS, IRD, Ales, France

    Bob E. Saint-Fleur, Sam Allier, Emilien Lassara, Antoine Rivet, Guillaume Artigue & Anne Johannet

  2. HydroSciences Montpellier, Univ. Montpellier, CNRS, IRD, Montpellier, France

    Séverin Pistre

Authors
  1. Bob E. Saint-Fleur
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sam Allier
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Emilien Lassara
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Antoine Rivet
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Guillaume Artigue
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Séverin Pistre
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Anne Johannet
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Anne Johannet.

Ethics declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Saint-Fleur, B.E., Allier, S., Lassara, E. et al. Towards a better consideration of rainfall and hydrological spatial features by a deep neural network model to improve flash floods forecasting: case study on the Gardon basin, France. Model. Earth Syst. Environ. 9, 3693–3708 (2023). https://doi.org/10.1007/s40808-022-01650-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s40808-022-01650-w

Keywords

Search

Navigation