Predicting sea surface salinity in a tidal estuary with machine learning

• Autorzy: Guillou Nicolas , Chapalain Georges , Petton Sébastien

Identyfikatory

DOI

10.1016/j.oceano.2022.07.007

• Języki publikacji: EN

Abstrakty

EN

As an indicator of exchanges between watersheds, rivers and coastal seas, salinity may provide valuable information about the exposure, ecological health and robustness of marine ecosystems, including especially estuaries. The temporal variations of salinity are traditionally approached with numerical models based on a physical description of hydrodynamic and hydrological processes. However, as these models require large computational resources, such an approach is, in practice, rarely considered for rapid turnaround predictions as requested by engineering and operational applications dealing with the ecological monitoring of estuaries. As an alternative efficient and rapid solution, we investigated here the potential of machine learning algorithms to mimic the non-linear complex relationships between salinity and a series of input parameters (such as tide-induced free-surface elevation, river discharges and wind velocity). Beyond regression methods, the attention was dedicated to popular machine learning approaches including MultiLayer Perceptron, Support Vector Regression and Random Forest. These algorithms were applied to six-year observations of sea surface salinity at the mouth of the Elorn estuary (bay of Brest, western Brittany, France) and compared to predictions from an advanced ecological numerical model. In spite of simple input data, machine learning algorithms reproduced the seasonal and semi-diurnal variations of sea surface salinity characterised by noticeable tide-induced modulations and low-salinity events during the winter period. Support Vector Regression provided the best estimations of surface salinity, improving especially predictions from the advanced numerical model during low-salinity events. This promotes the exploitation of machine learning algorithms as a complementary tool to process-based physical models.

Słowa kluczowe

EN

multilayer perceptron; support vector regression; random forest; river plume; numerical model; Bay of Brest

• Wydawca: Polish Academy of Sciences, Institute of Oceanology ; Elsevier
• Czasopismo: Oceanologia
• Rocznik: 2023
• Tom: Vol. 65 (2)
• Strony: 318--332
• Opis fizyczny: Bibliog. 49 poz., map., rys., tab., wykr.

Twórcy

autor

Guillou Nicolas

• Cerema, DTecREM, HA, Technopôle Brest-Iroise, Plouzané, France

autor

Chapalain Georges

• Cerema, DTecREM, HA, Technopôle Brest-Iroise, Plouzané, France

autor

Petton Sébastien

• Ifremer, University of Brest, CNRS, IRD, LEMAR, Argenton, France

• https://orcid.org/0000-0001-6308-2590

Bibliografia

• 1. Agence de l’eau Loire Bretagne, 1997. Contrat de baie - La Rade de Brest et son bassin versant : Etat des lieā. URL: https://www.documentation.eauetbiodiversite.fr/notice/00000000015df039937aacbc62f2d250 (accessed 11.5.21).
• 2. Alizadeh, M.J., Kavianpour, M.R., Danesh, M., Adolf, J., Shamshirband, S., Chau, K.-W., 2018. Effect of river flow on the quality of estuarine and coastal waters using machine learning models. Engineering Appl. Comput. Fluid Mech. 12, 810-823. https://doi.org/10.1080/19942060.2018.1528480
• 3. Auffret, G., 1983. Dynamique sédimentaire de la marge continentale celtique - Evolution Cénozoïque - Spécificité du Pleistocène supérieur et de l’Holocène. Semantic Scholar [WWW Document], URL https://www.semanticscholar.org/paper/Dynamique-s%C3%A9dimentaire-la-marge-continentale-du-Auffret/1b555139d5867c69dbedecd6f4455e1d7a7e2094 (accessed 11.3.21).
• 4. Azencott, C.A., 2019. Introduction au Machine Learning, Dunod, Cambridge, UK., 227 pp.
• 5. Banque, Hydro, 2021. Banque Hydro. http://hydro.eaufrance.fr/indexs.php (accessed on 05/2021).
• 6. Beudin, A., Chapalain, G., Guillou, N., 2014. Modelling dynamics and exchanges of fine sediments in the bay of Brest. La Houille Blanche 47-53. https://doi.org/10.1051/lhb/2014062
• 7. Bowden, G.J., Maier, H.R., Dandy, G.C., 2005. Input determination for neural network models in water resources applications. Part 2. Case study: forecasting salinity in a river. J. Hydrol. 301, 93-107. https://doi.org/10.1016/j.jhydrol.2004.06.020
• 8. Chauvaud, L., Jean, F., Ragueneau, O., Thouzeau, G., 2000. Longterm variation of the Bay of Brest ecosystem: benthic-pelagic coupling revisited. Mar. Ecol. Prog. Ser. 200, 35-48. https:// doi.org/10.3354/MEPS200035
• 9. Chen, L., Roy, S.B., Hutton, P.H., 2018. Emulation of a processbased estuarine hydrodynamic model. Hydrol. Sci. J. 63, 783- 802. https://doi.org/10.1080/02626667.2018.1447112
• 10. Chen, W., Liu, W., Huang, W., Liu, H., 2017. Prediction of Salinity Variations in a Tidal Estuary Using Artificial Neural Network and Three-Dimensional Hydrodynamic Models Comp. Water Energy Environ. Eng. 6 (1), 107-108.
• 11. Choi, K.W., Lee, J.H.W., 2004. Numerical determination of flushing time for stratified water bodies. J. Marine Syst. 50, 3-4. https://doi.org/10.1016/J.JMARSYS.2004.04.005
• 12. Chung, F.I., Seneviratne, S.A., 2009. Developing Artificial Neural Networks to Represent Salinity Intrusions in the Delta, in: World Environmental and Water Resources Congress 2009. In: Presented at the World Environmental and Water Resources Congress 2009. American Society of Civil Engineers, Kansas City, Missouri, United States, 1-10. https://doi.org/10.1061/41036(342)483
• 13. Cortes, C., Vapnik, V., 1995. Support-vector networks. Mach. Learn. 20, 273-297. https://doi.org/10.1007/BF00994018
• 14. Cruz, E.R., Nolasco, R., Padin, X.A., Gilcoto, M., Babarro, J.M.F., Dubert, J., Pérez, F.F., 2021. A High-Resolution Modeling Study of the Circulation Patterns at a Coastal Embayment: Ría de Pontevedra (NW Spain) Under Upwelling and Downwelling Conditions. Front. Mar. Sci. 8, 661250. https://doi.org/10.3389/fmars.2021.661250
• 15. Drucker, H., Burges, C.J.C., Kaufman, L., Smola, A., Vapnik, V., 1997. Support Vector Regression Machines. Neural Information Processing Systems 9, MIT Press, 155-161.
• 16. Ducrocq, V., Bouttier, F., Malardel, S., Montmerle, T., Seity, Y., 2005. Le projet AROME. La Houille Blanche 91 (2), 39-43. https://doi.org/10.1051/lhb:200502004
• 17. Dyer, K.R., 1973. Estuaries: A physical introduction DYER, K. R. 1973. Wiley-Interscience, New York, London, xv + 140 pp.
• 18. Frère, L., Paul-Pont, I., Rinnert, E., Petton, S., Jaffré, J., Bihanic, I., Soudant, P., Lambert, C., Huvet, A., 2017. Influence of environmental and anthropogenic factors on the composition, concentration and spatial distribution of microplastics: A case study of the Bay of Brest (Brittany, France). Environ. Pollut. 225, 211-222. https://doi.org/10.1016/j.envpol.2017.03.023
• 19. Guillou, N., 2007. Rôles de l’hétérogénéité des sédiments de fond et des interactions houle-courant sur l’hydrodynamique et la dynamique sédimentaire en zone subtidale - applications en Manche orientale et à la pointe de la Bretagne [WWW Document]. URL: https://www.calameo.com/books/001058329dee68c4a2d96 (accessed 11.3.21)
• 20. Guo, Q., Lordi, G.P., 2000. Method for quantifying freshwater input and flushing time in estuaries. J. Environ. Eng. 126, 675- 683.
• 21. He, M., Zhong, L., Sandhu, P., Zhou, Y., 2020. Emulation of a Process-Based Salinity Generator for the Sacramento—San Joaquin Delta of California via Deep Learning. Water 12, 2088. https://doi.org/10.3390/w12082088
• 22. Hsu, C.-W., Chang, C.-C., Lin, C.-J., 2010. A Practical Guide to Support Vector Classification, National Taiwan University Papers, Taipei, 16 pp.
• 23. Huang, W., Foo, S., 2002. Neural network modeling of salinity variation in Apalachicola River. Water Res. 36, 356-362. https://doi.org/10.1016/S0043-1354(01)00195-6
• 24. Keerthi, S.S., Lin, C.-J., 2003. Asymptotic behaviors of support vector machines with Gaussian kernel. Neural Comput 15, 1667- 1689. https://doi.org/10.1162/089976603321891855
• 25. Keras, 2021. Keras, Simple. Flexible. Powerful. https://keras.io (accessed on 09/2021).
• 26. Kingma, D.P., Ba, J., 2017. Adam: A Method for Stochastic Optimization. arXiv:1412.6980 [cs].
• 27. Lazure, P., Dumas, F., 2008. An external-internal mode coupling for a 3D hydrodynamical model for applications at regional scale (MARS). Adv. Water Resour. 31, 233-250. https://doi.org/10.1016/j.advwatres.2007.06.010
• 28. Le Pape, O., Del Amo, Y., Menesguen, A., Aminot, A., Quequiner, B., Tréguer, P., 1996. Resistance of a coastal ecosystem to increasing eutrophic conditions: the Bay of Brest (France), a semi-enclosed zone of Western Europe. Cont. Shelf Res. 16, 1885-1907.
• 29. Le Roy, R., Simon, B., 2003. Réalisation et validation d’un modèle de marée en Manche et dans le Golfe de Gascogne. Application à la réalisation d’un nouveau programme de réduction des sondages bathymétriques. (No. Rapport n°002/03). SHOM.
• 30. Maier, H.R., Dandy, G.C., 1996. The Use of Artificial Neural Networks for the Prediction of Water Quality Parameters. Water Resour. Res. 32, 1013-1022. https://doi.org/10.1029/96WR03529
• 31. Maier, H.R., Jain, A., Dandy, G.C., Sudheer, K.P., 2010. Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions. Environ. Modell. Softw. 25, 891-909. https://doi.org/10.1016/j.envsoft.2010.02.003
• 32. Nair, V., Hinton, G.E., 2010. Rectified Linear Units Improve Restricted Boltzmann Machines. In: Proceedings of the 27th International Conference on Machine Learning (ICML-10).
• 33. Nguyen, T.G., Tran, N.A., Vu, P.L., Nguyen, Q.-H., Nguyen, H.D., Bui, Q.-T., 2021. Salinity intrusion prediction using remote sensing and machine learning in data-limited regions: A case study in Vietnam’s Mekong Delta. Geoderma Regional 27, e00424. https://doi.org/10.1016/j.geodrs.2021.e00424
• 34. Petton, S., 2010. Etude des processus hydrodynamiques et hydro-sédimentaires affectant un estran de type marais salé de la rade de Brest (anse de Penfoul) colonisé par l’espèce invasive spartine (Spartina Alterniflora Loisel). In: Centre d’Etudes Techniques Maritimes et Fluviales, 37 pp.
• 35. Petton, S., Le Berre, D., Haurie, A., Pouvreau, S., 2016. HOMER Campaign : Mooring time series. https://doi.org/10.17882/43082
• 36. Petton, S., Le Roy, V., Bellec, G., Queau, I., Le Souchu, P., Pouvreau, S., 2018. Marine environmental station database of Daoulas bay. https://doi.org/10.17882/42493
• 37. Petton, S., Pouvreau, S., Dumas, F., 2020. Intensive use of Lagrangian trajectories to quantify coastal area dispersion. Ocean Dynam. 70, 541-559. https://doi.org/10.1007/s10236-019-01343-6
• 38. Poppeschi, C., Charria, G., Goberville, E., Rimmelin-Maury, P., Barrier, N., Petton, S., Unterberger, M., Grossteffan, E., Repecaud, M., Quéméner, L., Theetten, S., Le Roux, J.-F., Tréguer, P., 2021. Unraveling Salinity Extreme Events in Coastal Environments: A Winter Focus on the Bay of Brest. Front. Mar. Sci. 8, 966. https://doi.org/10.3389/fmars.2021.705403
• 39. Quéméneur, M., Kerouel, R., Aminot, A., 1984. Cycle de la matière organique dans l’estuaire de l’Elorn et relations avec les bactéries. Ifremer.
• 40. Rath, J.S., Hutton, P.H., Chen, L., Roy, S.B., 2017. A hybrid empirical-Bayesian artificial neural network model of salinity in the San Francisco Bay-Delta estuary. Environ. Modell. Softw. 93, 193-208. https://doi.org/10.1016/j.envsoft.2017.03.022
• 41. Robins, P.E., Lewis, M.J., Simpson, J.H., Howlett, E.R., Malham, S.K., 2014. Future variability of solute transport in a macrotidal estuary. Estuar. Coast. Shelf Sci. 151, 88-99. https://doi.org/10.1016/j.ecss.2014.09.019
• 42. Salomon, J.C., Breton, M., 1991. Numerical study of the dispersive capacity of the Bay of Brest, France, towards dissolved substances. Environ. Hydraul. 459-464.
• 43. SHOM, 2021. https://www.data.shom.fr (accessed on 05/2021).
• 44. Su, H., Wu, X., Yan, X.-H., Kidwell, A., 2015. Estimation of subsurface temperature anomaly in the Indian Ocean during recent global surface warming hiatus from satellite measurements: A support vector machine approach. Remote Sens. Environ. 160, 63-71. https://doi.org/10.1016/j.rse.2015.01.001
• 45. Tréguer, P., Goberville, E., Barrier, N., L’Helguen, S., Morin, P., Bozec, Y., Rimmelin-Maury, P., Czamanski, M., Grossteffan, E., Cariou, T., Répécaud, M., Quéméner, L., 2014. Large and local-scale influences on physical and chemical characteristics of coastal waters of Western Europe during winter. J. Marine Syst. 139, 79-90.
• 46. Vapnik, V., Golowich, S.E., Smola, A., 1996. Support vector method for function approximation, regression estimation and signal processing. In: Proceedings of the 9th International Conference on Neural Information Processing Systems, Guide Proceedings [WWW Document]. URL: https://dl.acm.org/doi/abs/10.5555/2998981.2999021 (accessed 3.7.22).
• 47. Vapnik, V.N., 1995. The Nature of Statistical Learning Theory. Springer, New York, USA, 189 pp.
• 48. WMO, 2021. World Meteorological Organization. Données d’observation des principales stations météorologiques URL: https://www.data.gouv.fr/fr/datasets/donnees-d-observationdes-principales-stations-meteorologiques (accessed on 05/2021).
• 49. Zhang, H., Shen, Y., Tang, J., 2021. Hydrodynamics and water renewal in the Pearl River Estuary, China: A numerical study from the perspective of water age. Ocean Eng. 237, 109639. https://doi.org/10.1016/j.oceaneng.2021.109639

Uwagi

Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2022-2023). (PL)