Advertisement

Hydrogeology Journal

, Volume 27, Issue 7, pp 2535–2550 | Cite as

Identification of transient contaminant sources in aquifers through a surrogate model based on a modified self-organizing-maps algorithm

  • Xuemin Xia
  • Nianqing Zhou
  • Lichun Wang
  • Xianwen Li
  • Simin JiangEmail author
Paper
  • 98 Downloads

Abstract

The identification of transient groundwater contaminant sources in terms of source locations, contaminant magnitudes, and active durations remains a challenge. The problem becomes more complex due to spatial heterogeneity, sparse observation data, concentration measurement errors, and unexpected uncertainty. This study addresses this challenge by proposing a modified self-organizing maps (SOM) algorithm; this algorithm can improve the physically-based models by reducing the computational burden more efficiently. The method sufficiently increases the accuracy and efficiency for identifying the contaminant source, because the trained SOM-based surrogate models can identify the source characteristics independently without necessarily operating a formal linked simulation-optimization model. The performance of the proposed method was assessed on a hypothetical heterogeneous aquifer model; the assessment considered unknown observation data, concentration measurement errors, and an unknown pumping well. The proposed SOM-based surrogate model can not only approximate the results from the groundwater flow and transport simulation models, but it can also be used in lieu of the optimization model in a more efficient way for identifying the unknown transient contaminant sources in groundwater systems.

Keywords

Self-organizing maps Surrogate model Contamination Inverse modeling Unexpected uncertainty 

Identification de sources transitoires de contamination dans les aquifères par un modèle de substitut basé sur un algorithme modifié de cartes d’organisation automatique

Résumé

L’identification de sources de contamination transitoires des eaux souterraines en termes d’emplacements des sources, de quantité de contaminant, et de durées actives reste un défi. Le problème devient plus complexe du fait de l’hétérogénéité spatiale, de données d’observations éparses, d’erreurs de mesure des concentrations, et de l’incertitude inattendue. Cette étude adresse ce défi en proposant un algorithme modifié de cartes d’organisation automatique (COA); cet algorithme peut améliorer les modèles basés sur la physique en réduisant plus efficacement la demande de calcul. La méthode augmente suffisamment la précision et l’efficacité pour identifier la source de contaminant, parce que les modèles de substitut entrainés basés sur les COA peuvent identifier les caractéristiques des sources indépendamment sans nécessairement utiliser un lien formel pour le modèle simulation-optimisation. La performance de la méthode proposée a été évaluée sur un modèle d’aquifère hétérogène hypothétique; l’évaluation a considéré des données observées inconnues, des erreurs sur les mesures de concentrations, et un puits de pompage inconnu. Le modèle de substitut basé sur les COA qui est proposé peut non seulement approximer les résultats de modèles de simulations d’écoulements des eaux souterraines et de transport, mais il peut aussi être utilisé à la place d’un modèle d’optimisation d’une manière plus efficace pour identifier les sources transitoires de contaminants, non connues, dans des systèmes hydrogéologiques.

Identificación de fuentes de contaminantes transitorias en acuíferos mediante un Modelo sustituto basado en un algoritmo modificado de mapas autoorganizados

Resumen

La identificación de las fuentes transitorias de contaminantes del agua subterránea en términos de ubicación de las fuentes, magnitudes de los contaminantes y duraciones activas sigue siendo un desafío. El problema se vuelve más complejo debido a la heterogeneidad espacial, la escasez de datos de observación, los errores de medición de la concentración y la incertidumbre imprevista. Este estudio aborda este desafío proponiendo un algoritmo modificado de mapas auto-organizados (SOM); este algoritmo puede mejorar los modelos basados físicamente al reducir la carga computacional de manera más eficiente. El método aumenta suficientemente la precisión y eficiencia para identificar la fuente del contaminante, ya que los modelos sustitutos basados en SOM pueden identificar las características de la fuente de manera independiente sin necesidad de operar un modelo formal de simulación-optimización. El rendimiento del método propuesto se evaluó sobre la base de un hipotético modelo de acuífero heterogéneo; la evaluación consideró datos de observación desconocidos, errores de medición de la concentración y un pozo de bombeo desconocido. El modelo propuesto, basado en SOM, no sólo puede aproximar los resultados de los modelos de simulación de flujo de agua subterránea y transporte, sino que también puede ser utilizado en lugar del modelo de optimización de una manera más eficiente para identificar las fuentes de contaminantes transitorios desconocidos en los sistemas de agua subterránea.

通过改进的自组织映射算法的替代模型识别含水层中的瞬态污染源

摘要

根据污染源位置,大小和污染持续时间识别瞬态地下水污染源仍然是一个挑战。由于空间非均质性,稀少的观测数据,浓度测量误差和不可预期的不确定性,识别问题变得更加复杂。本研究提出改进的自组织映射(SOM)算法来解决这一问题。该算法可通过有效减少计算负荷来改进基于物理机制的模型。因为经过训练的基于SOM的替代模型可以独立地识别源特征而不必运行已链接好的模拟优化模型,该方法充分提高了识别污染源的准确性和效率。通过假想的非均质含水层模型,论文评估了该方法的性能;该评估考虑了未知的观测数据,浓度测量误差和未知的抽水井。所提出的基于SOM的替代模型不仅可以接近地下水流和运移模拟模型的结果,而且还可以替代优化模型,以更有效的方式识别地下水系统中未知的瞬态污染源。

Identificação de fontes contaminantes transientes em aquíferos por meio de um modelo substituto baseado em mapas auto organizáveis modificados

Resumo

A identificação de fontes contaminantes transientes em termos de localização de fontes, magnitude de contaminantes, e duração ativa continua sendo um desafio. O problema se torna mais complexo devido a heterogeneidade espacial, escassez de dados de observação, erros de medidas de concentração, e incertezas inesperadas. Este estudo aborda este desafio através da proposta de um algoritmo de mapas auto organizáveis (MAO) modificados; este algoritmo pode aprimorar modelos baseados em aspectos físicos através da redução da carga computacional de forma mais eficiente. O método aumenta suficientemente a precisão e a eficiência para a identificação da fonte contaminante, pois os modelos substitutos baseados em MAO treinados podem identificar características da fonte independentemente, sem a necessidade de operar um modelo formal de simulação-otimização. O desempenho do método proposto foi avaliado com um modelo hipotético de aquífero heterogêneo; a avaliação considerou dados de observação desconhecidos, erros de medida de concentração, e um poço de bombeamento desconhecido. O modelo substituto baseado em MAO proposto não apenas é capaz de aproximar os resultados de simulações de modelos fluxo e transporte das águas subterrâneas, como também ser usado no lugar de modelos de otimização de uma forma mais eficiente, a fim de identificar fontes contaminantes transientes desconhecidas em sistemas aquíferos.

Notes

Acknowledgements

This work was supported by the National Key Research and Development Program of China (KZ0023020171537), the Fundamental Research Funds for the Central Universities (No. 22120190013), and National Natural Science Foundation of China (41502225). The authors would like to thank the editor and three anonymous reviewers and A.C. Bagtzoglou for their constructive and valuable comments and suggestions, which significantly improved the quality of this work.

References

  1. Ababou R, Bagtzoglou AC, Mallet A (2010) Anti-diffusion and source identification with the RAW scheme: a particle-based censored random walk. Environ Fluid Mech 10(1–2):41–76.  https://doi.org/10.1007/s10652-009-9153-4 CrossRefGoogle Scholar
  2. Amirabdollahian M, Datta B (2013) Identification of contaminant source characteristics and monitoring network design in groundwater aquifers: an overview. J Environ Prot 04(5):26–41.  https://doi.org/10.4236/jep.2013.45A004 CrossRefGoogle Scholar
  3. Andrade L, O’Dwyer J, O’Neill E, Hynds P (2018) Surface water flooding, groundwater contamination, and enteric disease in developed countries: a scoping review of connections and consequences. Environ Pollut 236:540–549.  https://doi.org/10.1016/j.envpol.2018.01.104 CrossRefGoogle Scholar
  4. Asher MJ, Croke BFW, Jakeman AJ, Peeters LJM (2015) A review of surrogate models and their application to groundwater modeling. Water Resour Res 51(8):5957–5973.  https://doi.org/10.1002/2015wr016967 CrossRefGoogle Scholar
  5. Ayvaz MT (2010) A linked simulation–optimization model for solving the unknown groundwater pollution source identification problems. J Contam Hydrol 117(1–4):46–59.  https://doi.org/10.1016/j.jconhyd.2010.06.004 CrossRefGoogle Scholar
  6. Ayvaz MT (2016) A hybrid simulation–optimization approach for solving the areal groundwater pollution source identification problems. J Hydrol 538:161–176.  https://doi.org/10.1016/j.jhydrol.2016.04.008 CrossRefGoogle Scholar
  7. Bagtzoglou AC, Atmadja J (2005) Mathematical methods for hydrologic inversion: the case of pollution source identification. Springer, Heidelberg, Germany, pp 65–96Google Scholar
  8. Bagtzoglou AC, Baun SA (2005) Near real-time atmospheric contamination source identification by an optimization-based inverse method. Inverse Problems Sci Eng 13(3):241–259.  https://doi.org/10.1080/10682760412331330163 CrossRefGoogle Scholar
  9. Guneshwor L, Eldho TI, Kumar AV (2018) Identification of groundwater contamination sources using Meshfree RPCM simulation and particle swarm optimization. Water Resour Manag 32(4):1517–1538.  https://doi.org/10.1007/s11269-017-1885-1 CrossRefGoogle Scholar
  10. Gurarslan G, Karahan H (2015) Solving inverse problems of groundwater-pollution-source identification using a differential evolution algorithm. Hydrogeol J 23(5):1–11.  https://doi.org/10.1007/s10040-015-1256-z CrossRefGoogle Scholar
  11. Hazrati YS, Datta B (2017a) Self-organizing map based surrogate models for contaminant source identification under parameter uncertainty. Int J Geomate 13(36):11–18.  https://doi.org/10.21660/2017.36.2750 CrossRefGoogle Scholar
  12. Hazrati YS, Datta B (2017b) Adaptive surrogate model based optimization (ASMBO) for unknown groundwater contaminant source characterizations using self-organizing maps. J Water Resour Protect 09(2):193–214.  https://doi.org/10.4236/jwarp.2017.92014 CrossRefGoogle Scholar
  13. Jiang S, Zhu G, Shi X, Zhou N (2010) Inverse analysis of hydrogeological parameters using hybrid Hooke-Jeeves and particle swarm optimization method (in Chinese). Adv Water Sci 21(5):606–612Google Scholar
  14. Jiang S, Zhang Y, Wang P, Zheng M (2013) An almost-parameter-free harmony search algorithm for groundwater pollution source identification. Water Sci Technol 68(11):2359–2366.  https://doi.org/10.2166/wst.2013.499 CrossRefGoogle Scholar
  15. Kalteh AM, Hjorth P, Berndtsson R (2008) Review of the self-organizing map (SOM) approach in water resources: analysis, modelling and application. Elsevier, Amsterdam, pp 835–845CrossRefGoogle Scholar
  16. Kohonen T (1982) Analysis of a simple self-organizing process. Biol Cybern 44(2):135–140.  https://doi.org/10.1007/bf00317973 CrossRefGoogle Scholar
  17. Kohonen T, Oja E, Simula O, Visa A (1996) Engineering applications of the self-organizing map. Proc IEEE 84(10):1358–1384.  https://doi.org/10.1109/5.537105 CrossRefGoogle Scholar
  18. Ley R, Casper MC, Hellebrand H, Merz R (2011) Catchment classification by runoff behaviour with self-organizing maps (SOM). Hydrol Earth Syst Sci 15(9):2947–2962.  https://doi.org/10.5194/hessd-8-3047-2011 CrossRefGoogle Scholar
  19. Mahdavi M, Fesanghary M, Damangir E (2007) An improved harmony search algorithm for solving optimization problems. Appl Math Comput 188(2):1567–1579.  https://doi.org/10.1016/j.amc.2006.11.033 CrossRefGoogle Scholar
  20. Mategaonkar M, Eldho TI (2012) Groundwater remediation optimization using a point collocation method and particle swarm optimization. Environ Model Softw 32(3):37–48.  https://doi.org/10.1016/j.envsoft.2012.01.003 CrossRefGoogle Scholar
  21. McDonald MGM, Harbaugh AW (1988) MODFLOW, a modular three-dimensional finite difference ground-water flow model. US Geol Surv Open-File Rep 83-875, Chapter A1Google Scholar
  22. Mirghani BY, Mahinthakumar KG, Tryby ME, Ranjithan RS, Zechman EM (2009) A parallel evolutionary strategy based simulation–optimization approach for solving groundwater source identification problems. Adv Water Resour 32(9):1373–1385.  https://doi.org/10.1016/j.advwatres.2009.06.001 CrossRefGoogle Scholar
  23. Nguyen TT, Kawamura A, Tong TN, Nakagawa N, Amaguchi H, Gilbuena R (2015) Clustering spatio–seasonal hydrogeochemical data using self-organizing maps for groundwater quality assessment in the Red River Delta, Vietnam. J Hydrol 522:661–673.  https://doi.org/10.1016/j.jhydrol.2015.01.023 CrossRefGoogle Scholar
  24. Prakash O, Datta B (2015) Optimal characterization of pollutant sources in contaminated aquifers by integrating sequential-monitoring-network design and source identification: methodology and an application in Australia. Hydrogeol J 23(6):1089–1107.  https://doi.org/10.1007/s10040-015-1292-8 CrossRefGoogle Scholar
  25. Sayeed M (2005) Hybrid genetic algorithm: local search methods for solving groundwater source identification inverse problems. J Water Res Plann Manag 131(1):45–57.  https://doi.org/10.1061/(asce)0733-9496(2005)131:1(45) CrossRefGoogle Scholar
  26. Simula O, Vesanto J, Alhoniemi E, Hollmen J (1999) Analysis and modeling of complex systems using the self-organizing map. Helsinki University of Technology, Finland, pp 1–16, http://users.ics.aalto.fi/jhollmen/Publications/nftt.pdf. Accessed June 2019
  27. Sreekanth J, Datta B (2011) Coupled simulation-optimization model for coastal aquifer management using genetic programming-based ensemble surrogate models and multiple-realization optimization. Water Resour Res 47(4):158–166.  https://doi.org/10.1029/2010wr009683 CrossRefGoogle Scholar
  28. Srivastava D, Singh RM (2014) Breakthrough curves characterization and identification of an unknown pollution source in groundwater system using an artificial neural network (ANN). Environ Forensic 15(2):175–189.  https://doi.org/10.1080/15275922.2014.890142 CrossRefGoogle Scholar
  29. Thi HAL, Nguyen MC (2014) Self-organizing maps by difference of convex functions optimization. Data Min Knowl Disc 28(5–6):1336–1365.  https://doi.org/10.1007/s10618-014-0369-7 CrossRefGoogle Scholar
  30. Vatanen T, Osmala M, Raiko T, Lagus K, Sysi-Aho M, Orešič M, Honkela T, Lähdesmäki H (2015) Self-organization and missing values in SOM and GTM. Neurocomputing 147:60–70.  https://doi.org/10.1016/j.neucom.2014.02.061 CrossRefGoogle Scholar
  31. Zhang D, Lu Z (2004) An efficient, high-order perturbation approach for flow in random porous media via Karhunen-Loève and polynomial expansions. J Comput Phys 194(2):773–794.  https://doi.org/10.1016/j.jcp.2003.09.015 CrossRefGoogle Scholar
  32. Zhao Y, Lu W, Xiao C (2016) A kriging surrogate model coupled in simulation–optimization approach for identifying release history of groundwater sources. J Contam Hydrol 185-186(Pt2):51–60.  https://doi.org/10.1016/j.jconhyd.2016.01.004 CrossRefGoogle Scholar
  33. Zheng C, Wang PP (1999) MT3DMS: a modular three-dimensional multispecies transport model for simulation of advection, dispersion, and chemical reactions of contaminants in groundwater systems; documentation and user’s guide. https://www.researchgate.net/publication/242586434_MT3DMS_A_Modular_Three-Dimensional_Multispecies_Transport_Model_for_Simulation_of_Advection_Dispersion_and_Chemical_Reactions_of_Contaminants_in_Groundwater_Systems_Documentation_and_User's_Guide. Accessed June 2019

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  • Xuemin Xia
    • 1
  • Nianqing Zhou
    • 1
    • 2
  • Lichun Wang
    • 3
  • Xianwen Li
    • 4
  • Simin Jiang
    • 1
    Email author
  1. 1.Department of Hydraulic EngineeringTongji UniversityShanghaiChina
  2. 2.Key Laboratory of Yangtze River Water EnvironmentMinistry of EducationShanghaiChina
  3. 3.Institute of Surface-Earth System ScienceTianjin UniversityTianjinChina
  4. 4.Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas, Ministry of EducationNorthwest A&F UniversityXianyangChina

Personalised recommendations