Advertisement

Hydrogeology Journal

, Volume 26, Issue 4, pp 1099–1115 | Cite as

Application of statistical classification methods for predicting the acceptability of well-water quality

  • Enrico Cameron
  • Giorgio Pilla
  • Fabio A. Stella
Paper

Abstract

The application of statistical classification methods is investigated—in comparison also to spatial interpolation methods—for predicting the acceptability of well-water quality in a situation where an effective quantitative model of the hydrogeological system under consideration cannot be developed. In the example area in northern Italy, in particular, the aquifer is locally affected by saline water and the concentration of chloride is the main indicator of both saltwater occurrence and groundwater quality. The goal is to predict if the chloride concentration in a water well will exceed the allowable concentration so that the water is unfit for the intended use. A statistical classification algorithm achieved the best predictive performances and the results of the study show that statistical classification methods provide further tools for dealing with groundwater quality problems concerning hydrogeological systems that are too difficult to describe analytically or to simulate effectively.

Keywords

Well Contamination Groundwater quality Machine learning Statistical classification 

Application de méthodes de classification statistique pour prévoir l’acceptabilité de la qualité de l’eau issue de forages

Résumé

L’application de méthodes de classification statistique est étudiée—en comparant également avec les méthodes d’interpolation spatiale—pour prédire l’acceptabilité de la qualité de l’eau issue de forages, dans une situation où un modèle quantitatif efficace d’un système hydrogéologique considéré ne peut être développé. Dans la zone prise en exemple, au nord de l’Italie, l’aquifère est. localement affecté par une eau saline, et la concentration en chlorures est. le principal indicateur de la présence d’eau salée et de la qualité des eaux souterraines. L’objectif est de prédire si la concentration en chlorures de l’eau issue d’un forage est supérieure à la valeur autorisée, de sorte que l’eau n’est pas conforme à l’usage souhaité. Un algorithme de classification statistique a permis d’obtenir les meilleures performances de prévision et les résultats de cette étude montrent que les méthodes de classification statistique fournissent des outils plus poussés pour appréhender les problèmes de qualité des eaux souterraines, pour les systèmes hydrogéologiques trop difficiles à décrire de manière analytique ou à simuler de manière efficace.

Aplicación de métodos de clasificación estadística para predecir la aceptabilidad de la calidad del agua de pozos

Resumen

Se investiga la aplicación de métodos de clasificación estadística, en comparación también con los métodos de interpolación espacial, para predecir la aceptabilidad de la calidad del agua de pozos en una situación donde no se puede desarrollar un modelo cuantitativo eficaz del sistema hidrogeológico considerado. En el área del ejemplo, en particular en el norte de Italia, el acuífero se ve afectado localmente por el agua salina y la concentración de cloruro es el principal indicador tanto de la ocurrencia de agua salada como de la calidad del agua subterránea. El objetivo es predecir si la concentración de cloruro en un pozo de agua excederá la concentración permitida de modo que el agua no sea apta para el uso previsto. Un algoritmo de clasificación estadística logró los mejores resultados predictivos y los resultados del estudio muestran que los métodos de clasificación estadística proporcionan más herramientas para tratar los problemas de calidad del agua subterránea en relación con sistemas hidrogeológicos que son demasiado difíciles de describir analíticamente o de simularlos eficazmente.

应用统计分类方法预测井水水质的可接受性

摘要

调查了统计分类方法的应用情况—还与空间插入方法进行了比较—以预测无法建立水文地质系统有效定量模型的情况下水井水质的可接受性。特别是在意大利北部的研究案例区,含水层局部受到咸水的影响,氯化物的含量是出现盐水和地下水水质的主要指示物。目的就是预测水井中的氯化物含量是否超过允许的含量而使水不能使用。统计分类算法预测结果最好,研究结果显示,统计分类方法为处理很难解析描述或有效模拟的水文地质系统地下水水质问题提供了进一步的工具。

Utilização de métodos de classificação estatística para previsão de aceitabilidade de qualidade da água dos poços

Resumo

A utilização de métodos de classificação estatística é investigada—em comparação também aos métodos de interpolação espacial—para prever a aceitabilidade da qualidade de água de poços em uma situação onde um modelo quantitativo efetivo do sistema hidrogeológico sob consideração não pode ser desenvolvido. Na área piloto no norte da Itália, em particular, o aquífero é localmente afetado por água salina e a concentração de cloreto é o principal indicador de ocorrência de água salgada e qualidade das águas subterrâneas. O objetivo é prever se a concentração de cloreto em um poço de abastecimento excederá a concentração permitida, assim a água não se adequaria ao uso pretendido. Um algoritmo de classificação estatística alcançou os melhores desempenhos de previsão e os resultados do estudo demonstram que os métodos de classificação estatística fornecem ferramentas adicionais para lidar com os problemas da qualidade das águas subterrâneas no viés dos sistemas hidrogeológicos que são muito difíceis de se descrever analiticamente ou se simular efetivamente.

Notes

Acknowledgements

The authors wish to thank Dr. Jean-Michel Lemieux and two anonymous reviewers for their suggestions for improving the paper.

References

  1. AGIP (1994) Acque dolci sotterranee: inventario dei dati raccolti dall’AGIP durante la ricerca di idrocarburi in Italia (dal 1971 al 1990) [Underground freshwater: inventory of the data collected by AGIP during hydrocarbon exploration in Italy (from 1971 to 1990)]. AGIP, Milan, ItalyGoogle Scholar
  2. Bárdossy A, Giese H, Grimm-Strele J (1999) Interpolation of groundwater quality parameters using geological and land use classification. In: Gómez-Hernández JJ, Soares AO, Froidevaux R (eds) geoENV II: geostatistics for environmental applications. Springer, Dordrecht, The Netherlands, pp 247–258CrossRefGoogle Scholar
  3. Barzegar R, Asghar Moghaddam A, Adamowski J, Fijani E (2016) Comparison of machine learning models for predicting fluoride contamination in groundwater. Stoc Environ Res Risk Assess 30(2016):1–14.  https://doi.org/10.1007/s00477-016-1338-z Google Scholar
  4. Bersan M, Pilla G, Dolza G, Torrese P, Ciancetti G (2010) The uprising of deep saline waters into the Oltrepò Pavese (northern Italy) aquifer: early results. Italian J Eng Geol Environ 1:7–22.  https://doi.org/10.4408/IJEGE.2010-01.O-01 Google Scholar
  5. Boni A (1967) Note illustrative della Carta Geologica d’Italia F. 59 Pavia [Illustrative notes of the Italian Geological Map F Sheet 59 Pavia]. Stabilimento L. Salomone, RomeGoogle Scholar
  6. Braga G, Cerro A (1988) Le strutture sepolte della pianura pavese e le relative influenze sulle risorse idriche sotterranee [The buried structures of the Pavia alluvial plain and their influences on the ground water resources (south western Lombardy, Italy)]. Atti Ticinensi Sci Terra 31:421–433Google Scholar
  7. Bramer M (2013) Principles of data mining. Springer, LondonCrossRefGoogle Scholar
  8. Conti A, Sacchi E, Chiarle M, Martinelli G, Zuppi GM (2000) Geochemistry of the formation water of the Po plain (northern Italy): an overview. Appl Geochem 15:51–65.  https://doi.org/10.1016/S0883-2927(99)00016-5 CrossRefGoogle Scholar
  9. Cleary JG, Trigg L (1995) K*: an instance-based learner using an entropic distance measure. Proc Machine Learning Conference, Tahoe City, CA, 1995, pp 108–114. http://www.cs.waikato.ac.nz/ml/publications/1995/Cleary95-KStar.pdf. Accessed 11th April 2017
  10. Eberly D (2018) Thin plate splines. http://www.geometrictools.com/Documentation/ThinPlateSplines.pdf. Accessed 25th January 2018
  11. Isaaks EH, Mohan Srivastava R (1989) An introduction to applied geostatistics. Oxford University Press, New YorkGoogle Scholar
  12. Khalil A, Almasri MN, McKee M, Kaluarachchi JJ (2005) Applicability of statistical learning algorithms in groundwater quality modeling. Water Resour Res 41:W05010.  https://doi.org/10.1029/2004WR003608 Google Scholar
  13. Kovarik K (2000) Numerical models in groundwater pollution. Springer, Heidelberg, GermanyGoogle Scholar
  14. Li J Heap AD (2008) A review of spatial interpolation methods for environmental scientists. Record 2008/23. Geoscience Australia, Canberra, Australia. http://corpdata.s3.amazonaws.com/68229/Rec2008_023.pdf. Accessed 11th April 2017
  15. Liu J, Chang M, Ma X (2009) Groundwater quality assessment based on support vector machine. In: Zhang H, Zhao R, Zhao H (eds) 2nd International Symposium of HAIHE Basin Integrated Water and Environment Management 2009. Aussino, Riverwood, Australia. http://www.seiofbluemountain.com/upload/product/201005/2009shzyhy03a1.pdf. Accessed 11 April 2017, pp 167–173
  16. Nolan BT, Fienen MN, Lorenz DL (2015) A statistical learning framework for groundwater nitrate models of the Central Valley, California, USA. J Hydrol 531(3):902–911.  https://doi.org/10.1016/j.jhydrol.2015.10.025 CrossRefGoogle Scholar
  17. Pellegrini L, Vercesi PL (1995) Considerazioni morfotettoniche sulla zona a sud del Po tra Voghera (PV) e Sarmato (PC) [Considerations about the morphotectonic setting of the area located south of the River Po between Voghera (PV) and Sarmato (PC)]. Atti Ticinensi Sci Terra 38:95–118Google Scholar
  18. Pilla G, Sacchi E, Ciancetti G (2007a) Studio idrogeologico, idrochimico ed isotopico delle acque sotterranee del settore di pianura dell’Oltrepò Pavese (Pianura lombarda meridionale) [Hydrogeological, hydrochemical and isotopic study of the underground water in the Oltrepò Pavese plain (southern Lombardy plain)]. Giornale Geol Appl 5:59–74.  https://doi.org/10.1474/GGA.2007-05.0-05-0167 Google Scholar
  19. Pilla G, Sacchi E, Ciancetti G (2007b) Hydrochemical and isotopic groundwater investigation in the Oltrepo region (Po valley, northern Italy). In: IAEA proceedings series 2, pp 49–58. http://www-pub.iaea.org/MTCD/publications/PDF/Pub1310Vol2_web.pdf. Accessed 11 April 2017
  20. Pilla G, Torrese P, Bersan M (2010) Application of hydrochemical and preliminary geophysical surveys within the study of the saltwater uprising occurring in the Oltrepò Pavese plain aquifer. Boll Geofis Teor Appl 51(4):301–323Google Scholar
  21. Pilla G, Torrese P, Bersan M (2015) The uprising of deep saline Paleo-waters into the Oltrepò Pavese aquifer (northern Italy): application of hydro-chemical and shallow geophysical surveys. In: Lollino G, Arattano M, Rinaldi M, Giustolisi O, Marechal JC, Grant GE (eds) Engineering geology for society and territory, vol 3. Springer, Cham, Switzerland, pp 393–397Google Scholar
  22. Regione Lombardia, Eni Divisione AGIP (2002) Geologia degli acquiferi padani della Regione Lombardia [Geology of the Padanian plain aquifers of Lombardy]. S.EL.CA., Firenze, ItalyGoogle Scholar
  23. Saghebian Medi S, Taghi Sattari M, Mirabbasi R, Pal M (2014) Groundwater quality classification by decision tree method in Ardebil region, Iran. Arab J Geosci 7(11):4767–4777CrossRefGoogle Scholar
  24. Shepard D (1968) A two-dimensional interpolation function for irregularly spaced data. In: Proceedings of the 1968 23rd ACM National Conference, pp 517–523Google Scholar
  25. Sibson R (1981) A brief description of natural neighbour interpolation. In: Barnett V (ed) Interpreting multivariate data. Wiley, Chichester, UK, pp 21–36Google Scholar
  26. Sun N (1996) Mathematical modeling of groundwater pollution. Springer, New YorkCrossRefGoogle Scholar
  27. UNEP/DEWA DFID (2003) Groundwater and its susceptibility to degradation: a global assessment of the problem and options for management. http://wedocs.unep.org//handle/20.500.11822/8035. Accessed 11 April 2017
  28. Van der Perk M (2013) Soil and water contamination, 2nd edn. CRC, Boca Raton, FLGoogle Scholar
  29. Witten IH, Frank E, Mark AH (2011) Data mining: practical machine learning tools and techniques, 3rd edn. Morgan Kaufmann, Burlington, MAGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  • Enrico Cameron
    • 1
  • Giorgio Pilla
    • 2
  • Fabio A. Stella
    • 3
  1. 1.GeoStudioMorbegnoItaly
  2. 2.Department of Earth and Environmental SciencesUniversity of PaviaPaviaItaly
  3. 3.Department of Informatics, System Science and CommunicationUniversity of Milano BicoccaMilanItaly

Personalised recommendations