Advertisement

Mathematical Geosciences

, Volume 44, Issue 6, pp 739–763 | Cite as

Estimation of Hydraulic Conductivity and Its Uncertainty from Grain-Size Data Using GLUE and Artificial Neural Networks

  • Bart Rogiers
  • Dirk Mallants
  • Okke Batelaan
  • Matej Gedeon
  • Marijke Huysmans
  • Alain Dassargues
Article

Abstract

Various approaches exist to relate saturated hydraulic conductivity (K s) to grain-size data. Most methods use a single grain-size parameter and hence omit the information encompassed by the entire grain-size distribution. This study compares two data-driven modelling methods—multiple linear regression and artificial neural networks—that use the entire grain-size distribution data as input for K s prediction. Besides the predictive capacity of the methods, the uncertainty associated with the model predictions is also evaluated, since such information is important for stochastic groundwater flow and contaminant transport modelling.

Artificial neural networks (ANNs) are combined with a generalised likelihood uncertainty estimation (GLUE) approach to predict K s from grain-size data. The resulting GLUE-ANN hydraulic conductivity predictions and associated uncertainty estimates are compared with those obtained from the multiple linear regression models by a leave-one-out cross-validation. The GLUE-ANN ensemble prediction proved to be slightly better than multiple linear regression. The prediction uncertainty, however, was reduced by half an order of magnitude on average, and decreased at most by an order of magnitude. This demonstrates that the proposed method outperforms classical data-driven modelling techniques. Moreover, a comparison with methods from the literature demonstrates the importance of site-specific calibration. The data set used for this purpose originates mainly from unconsolidated sandy sediments of the Neogene aquifer, northern Belgium. The proposed predictive models are developed for 173 grain-size K s-pairs. Finally, an application with the optimised models is presented for a borehole lacking K s data.

Keywords

Early stopping Cross-validation Generalised likelihood uncertainty estimation Artificial neural networks Sedimentary aquifer Data-driven modelling Likelihood measures Principal component analysis GLUE-ANN 

Notes

Acknowledgements

The authors are grateful to ONDRAF/NIRAS, the Belgian Agency for Radioactive Waste and Enriched Fissile Materials, for providing the data. Findings and conclusions in this paper are those of the authors and do not necessarily represent the official position of ONDRAF/NIRAS.

References

  1. Agyare WA, Park SJ, Vlek PL (2007) Artificial neural network estimation of saturated hydraulic conductivity. Vadose Zone J 6(2):423–431 CrossRefGoogle Scholar
  2. Asefa T (2009) Ensemble streamflow forecast: A GLUE-based neural network approach. J Am Water Resour Assoc 45(5):1155–1163 CrossRefGoogle Scholar
  3. Basheer IA, Hajmeer M (2000) Artificial neural networks: Fundamentals computing design and application. J Microbiol Methods 43(1):3–31 CrossRefGoogle Scholar
  4. Beerten K, Wemaere I, Gedeon M, Labat S, Rogiers B, Mallants D, Salah S, Leterme B (2010) Geological hydrogeological and hydrological data for the Dessel disposal site Project near surface disposal of category A waste at Dessel—Version 1, NIROND-TR 2009-05 E, 261 Google Scholar
  5. Berry M, Linoff G (1997) Data mining techniques for marketing sales and customer support. Wiley, New York, 464 pp Google Scholar
  6. Beven K, Freer J (2001) Equifinality data assimilation and uncertainty estimation in mechanistic modelling of complex environmental systems using the GLUE methodology. J Hydrol 249(1–4):11–29 CrossRefGoogle Scholar
  7. Carman PC (1938) The determination of the specific surface of powders. J Soc Chem Ind Trans 57:225–234 Google Scholar
  8. Carman PC (1956) Flow of gases through porous media. Butterworths Scientific Publications, London Google Scholar
  9. Carrier WD (2003) Goodbye Hazen; hello Kozeny–Carman. Journal of Geotechnical and Geoenvironmental Engineering 129(11):1054 CrossRefGoogle Scholar
  10. Castejón Limas M, Ordieres Meré JB, González Marcos A, de Pisón Ascacibar FJM, Pernía Espinoza AV, Alba Elías F (2010) AMORE: A MORE flexible neural network package. R package version 0.2-12. http://CRAN.R-project.org/package=AMORE
  11. Coppola EA, Rana AJ, Poulton MM, Szidarovszky F, Uhl VW (2005) A neural network model for predicting aquifer water level elevations. Ground Water 43(2):231–241 CrossRefGoogle Scholar
  12. Cronican A, Gribb M (2004) Literature review: Equations for predicting hydraulic conductivity based on grain-size data. Supplement to technical note entitled: Hydraulic conductivity prediction for sandy soils. Ground Water 42(3):459–464 CrossRefGoogle Scholar
  13. Darcy H (1856) Les Fontaines Publiques de la Ville de Dijon. Dalmont, Paris Google Scholar
  14. Fletcher L, Katkovnik V, Steffens F, Engelbrecht A (1998) Optimizing the number of hidden nodes of a feedforward artificial neural network. In: Neural networks proceedings IEEE world congress on computational intelligence, the 1998 IEEE international joint conference, vol 2, pp 1608–1612 Google Scholar
  15. Gunst FR, Mason LR (1980) Regression analysis and its applications: A data oriented approach. Dekker, New York, 402 pp Google Scholar
  16. Haykin S (1999) Neural networks—A comprehensive foundation, 2nd edn. Prentice Hall, New Jersey Google Scholar
  17. Hazen A (1892) Some physical properties of sands and gravels. Massachusetts State Board of Health Annual Report 539-556 Google Scholar
  18. Henseler J (1995) Back propagation. In: Braspenning PJ, Thuijsman F, Weijters AJMM (eds) Artificial neural networks; an introduction to ANN theory and practice, vol 10. Springer, Berlin, pp 37–66 Google Scholar
  19. Hill T, Lewicki P (2007) STATISTICS: Methods and applications. StatSoft, Tulsa Google Scholar
  20. Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2(5):359–366 CrossRefGoogle Scholar
  21. Hornik K (1991) Approximation capabilities of multilayer feedforward networks. Neural Netw 4(2):251–257 CrossRefGoogle Scholar
  22. Jain A, Kumar AM (2007) Hybrid neural network models for hydrologic time series forecasting. Appl Soft Comput 7(2):585–592 CrossRefGoogle Scholar
  23. Joorabchi A, Zhang H, Blumenstein M (2009) Application of artificial neural networks to groundwater dynamics in coastal aquifers. In: Proceedings of the 10th international coastral symposium. J Coast Res 56:966–970 Google Scholar
  24. Khalil B, Ouarda TBMJ, St-Hilaire A (2011) Estimation of water quality characteristics at ungauged sites using artificial neural networks and canonical correlation analysis. J Hydrol 405(3–4):277–287 CrossRefGoogle Scholar
  25. Kişi Ö (2007) Streamflow forecasting using different artificial neural network algorithms. J Hydrol Eng (October):532–539 Google Scholar
  26. Kleinbaum DG, Kupper LL, Muller KE (2007) Applied regression analysis and other multivariable methods. Cengage learning, 906 pp Google Scholar
  27. Klute A (1965) Laboratory measurements of hydraulic conductivity of saturated soil. In: Black CA et al (eds) Methods of soil analysis. Part 1, Agronomy, vol 9, pp 210–220 Google Scholar
  28. Kozeny J (1927) Ueber kapillare Leitung des Wassers im Boden. Sitzungsber Akad Wiss Wien 136(2a):271 Google Scholar
  29. Linderman M, Liu J, Qi J, An L, Ouyang Z, Yang J, Tan Y (2004) Using artificial neural networks to map the spatial distribution of understorey bamboo from remote sensing data. Int J Remote Sens 25(9):1685–1700 CrossRefGoogle Scholar
  30. Minasny B, Hopmans JW, Harter T, Eching SO, Tuli A, Denton MA (2004) Neural networks prediction of soil hydraulic functions for alluvial soils using multistep outflow data. Soil Sci Soc Am J 68(2):417–429 CrossRefGoogle Scholar
  31. Morshed J, Kaluarachchi JJ (1998) Application of artificial neural network and genetic algorithm in flow and transport simulations. Adv Water Resour 22(2):145–158 CrossRefGoogle Scholar
  32. Nakhaei M (2005) Estimating the saturated hydraulic conductivity of granular material using artificial neural network based on grain size distribution curve. J Sci Islam Repub Iran 16(1):55–62 Google Scholar
  33. Pachepsky YA, Rawls WJ, Timlin DJ (1999) The current status of pedotransfer functions: Their accuracy reliability and utility in field- and regional-scale modeling. In: Corwin DL, Loague K, Ellsworth TR (eds) Assessment of non-point source pollution in the vadose zone: Geophysical monograph, vol 108. American Geophysical Union, Washington, pp 223–234 CrossRefGoogle Scholar
  34. Prechelt L (1998) Automatic early stopping using cross validation: Quantifying the criteria. Neural Netw 11(4):761–767 CrossRefGoogle Scholar
  35. R Development Core Team (2010) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/
  36. Sarkar D (1995) Methods to speed up error back-propagation learning algorithm. ACM Comput Surv 27(4):519–542 CrossRefGoogle Scholar
  37. Sarle WS (ed) (1997) Neural Network FAQ. Periodic posting to the Usenet newsgroup comp.ai.neural-nets. ftp://ftp.sas.com/pub/neural/FAQ.html, visited on 11/11/2010
  38. Schaap M, Leij FJ (1998) Using neural networks to predict soil water retention and soil hydraulic conductivity. Soil Tillage Res 47(1–2):37–42 CrossRefGoogle Scholar
  39. Schaap MG, Leij FJ, van Genuchten MT (1998) Neural network analysis for hierarchical prediction of soil hydraulic properties. Soil Sci Soc Am J 62:847–855 CrossRefGoogle Scholar
  40. Schaap M, Leij F, Van Genuchten MTh (2001) Rosetta: A computer program for estimating soil hydraulic parameters with hierarchical pedotransfer functions. J Hydrol 251(3–4):163–176 CrossRefGoogle Scholar
  41. Scott D (1979) On optimal and data-based histograms. Biometrika 66:605–610 CrossRefGoogle Scholar
  42. Soetens T (2008) Oriënterend bodemonderzoek in het kader van de aankoop van het terrein voor de oppervlakteberging van Umicore. NIRAS/ONDRAF 245.090-MER000, 2008-1119 herz.1 Google Scholar
  43. Stedinger JR, Vogel RM, Lee SU, Batchelder R (2008) Appraisal of the generalized likelihood uncertainty estimation (GLUE) method. Water Resour Res 44:W00B06 CrossRefGoogle Scholar
  44. Swingler K (1996) Applying neural networks: A practical guide. Academic Press, London Google Scholar
  45. Tiwari MK, Chatterjee C (2010) Uncertainty assessment and ensemble flood forecasting using bootstrap based artificial neural networks (BANNs). J Hydrol 382(1–4):20–33 CrossRefGoogle Scholar
  46. Tollenaere T (1990) SuperSAB fast adaptive back propagation with good scaling properties. Neural Netw 3(5):561–573 CrossRefGoogle Scholar
  47. Valverde Ramírez MC, Campos Velho HF, de Ferreira NJ (2005) Artificial neural network technique for rainfall forecasting applied to the Sao Paulo region. J Hydrol 301:146–162 CrossRefGoogle Scholar
  48. Van De Genachte G, Mallants D, Ramos J, Deckers JA, Feyen J (1996) Estimating infiltration parameters from basic soil properties. Hydrol Process 10(5):687–701 CrossRefGoogle Scholar
  49. Vienken T, Dietrich P (2011) Field evaluation of methods for determining hydraulic conductivity from grain-size data. J Hydrol 400(1–2):58–71 CrossRefGoogle Scholar
  50. Wang T, Wedin D, Zlotnik VA (2009) Field evidence of a negative correlation between saturated hydraulic conductivity and soil carbon in a sandy soil. Water Resour Res 45(7):W07503. doi: 10.1029/2008WR006865 CrossRefGoogle Scholar
  51. Wemaere I, Marivoet J, Labat S (2008) Hydraulic conductivity variability of the Boom clay in north-east Belgium based on four core drilled boreholes. Phys Chem Earth 33(S1):24–36 Google Scholar
  52. Wemaere I, Marivoet J, Labat S, Beaufays R, Maes T (2002) Mol-1 borehole (April–May 1997): Core manipulations and determination of hydraulic conductivities in the laboratory. SCK•CEN Report R-3590, 56 pp Google Scholar
  53. Zou R, Lung WS, Wu J (2007) An adaptive neural network embedded genetic algorithm approach for inverse water quality modeling. Water Resour Res 43:W08427. doi: 10.1029/2006WR005158 CrossRefGoogle Scholar

Copyright information

© International Association for Mathematical Geosciences 2012

Authors and Affiliations

  • Bart Rogiers
    • 1
    • 2
  • Dirk Mallants
    • 5
  • Okke Batelaan
    • 2
    • 3
  • Matej Gedeon
    • 1
  • Marijke Huysmans
    • 2
  • Alain Dassargues
    • 2
    • 4
  1. 1.Institute for Environment, Health and SafetyBelgian Nuclear Research Centre (SCK•CEN)MolBelgium
  2. 2.Dept. of Earth and Environmental SciencesKU LeuvenHeverleeBelgium
  3. 3.Dept. of Hydrology and Hydraulic EngineeringVrije Universiteit BrusselBrusselsBelgium
  4. 4.Hydrogeology and Environmental Geology, Dept. of Architecture, Geology, Environment and Civil Engineering (ArGEnCo) and AquapoleUniversité de LiègeLiègeBelgium
  5. 5.Groundwater Hydrology ProgramCSIRO Land and WaterGlen OsmondAustralia

Personalised recommendations