Modeling multidimensional flow in wettable and water-repellent soils using artificial neural networks

Summary This study examined the use of three different classes of artificial neural networks for modeling water flow in wettable and water-repellent soils, using both synthetic numerical data and experimentally measured data. The 1D self-organizing maps (SOM) successfully rendered the moisture contour in the transition zone of the wetting plumes for all soil types at different flow rates. Due to SOMs inability to generate external output data, multilayer perceptrons (MLP) and modular neural networks (MNN), respectively, were combined with SOM to predict the moisture contour for both wettable and water-repellent soils. Due to dimensionality reduction, the 1D SOM failed to capture high moisture content classes of water-repellent soils with anomalous wetting patterns, whereas spatial moment analysis succeeded in providing an accurate, albeit indirect, description. Hence, the MLP and MNN networks were applied to predict the spatial moments. The comparison between the predicted and the experimental measures demonstrated the capability of the MLP and SOM to predict the spatial moments. Comparison of the two different artificial neural networks indicated no significant difference between their results.

[1]  Ivica Kostanic,et al.  Principles of Neurocomputing for Science and Engineering , 2000 .

[2]  P. Moerland,et al.  Some Methods for Training Mixtures of Experts , 1997 .

[3]  N. J. de Vos,et al.  Multiobjective training of artificial neural networks for rainfall‐runoff modeling , 2008 .

[4]  Tammo S. Steenhuis,et al.  Preferential flow in water-repellent sands , 1998 .

[5]  Brent Clothier,et al.  The breakdown of water repellency and solute transport through a hydrophobic soil , 2000 .

[6]  Niels Schütze,et al.  Self‐organizing maps with multiple input‐output option for modeling the Richards equation and its inverse solution , 2005 .

[7]  X. Pang,et al.  Approaches to characterize the degree of water repellency , 2000 .

[8]  Wesley W. Wallender,et al.  Performance evaluation of methods for the estimation of soil hydraulic parameters and their suitability in a hydrologic model , 2006 .

[9]  M. Deurer,et al.  Modeling Water Movement in Heterogeneous Water‐Repellent Soil: 1. Development of a Contact Angle–Dependent Water‐Retention Model , 2007 .

[10]  A. Warrick,et al.  Subsurface Water Distribution from Drip Irrigation Described by Moment Analyses , 2007 .

[11]  B. S. Thandaveswara,et al.  A non-linear rainfall–runoff model using an artificial neural network , 1999 .

[12]  Yunmei Chen,et al.  Smoothing and Edge Detection by Time-Varying Coupled Nonlinear Diffusion Equations , 2001, Comput. Vis. Image Underst..

[13]  Holger R. Maier,et al.  Optimal division of data for neural network models in water resources applications , 2002 .

[14]  Yakov A. Pachepsky,et al.  Estimating saturated soil hydraulic conductivity using water retention data and neural networks , 1999 .

[15]  L. Debano.  Water repellency in soils: a historical overview , 2000 .

[16]  U. C. Kothyari,et al.  Modeling of the daily rainfall-runoff relationship with artificial neural network , 2004 .

[17]  John Scott Bridle,et al.  Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition , 1989, NATO Neurocomputing.

[18]  R. Shakesby,et al.  Spatial variability of soil hydrophobicity in fire-prone eucalyptus and pine forests, Portugal , 1998 .

[19]  G. D. Rooij.  A three-Region Analytical Model of Solute Leaching in a Soil with a Water-Repellent Top Layer , 1995 .

[20]  Stephen Grossberg,et al.  A neural network architecture for figure-ground separation of connected scenic figures , 1991, Neural Networks.

[21]  C. Ritsema,et al.  Occurrence of soil water repellency in arid and humid climates , 2000 .

[22]  Uwe Petersohn,et al.  New strategy for optimizing water application under trickle irrigation , 2002 .

[23]  David A. DiCarlo,et al.  Can Continuum Extensions to Multiphase Flow Models Describe Preferential Flow? , 2010 .

[24]  Jan M. H. Hendrickx,et al.  Preferential flow mechanism in a water repellent sandy soil , 1993 .

[25]  Kuolin Hsu,et al.  Artificial Neural Network Modeling of the Rainfall‐Runoff Process , 1995 .

[26]  Magnus Persson,et al.  Predicting the Dielectric Constant-Water Content Relationship Using Artificial Neural Networks , 2002 .

[27]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[28]  M. Schaap,et al.  Neural network analysis for hierarchical prediction of soil hydraulic properties , 1998 .

[29]  C. Ritsema,et al.  Preferential flow in water repellent sandy soils: principles and modeling implications , 2000 .

[30]  Kenji Suzuki,et al.  Neural Edge Enhancer for Supervised Edge Enhancement from Noisy Images , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[31]  R. Glass,et al.  On the porous‐continuum modeling of gravity‐driven fingers in unsaturated materials: Extension of standard theory with a hold‐back‐pile‐up effect , 2002 .

[32]  R. Carsel,et al.  Developing joint probability distributions of soil water retention characteristics , 1988 .

[33]  Mohammad N. Almasri,et al.  Modular neural networks to predict the nitrate distribution in ground water using the on-ground nitrogen loading and recharge data , 2005, Environ. Model. Softw..

[34]  J. Feyen,et al.  Effects of soil water repellency on infiltration rate and flow instability , 2000 .

[35]  M. Schaap,et al.  ROSETTA: a computer program for estimating soil hydraulic parameters with hierarchical pedotransfer functions , 2001 .

[36]  Michael I. Jordan,et al.  Hierarchical Mixtures of Experts and the EM Algorithm , 1994, Neural Computation.

[37]  Y. Hong,et al.  Self‐organizing nonlinear output (SONO): A neural network suitable for cloud patch–based rainfall estimation at small scales , 2005 .

[38]  Y. Pachepsky,et al.  Artificial Neural Networks to Estimate Soil Water Retention from Easily Measurable Data , 1996 .

[39]  Geoffrey E. Hinton,et al.  Adaptive Mixtures of Local Experts , 1991, Neural Computation.

[40]  Rao S. Govindaraju,et al.  Prediction of watershed runoff using Bayesian concepts and modular neural networks , 2000 .

[41]  Robert J. Glass,et al.  On the continuum‐scale modeling of gravity‐driven fingers in unsaturated porous media: The inadequacy of the Richards Equation with standard monotonic constitutive relations and hysteretic equations of state , 2001 .

[42]  Ruben Juanes,et al.  A phase field model of unsaturated flow , 2009 .

[43]  Peter Wallbrink,et al.  Effects of differing wildfire severities on soil wettability and implications for hydrological response , 2006 .

[44]  A. M. Kalteh,et al.  Review of the self-organizing map (SOM) approach in water resources: Analysis, modelling and application , 2008, Environ. Model. Softw..

[45]  Naftali Lazarovitch,et al.  Water distribution under trickle irrigation predicted using artificial neural networks , 2009 .

[46]  R. Wallach,et al.  Soil water repellency induced by long-term irrigation with treated sewage effluent. , 2005, Journal of environmental quality.

[47]  R. E. Yoder,et al.  Simulation of nitrate distribution under drip irrigation using artificial neural networks , 2003, Irrigation Science.

[48]  Coen J. Ritsema,et al.  How water moves in a water repellent sandy soil: 2. Dynamics of fingered flow , 1994 .

[49]  N. Chang,et al.  Short-term streamflow forecasting with global climate change implications – A comparative study between genetic programming and neural network models , 2008 .

[50]  null null,et al.  Artificial Neural Networks in Hydrology. II: Hydrologic Applications , 2000 .

[51]  Jan M. H. Hendrickx,et al.  Unstable Wetting Fronts in Water-Repellent Field Soils , 1993 .

[52]  J. Nieber,et al.  Dynamic Capillary Pressure Mechanism for Instability in Gravity-Driven Flows; Review and Extension to Very Dry Conditions , 2005 .

[53]  A. Warrick,et al.  Neuro-Drip: estimation of subsurface wetting patterns for drip irrigation using neural networks , 2010, Irrigation Science.

[54]  M. Bíl,et al.  On the incompatibility of Richards' equation and finger‐like infiltration in unsaturated homogeneous porous media , 2009 .

[55]  R Govindaraju,et al.  ARTIFICIAL NEURAL NETWORKS IN HYDROLOGY: II, HYDROLOGIC APPLICATIONS , 2000 .

[56]  N. Null.  Artificial Neural Networks in Hydrology. I: Preliminary Concepts , 2000 .

[57]  R. Wallach,et al.  Unstable finger-like flow in water-repellent soils during wetting and redistribution – The case of a point water source , 2008 .

[58]  R. Shakesby,et al.  Soil water repellency: its causes, characteristics and hydro-geomorphological significance , 2000 .

[59]  Kuolin Hsu,et al.  Self‐organizing linear output map (SOLO): An artificial neural network suitable for hydrologic modeling and analysis , 2002 .

[60]  Martin T. Hagan,et al.  Neural network design , 1995 .

[61]  Van Genuchten,et al.  A closed-form equation for predicting the hydraulic conductivity of unsaturated soils , 1980 .

[62]  R. Glass,et al.  On the porous continuum‐scale modeling of gravity‐driven fingers in unsaturated materials: Numerical solution of a hypodiffusive governing equation that incorporates a hold‐back‐pile‐up effect , 2003 .

[63]  Paul L. G. Vlek,et al.  Artificial Neural Network Estimation of Saturated Hydraulic Conductivity , 2007 .

[64]  Klaas Oostindie,et al.  Exponential increase of publications related to soil water repellency , 2005 .

[65]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[66]  Amin Elshorbagy,et al.  Spiking modular neural networks: A neural network modeling approach for hydrological processes , 2006 .

[67]  Ravi Kothari,et al.  Spatial characterization of remotely sensed soil moisture data using self organizing feature maps , 1999, IEEE Trans. Geosci. Remote. Sens..