Applicability of statistical learning algorithms in groundwater quality modeling

Khalil, Abedalrazq F.; Almasri, Mohammad N.; McKee, Mac; Kaluarachchi, Jagath J.

doi:10.1029/2004wr003608

Cited by 132 publications

(78 citation statements)

References 68 publications

(121 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, these kinds of models are not always efficient, especially in hydrogeological heterogeneous models which simulate complex and non linear phenomena. For such models, modern statistical learning algorithms can show much better ability to build accurate models with strong predictive capabilities (Khalil et al 2004;Marrel et al 2006).…”

Section: Response Surfacesmentioning

confidence: 99%

Global sensitivity analysis for a numerical model of radionuclide migration from the RRC “Kurchatov Institute” radwaste disposal site

Волкова

Iooss

Dorpe

2006

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Today, in different countries, there exist sites with contaminated groundwater formed as a result of inappropriate handling or disposal of hazardous materials or wastes. Numerical modeling of such sites is an important tool for a correct prediction of contamination plume spreading and an assessment of environmental risks associated with the site. Many uncertainties are associated with a part of the parameters and the initial conditions of such environmental numerical models. Statistical techniques are useful to deal with these uncertainties. This paper describes the methods of uncertainty propagation and global sensitivity analysis that are applied to a numerical model of radionuclide migration in a sandy aquifer in the area of the RRC ''Kurchatov Institute'' radwaste disposal site in Moscow, Russia. We consider 20 uncertain input parameters of the model and 20 output variables (contaminant concentration in the observation wells predicted by the model for the end of 2010). Monte Carlo simulations allow calculating uncertainty in the output values and analyzing the linearity and the monotony of the relations between input and output variables. For the non monotonic relations, sensitivity analyses are classically done with the Sobol sensitivity indices. The originality of this study is the use of modern surrogate models (called response surfaces), the boosting regression trees, constructed for each output variable, to calculate the Sobol indices by the Monte Carlo method. It is thus shown that the most influential parameters of the model are distribution coefficients and infiltration rate in the zone of strong pipe leaks on the site. Improvement of these parameters would considerably reduce the model prediction uncertainty.

show abstract

Section: Response Surfacesmentioning

confidence: 99%

Global sensitivity analysis for a numerical model of radionuclide migration from the RRC “Kurchatov Institute” radwaste disposal site

Волкова

Iooss

Dorpe

2006

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

show abstract

“…Over the past few years there have been various applications of SVM in hydrologic science [Dibike et al, 2001;See and Abrahart, 2001;Liong and Sivapragasam, 2002;Asefa and Kemblowski, 2002;Gill et al, 2003Gill et al, , 2006Asefa et al, 2004;Khalil, 2005;Khalil et al, 2005aKhalil et al, , 2005bKhalil et al, , 2005cKhalil et al, , 2006Asefa et al, 2005]. It is a far more robust machine than ANNs [Khalil, 2005;Khalil et al, 2005a], and it has been seen to reduce the dimensionality of the calibration procedure considerably as compared to a physical/conceptual model, e.g., Sacramento soil moisture accounting (SACSMA) model. There are three SVM parameters that must be determined through calibration: trade-off C, tolerance e, and kernel parameter g. The model tested here is the same as the one already used in a previous study for soil moisture prediction [Khalil et al, 2005b].…”

Section: Case Study Iii: Svm Soil Moisture Modelmentioning

confidence: 99%

Multiobjective particle swarm optimization for parameter estimation in hydrology

Gill

Kaheil

Khalil

et al. 2006

Water Resources Research

Self Cite

164

103

View full text Add to dashboard Cite

[1] Modeling of complex hydrologic processes has resulted in models that themselves exhibit a high degree of complexity and that require the determination of various parameters through calibration. In the current application we introduce a relatively new global optimization tool, called particle swarm optimization (PSO), that has already been applied in various other fields and has been reported to show effective and efficient performance. The PSO approach initially dealt with a single-objective function but has been extended to deal with multiobjectives in a form called multiobjective particle swarm optimization (MOPSO). The algorithm is modified to account for multiobjective problems by introducing the Pareto rank concept. The new MOPSO algorithm is tested on three case studies. Two test functions are used as the first case study to generate the true Pareto fronts. The approach is further tested for parameter estimation of a well-known conceptual rainfall-runoff model, the Sacramento soil moisture accounting model having 13 parameters, for which the results are very encouraging. We also tested the MOPSO algorithm to calibrate a three-parameter support vector machine model for soil moisture prediction.

show abstract

“…ANN models have proved to be superior to mechanistic models when data are limited and numerous assumptions have to be made to solve the physically based equations. For instance, in-stream nitrogen concentrations, which are affected by land use/land classification, vegetation dynamics and in-stream nitrogen transformations, were simulated reasonably well with the ANN approach (Lek et al 1999, Khalil et al 2005.…”

Section: Link Between Swat and Annmentioning

confidence: 83%

Incorporating water quantity and quality modelling into forest management

Li¹,

Nour²,

Smith³

et al. 2008

The Forestry Chronicle

View full text Add to dashboard Cite

Under the authority of the Millar Western Forest Products Ltd. (MWFP) Forest Management Agreement, the company has the right to harvest trees, grow trees, manage the forest and plan activities that assure forest productivity and industry profitability without jeopardizing the quality of the environment. Thus, as part of obtaining provincial government approval, the company has to submit a Detailed Forest Management Plan that includes a comprehensive assessment of the environmental implications of forestry operations and the mitigation of impacts. Forest management planning for environmental sustainability will become more and more difficult with increased land use pressures from other industries, agriculture and recreation. Therefore, the planning process will require increasingly more sophisticated modelling tools to identify and avoid significant impacts. The Forest Watershed and Riparian Disturbance (FORWARD) project proposes a hybrid modelling tool that relies on inexpensive remote sensing data, with few ground truthing requirements, to model streamflow, suspended solids and nutrients in streams on the Boreal Plain. Incorporating modelling tools into the MWFP planning process provides MWFP additional strategies to operate in an environmentally sensitive manner. Thus, the company can maintain an allowable cut, while ensuring that ecological and physical values are considered.

show abstract

Applicability of statistical learning algorithms in groundwater quality modeling

Cited by 132 publications

References 68 publications

Global sensitivity analysis for a numerical model of radionuclide migration from the RRC “Kurchatov Institute” radwaste disposal site

Global sensitivity analysis for a numerical model of radionuclide migration from the RRC “Kurchatov Institute” radwaste disposal site

Multiobjective particle swarm optimization for parameter estimation in hydrology

Incorporating water quantity and quality modelling into forest management

Contact Info

Product

Resources

About