Uncertainty and Sensitivity Analysis of Solute Contaminant Transport Simulation in Groundwater (Case Study: Qazvin Plain)

Document Type : Case study


1 Former Gratuated Student, College of Water and Hydraulic Structural Engineering, Dept. of Civil, Water and Environmental Engineering, Shahid Beheshti University of Tehran, Iran

2 Assoc. Prof. of Water Engineering, Dept. of Civil, College of Water and Environmental Engineering, Shahid Beheshti University, Tehran, Iran


Groundwater contaminant transport modeling is a useful tool for identifying how pollutants fate and transport in porous aquifer environments. These models include several parameters, which are often estimated based on personal judgment or in the best case, based on limited field measurements. Therefore, the input data of simulation models are not accurate and contain several errors. The purpose of this study is parameter uncertainty and sensitivity analysis in the groundwater solute contaminants transport modeling using probability theory. First, governing equations for groundwater flow and solute contaminant transport have been presented. Then, using MOFLOW for modeling groundwater flow and MT3DMS for modeling solute contaminant transport in a hypothetical problem and using effective parameters in a case study (Qazvin plain), uncertainty analysis through the Monte Carlo method was done. To illustrate the uncertainty analysis, the Complementary Cumulative Distribution Functions (CCDF) of Chloride and Nitrate graphs have been computed. Then using random samples, generated in uncertainty analysis step, local and global sensitivity analysis of solute transport model parameters have been determined.  Result: Using maximum concentration of solute contaminant as a model output, the results of the local sensitivity analysis show that the most sensitive parameters are hydraulic conductivity (K), decay rate constant (λ), porosity (θ), distribution coefficient (Kd), and dispersivity (D) respectively. While using time to maximum concentration as output variable, leads to the following order of sensitivity: K, Kd, θ, D, and λ. On the other hand, the global sensitivity analysis using maximum concentration shows that the order of sensitivity is: K, λ, θ, D, and Kd, and using time to maximum concentration it is: K, Kd, D, λ, and θ respectively. According to the CCDF of Chloride, concentrations of 5%, 50% and 95% equal 205.5, 196 and 185.4 mg/L respectively. Also, according to the CCDF of Nitrate the concentrations of 5%, 50%, and 95% equal 56, 54.125 and 51.5 mg/L respectively. All five parameters are sensitivite in solute transport modeling. The local and global sensitivity analysis show more or less the same results. In general, the sensitivity ranking of parameters is K, λ, Kd, θ, and D.


Abkhan Consulting Engineers Company. 2013. Studies on updating the water resources balance report of Namak Lake catchment area, water balance report of Qazvin plain. Tehran Regional Water Company, Ministry of Energy, Tehran, Iran. (In Persian)
Behrouz, M. 2018. Uncertainty and sensitivity analysis of flood control levees design considering evidence theory. PhD Thesis, Shahid Beheshti University. Tehran, Iran. (In Persian)
Epstein, B. 1948. Some application of the Mellin transform in statistics, Annals of Mathematical Statistics, 19, 370-379.
Gelhar, L. W., Mantoglou, A., Welty, C. & Rehfeldt, K. R. 1985. EPRI report EA-4190, Elut. Power Res. Inst. California, Water Resource Research, 28, 1955-1974.
Gorelick, S. M. 1983. A review of distributed parameter groundwater management modeling methods. Water Resources Research, 19, 305-319.
Helton, J. C. 1993. Uncertainty and sensitivity analysis techniques for use in performance assessment for radioactive waste disposal. Reliability Engineering & System Safety, 42, 327-367.
Mays, L. W. & Tung, Y. K. 1992. Hydrosystems Engineering & Management. McGraw Hill. USA.
McDonald, M. G. & Harbaugh, A. W. 2003. The history of MODFLOW. Groundwater, 41(2), 280-283.
Rojas, R., Feyen, L. & Dassargues, A. 2008. Conceptual model uncertainty in groundwater modeling: combining generalized likelihood uncertainty estimation and Bayesian model averaging. Water Resources Research, 44, W12418.
Tung, Y. K. & Yen, B. C. 2005. Hydrosystems engineering uncertainty analysis, McGraw-Hill, New York.
Park, C. S. 1987. The Mellin transform in probabilistic cash flow modeling. The Engineering Economist, 32(2), 115-134.
Zheng, CH. 1990. A Modular Three-Dimensional Multispecies Transport Model (MT3DMS). S.S. Papadopulos and Associates, Inc. with partial support from the U.S. Environmental Protection Agency (USEPA). USA.
Zheng, Ch., Bennett, G. D. 2002. Applied Contaminant Transport Modeling. 2nd Edition. Wiley Interscience.