Research Of Solute Transport Law In Saturated Soil And Unsaturated Soil Based On Uncertainty Of Permeability Coefficient

Groundwater is valuable and non-renewable resources. Along with the developmentof the Chinese economy, the speeding up of the industrialization and the damage to thegroundwater environment by geotechnical engineering construction, more and moregroundwater resources have being contaminated，which is a serious threat to human healthand survival environment. Groundwater pollution has caused widespread public concern.To understand the migration of pollutants in soil and accurately capture the pollution rangeof pollutants is the key to solve the groundwater pollution. And the numerical simulationcan clearly show the distribution of pollutants in soil. Numerical simulation ofgroundwater solute transport has become a new research direction in the field ofgroundwater and Geo-environmental Engineering. And it attracted more and more scholarsto join this project.As the carrier of the groundwater, soil directly affects the migration of pollutants ingroundwater. But because of the complexity of the environment in the process of soildeposits, the texture and structure of soil presents corresponding complexity anduncertainty. And performance is that the parameters for each point in the soil have bothrandomness and certain relevance. Permeability coefficient is one of the importantparameters affecting solute transport. This paper had simulated the migration process of Cl-leaking to into the saturated soil layer with the groundwater considering the uncertainty ofpermeability coefficient. The randomization of permeability coefficient is completed byLatin Hypercube method.In this paper, Bayesian Statistical Method is used to modify the statistics ofpermeability coefficient, and randomization of permeability coefficient is completed basedon the revised statistics. Then, The stochastic model of unsaturated soil solute transport isbuilding by HYDRUS-1D software. The saturated soil solute transport model is stablishedby GMS software on the basis of the unsaturated soil solute transport model. Finally themodel calculation results and the actual monitoring data is analyzed, drawing the followingconclusions:1) when a random simulation is performed, and the data is so little thatparameter information is not enough, we can firstly use Bayesian Statistical Methods tomodify the statistics of parameter and then conduct the randomization of parameter. Thismethod is more effective than conducting the randomization of parameter directly.2)When the permeability coefficient of unsaturated soil layer thicker is smaller and the thickness islarger, the barrier action of soil on Cl-is greater and the effect of the unsaturated soil layeron pollutant concentration can not be ignored.3) During the numerical simulation ofsaturated soil solute transport, it is more scientific to take the uncertainty of permeabilitycoefficient in to consideration compared with the case ignoring the uncertainty ofpermeability coefficient. For one thing, the result of random model calculation is closer toactual situation and can reflect influence of parameters uncertainty on solute transport; foranother, the value of permeability coefficient has significant effects on the pollutantconcentration, it can greatly reduce the error caused by improper permeability coefficientvalues using the mean concentration distribution of stochastic model instead of theconcentration distribution of certainty model.