| Soil as a porous media exists widely in nature,and correctly understanding the migration and distribution of solutes and water flows in porous media is of great significance for environmental protection issues such as groundwater pollution prevention and control and rational application of water and fertilizer.However,due to the heterogeneity and strong spatial variability of porous media,solutes often show irregular transport phenomena during transport and are often affected by preferential flow,and the reaction migration process of pollutants changes with depth also increases the difficulty of studying solute migration.Therefore,given the above problems,in this thesis,we investigated the following four issues:analysis of the influence of pore characteristics on the hydraulic properties and preferential flow of heterogeneous media based on pore-scale simulation,experimental study of solute migration in heterogeneous media under the influence of preferential flow,comparative analysis of solute migration process in heterogeneous media under the influence of different mathematical models,and solute transport model and application considering the change of reaction process with depth under the influence of preferential flow.The research outcome will promote the enrichment and development of solute transport theories in heterogeneous media and provide model theoretical guidance for practical remediation work.The results of the experiment and numerical simulation are analyzed as follows:（1）The influence of pore characteristics on hydraulic properties in porous media is studied based on pore-scale simulation.The effects of pore size,pore connectivity,anisotropy on field distribution,and preferential channel flow path were discussed.The results show that the media’s porosity and intrinsic permeability increase with the increase of the average pore sizeμ.,Conversely,it decreases with the rise in the standard deviation of the pore size,which is linearly increasing with the coordination number.In the case of the same pore size,the uniform distribution of the pore size,the normal distribution and the lognormal distribution of the three distributions,the smallest porosity and the inherent permeability appear under the lognormal distribution conditions.The largest porosity appears under the normal distribution conditions,and the maximum inherent permeability appears under the lognormal distribution conditions.As the coefficient of variation of the pore size increases,the flow path in the pore network becomes more tortuous.Similarly,the degree of tortuousness gradually increases.For example,when the coefficient of variation increases from 0 to1,the tortuousness of the dominant path increases from 1 to 1.71,and the ratio of the flow to the total flow rate along the dominant path increases from 2%to 7.8%.Consequently,as the aperture ratio in the horizontal and vertical directions gradually increases,more pores in the horizontal direction participate in the seepage process,leading to a gradual increase in the tortuousness of the flow path.At the same time,this effect is weakened when the coefficient of aperture variation is large.In addition,the ratio of anisotropy of intrinsic permeability Kh/Kvto the ratioμdh/μdv of pore size in the horizontal and vertical directions shows a linear positive correlation at logarithmic coordinates.（2）A series of soil column solute penetration experiments were carried out,and the irregular migration of solutes was explored and evaluated based on controlling factors such as connectivity,size,density,and recharge strength of large pore structures.The results show that recharge flux has an important influence on the solute migration process.As such the greater the recharge flux,the more significant the influence of the priority flow.Under certain conditions of the recharge flux,the greater the proportion of large pore channels,the greater the contribution of the priority flow to solute migration.Also,the penetration degree of large pores has a significant impact on solute migration.In the case where there is equal number of large pores and the recharge flux,the large pores that run through create a significant bimodal distribution of the solute penetration curve,but not under the same conditions.Large pores that do not penetrate do not lead to any preferential migration of the solute penetration curve;otherwise,the preferential migration of solutes is more likely to occur in porous media with finer particle size.（3）The solute migration process in heterogeneous media under the influence of preferential flow was simulated,and the five most widely used models were compared and analyzed.The Advection-diffusion model（ADE）,Mobile-immobile model（MIM）,Mobile-mobile model（MM）,Stochastic advection-diffusion model（SADE）,and continuous-time random walking model（CTRW）were employed to simulate the solute transport process in soil columns under different conditions.The fitted results of each model were compared and analyzed.The simulation difference is explored from the model mechanism,and the advantages and disadvantages of the models for the solute transport process fitting are compared.The results show that ADE,MIM,MM,SADE,and CTRW can adequately describe the solute transport process when the solute migration is not significantly affected by the priority flow.Also,the ADE,MIM,and SADE models perform relatively poorly in describing solute transport when solute migration is greatly affected by the priority flow.Meanwhile,the MM and CTRW were found to adequately describe the solute transport process where the double-peak phenomenon of solute transport under the influence of MM on the priority flow is better.However,the MM model is considered to be inferior to the CTRW model in terms of the front end and tail end of the solute penetration curve;ADE,MIM,MM,SADE,and CTRW have their advantages and disadvantages in the physical background,mathematical form,number of parameters,the difficulty of solution and extrapolation applicability of actual problems,and in the selection of models,it is necessary to start from the problem,comprehensively consider the problem requirements and the difficulty and accuracy of the solution.（4）In view of the problem that the solute reaction process in heterogeneous media changes with depth,this thesis generalizes the reaction process as a deep inflection function,and based on the Mobile-mobile model（MM）,a reactive solute transport model considering the depth of the reaction process is established.The effect of the mass transfer coefficient and the flow rate of each region on the solute penetration curve is compared and analyzed.The comparative analysis shows that the size of the mass transfer coefficient determines the size of the peak of the penetration curve and the number of concentration peaks.The main flow rate determines the shape of the penetration curve.The speed of slow velocity domain（SVD）determines the mode of the penetration curve bimodal;the penetration curve usually shows a unipolar distribution under the small SVD flow rate;the difference between the two peaks gradually increases with the increase of soil depth.The results of the study on the effect of the depth-dependent reaction coefficient on pollutant migration show that the depth-dependent penetration curve is located between the penetration curves with the two constant reaction coefficients（at the surface and the observation point）.In the case where the solute reaction changes with the depth of the medium without consideration for the process,a certain degree of overestimation or underestimation of the simulation prediction results occurs.In addition,this thesis further uses field observation data to confirm the model’s applicability.The results show that the model established in this thesis can adequately characterize the change of solute concentration at different depths and also simulate the influence of priority flow by taking into account the depth-related reaction process. |
