Spatio-Temporal Dynamics Of Reactive Oxygen Species In Riparian Aquifers:Experiments And Modeling

Reactive oxygen species,mainly including hydroxyl radicals（·OH）,hydrogen peroxide（H₂O₂）and superoxide ions（·O₂^-）,are molecular oxygen（O₂）to water（H₂O）Intermediate product of one-electron sequential reduction process.ROS in the subsurface environment has been paid more and more attention by researchers in the past ten years because of its significant environmental effects.Existing studies usually focus on the production mechanism and environmental effects of ROS in indoor experimental systems or the temporal and spatial distribution of ROS in natural environments.The one-electron reduction process of oxygen is the main production mechanism of ROS,and ROS has obvious effects on the cycle of variable valence elements,the degradation of organic pollutants,and the evolution of microbial communities.On the other hand,the solute transport driven by water flow creates conditions for the contact of reduced species and O₂.We mention that the recent researches rarely consider other classical biogeochemical reactions of related species in the ROS reaction networks.The spatiotemporal dynamics of reactive oxygen species under complex natural conditions is still insufficiently understood.The natural occurrence and environmental significance of ROS in aquifers have always been emphasized.Nonetheless,the current understanding of the mechanism of ROS dynamics in the actual process is still fragmented,and the environmental contribution of ROS is still not clear.The lack of numerical simulation also hinders the understanding of the spatial and temporal dynamics of reactive oxygen species and the evaluation of environmental effects.Referring to the realistic riparian aquifer,coupled with experiments and numerical simulation,this study explored and sorted out the controlling mechanism of the spatio-temporal dynamics of ROS in the riparian aquifer,quantified the ROS production and consumption,and identified the main controlling factors.The main work and conclusions are as follows:（1）The tank experiment and corresponding numerical simulation were carried out to simulate the interactive process of riparian aquifers which coupling ROS production with traditional biogeochemical processes.In the experiment,we chose H₂O₂ to represent ROS.The results of sand tank experiments and simulations showed that the concentration of H₂O₂ in the subsurface environment is related to the concentration of DO（Dissolved Oxygen）and the reactivity of reduced species.In addition,other electron acceptors with weaker reactivity than DO,represented by NO₃^-,could reach the deep part of the aquifer before DO,and weaken the production potential of ROS by preferentially consuming high reactivity reduced species.At the same time,water flow conditions affected the supply dynamics of DO and indirectly shapes the spatiotemporal distribution of ROS.In addition,DOC also competed DO by Aerobic organisms with reduced species and restricted the production of ROS.Based on this,the mechanism of ROS spatio-temporal dynamics can be further summarized as follows:DO transport by water flow and coexists transiently with different reactive reduced species,which constitutes the natural production region of ROS.Water flow conditions,hydration conditions,and the composition characteristics of reduced species shape the ROS dynamics with high spatiotemporal instability by regulating the DO supply/consumption and the reactivity of reduced species.（2）Due to the significant impact of DO supply on ROS,quantifying DO supply dynamics in aquifers is a prerequisite for further evaluation of ROS spatiotemporal dynamics.The results of series of field work showed that a large amount of DO in riparian aquifers come from the air,however this source of DO is neglected in current numerical simulations,which leads to misestimation of the concentration and distribution of DO in the aquifer.Bearing all that in mind,we developed a model considering air-liquid two-phase flow,air entrapment and multi-phase transport of DO to to investigate the DO dynamics in the riparian aquifer.The reliability and applicability of the model were verified by previous experimental data and field observation results.The simulation results showed that the river,the entrapped air and the vadose zones all provide a significant amount of DO to the riparian aquifer,and there are also noticeable differences in the spatial distribution of DO from the three different sources.The amplitude and period of the river level fluctuation,the base flow conditions,the permeability coefficient of the aquifer and the DO concentration in the river water significantly affected the DO supply dynamics of the above pathways.This work provided a useful tool for estimating DO supply to riparian aquifers and lays the foundation for the quantification of ROS.（3）In general,ROS have high reactivity,and there are great differences in the time scales in the reaction network of ROS production and consumption.This feature is called stiffness in computational science.The numerical solution of stiff systems requires very fine time or space discretization based on the smallest scale to avoid numerical oscillations,which requires a huge amount of CPU time and memory and greatly increases the computational cost.Due to the stiffness of the ROS reaction network,the computational cost of local-scale ROS reactive transport model is too large to be acceptable.To overcome this dilemma,we introduced the computational singular perturbation method into this field,combined this method with the reaction transport modeling process,and developed an efficient and high-precision method for highly complex systems including ROS dynamics in subsurface environments.Based on time scale analysis,this method decoupled all components and reactions in the system into fast components and slow components,and retained the slow components in the subsequent numerical solution,and the fast components were replaced by the slow components.The new method greatly improved the computational efficiency of the reaction transport model while maintaining the simulation accuracy.After a case test,the method improved the computational efficiency by 80 times under the condition of meeting the accuracy requirements.This method made the local-scale ROS reactive transport model unrestricted by the computational cost.In addition,this method could also be applied to the modeling of biogeochemical processes which were troubled by stiffness,to predict behaviors of complex earth systems and optimize the technique of groundwater remediation.（4）Previous researches have found naturally occurring ROS in in riparian aquifers.We take the riparian aquifer at Rifle site,Colorado,USA as representative,to investigate the mechanism of ROS spatiotemporal dynamics in local-scale.The environmental contribution of ROS was evaluated by what if the aquifer is polluted by rhodamine b.This work integrated local-scale water flow and biogeochemical processes through numerical simulation,and a large number of field observation data and experimental results were used to guide the model development and verify the reliability.The simulation results of the model were fixed well with the observation results under multiple scenarios.These also indicated that the model has great potential in quantifying and predicting the ROS dynamic and environmental effects in the field,and the verified the spatio-temporal characteristic of ROS summarized in this thesis.The model established in this work reconstructs the spatio-temporal dynamic of ROS during the river-groundwater interaction within riparian aquifers,and determines the annual H₂O₂and·OH production and environmental effects.The environmental contribution of ROS includes accelerating the oxidation process,promoting CO₂ emissions and enhencing the degradation of pollutants.The results of sensitivity analysis showed that a large and efficient DO supply and a large and highly reactive reduced species will constitute a ROS hotspot with significant environmental effects in the aquifer.In summary,this thesis focuses on the spatio-temporal dynamics of ROS in the riparian aquifers.With a serious of laboratory experiments,method development and numerical simulation,this work further improves the understanding of the mechanism of ROS spatio-temporal dynamics in the riparian aquifer.This thesis also establishes the local-scale model to quantify the ROS dynamics and environmental effects in realistic aquifers,develops the acknowledge of ROS in the subsurface environment from qualitative to quantitative.Specifically,this thesis advances the understanding of ROS spatio-temporal dynamics in the subsurface environment from under what conditions ROS can be produced,to within what process,how much ROS can be produced in how long and in a wide range,and how much environmental effects can be contributed.In addition,the results of this work can also be used to guide the optimization of ROS-based groundwater remediation process,which has practical application value.