| Subsurface flow constructed wetlands(SSF CWs)have been widely used around the world for the natural design,low energy consumption operation,low maintenance cost and good treatment effect on contaminants in wastewater.However,clogging is a serious problem which could hamper successful operation of SSF CWs.Clogging can lead to malfunction in the hydraulic systems,together with a reduction in the porosity,affecting the operation of wetlands and finally shortening the functional lifespan of SSF CWs.The study of clogging process will help to construct treatment wetlands as a sewage treatment technology,which is still regarded as the “black box system”,and provide a more accurate reference for the management and maintenance of SSF CWs.However,there are few studies on the occurrence and development of clogging.Reason is as follows,first,regular monitoring of changes within the wetland matrix is needed for the research on clogging in wetlands,however,which is limited by a lack of real-time monitoring technology in SSF CWs.Without adequate monitoring technology,changes within the wetland matrix can be measured only through substrates sample testing,which could cause disruption to the internal environment in substrates around the sample sites.Therefore,data based on regular sampling of substrates in a SSF CW system over time to reveal the temporal inner changes are still notably lacking.Secondly,the physical,chemical and biological reactions in wetlands are complex,with accumulated solids composed of widely diverse matter in the matrix.It is difficult to explore the relationship between these internal changes and the performance of wetlands,even under the premise of knowing the internal changes in wetland matrix.Thirdly,models constructed by the current research to reveal clogging regards welands as a whole,and can not accurately describe the temporal and spatial development of clogging.In this paper,the vertical flow type of SSF CWs was chosen to construct laboratory systems for studying the physical clogging process caused by the adsorption and interception of the particles from influent in SSF CWs.First,the influence of matrix type,size and hydraulic retention time on physical clogging in the laboratory systems was studied to determine the design and operation parameters of the experiment.Next,temporal and spatial distribution of the solids accumulation in the laboratory systems was discussed based on periodic sampling and analysis of the substrates during the whole experiment.Further,through periodic monitoring of functional indexes(pollutant removal rate,porosity and permeability)of the laboratory systems,changes of these functional indexes along with solids accumulation in the systems were analyzed.And the relationship between the internal changes of the systems and the change of external function indexes were described quantitatively.Finally,through tracer tests,variation of hydraulic behavior in the laboratorysystems was analyzed and simulated by Computational Fluid Dynamics(CFD)code.Based on the simulation results,temporal and spatial distribution models were set up to describe the physical clogging process in the laboratory systems.The main results obtained can be summarized as follows:1?Although the internal porous structure substrates can provide a higher initial porosity for wetlands,the porosity decreases more rapidly than non-porous structural substrates.Larger particle size would lead to the higher initial porosity and infiltration rate of the wetland system,and the lower porosity decreasing rate over running time.Infiltration rate of systems with substrate diameter larger than 5mm increased at the early stage of the experiment,then decreased sharply.However,infiltration rate of the substrate with diameter smaller than 5mm kept decreasing during the whole experimental period.The effect of hydraulic retention time on porosity is not significant,but it has a significant effect on the removal rate of total suspended particulates(TSS)of the wetland systems.In summary,the following physical clogging test was operated chosing gravel with diameter of 3-4 mm as substrate.And the hydraulic retention time was set at 2 hours.2?During the experiment,being absorbed on gravel surface with particle size less than 20 ?m was the main way for particulates accumulation,and the system running remained stable.Particles in the top and the bottom layers of matrix were accumulated at a constant rate.But for the middle layer,there was no significant correlation between the accumulation of particulate matter and running time.At the end of the experiment,the content of large size particles in the pores of the top layer began to increase.This phenomenon was not observed in the middle and bottom layers of the matrix.The size distribution of particles adsorbed on the surface of gravel was steady and did not show any significant change.Although the adsorption of particles during the experiment is the main accumulation mode in the laboratory systems,the adsorption rate of particles is lower than the rate of particles accumulating in pores.3 ? TSS reduction rate remained relatively stable at a mean of 46.44%,under constant operation conditions,with no increasing or descending trend observed throughout operation time in the top layer of vertical-flow CWs in laboratory.Correspondingly,the solids accumulation rate was uniform(5E-5 g/g gravel per day)under the condition that the contributions of biomass was excluded.And in systems where the density of accumulated solids could be recognized as homogeneous,porosity would decrease at a uniform rate(2.36E-4 per day).In addition,the infiltration rate rapidly declined during the early stage of operation,but fluctuated during the middle stage because of flushing and re-accumulation.And finally,the infiltration rate would start presenting a downward trend.A significant correlation between accumulated solids and porosity was confirmed at the0.01 level with the Pearson correlation coefficient being-0.776.In addition,porosity also had a correlation with infiltration rate at the 0.05 level,although no direct correlations were observed between accumulated solids and infiltration rate(P=0.109).The relationship between TSS reduction rate and accumulated solids;accumulated solids and effective porosity;effective porosity and infiltration rate were found to be effectively described using formulas with simulated values showing a good fit with measured values.4?During the experiment,the porosity of the laboratory systems decreased from 36.69% to31.6%.The nominal hydraulic retention time reduced from the initial 137.14 min to 118.1min correspondingly.The actual average hydraulic retention time obtained by tracer tests also decreased alongwith the development of clogging,from the initial 106.62 min to 77.4 min.Both of the nominal retention time and the actual average retention time were significantly negatively correlated with running time at 0.01 level(Pearson coefficients were-0.885 and-0.773,respectively).The linear fitting results showed that the actual hydraulic retention time decreased faster than the nominal hydraulic retention time during the experiment.The calculation results of four indexes indicating short flow,mixed flow,effective utilization of volume and hydraulic efficiency showed that serious short flow and mixed flow occurred in the systems with dead zone existing,and the overall hydraulic efficiency was poor.But variation of the above four indicators were not significant during the experiment.CFD code could be used to simulate the hydraulic retention time.The simulation results of flow field and hydraulic retention time showed that the effective volume of the system did not change significantly during the experiment,but the flow velocity in the effective watershed is gradually accelerated and the time required to flow through the matrix was gradually shortened.5?Based on the simulation results of CFD,the spatial and temporal distribution model of particle accumulation and TSS concentration in the laboratory systems were constructed.Further,according to the quantitative relationship between the accumulation of particulate matter and the porosity,porosity and infiltration rate,the spatial and temporal distribution model of porosity and infiltration model were constructed to reveal the physical clogging process.As the interception of TSS by the bottom of the experimental column was not considered in the models,there was a deviation between the simulated data and the measured data of accumulated solids in the bottom layer of substrates. |
PDF Full Text Request