Application of subsurfacewastewater infiltration system to on-site treatment of domestic sewage under high hydraulic loading rate

Ying-hua Li，Hai-bo Li*，Xin-yang Xu，Xuan Gong，Yong-chun ZhouSchool of Resources and Civil Engineering，Northeastern University，Shenyang 110004，PRChina Received 17 February 2014；accepted 4 January 2015 Available online 7 February 2015



Application of subsurfacewastewater infiltration system to on-site treatment of domestic sewage under high hydraulic loading rate

Ying-hua Li，Hai-bo Li*，Xin-yang Xu，Xuan Gong，Yong-chun Zhou
School of Resources and Civil Engineering，Northeastern University，Shenyang 110004，PRChina Received 17 February 2014；accepted 4 January 2015 Available online 7 February 2015

Abstract

In order toenhance thehydraulic loading rate（HLR）of a subsurfacewastewater infiltration system（SW IS）used in treating domestic sewage，the interm ittent operation mode was employed in the SW IS.The results show that the interm ittent operation mode contributes to the improvementof the HLR and the pollutant removal rate.When thewetting-drying ratio（RWD）was1.0，the pollutant removal rate increased by（13.6±0.3）%for NH3-N，（20.7±1.1）%for TN，（18.6±0.4）%for TP，（12.2±0.5）%for BOD，（10.1±0.3）%for COD，and（36.2±1.2）%for SS，com pared w ith pollutant removal rates under the continuous operationmode.The pollutant removal rate declined w ith the increase of the HLR.Theeffluentqualitymet The ReuseofUrban RecyclingWater-Water Quality Standard for Scenic Environment Use（GB/T 18921-2002）evenwhen the HLRwasashigh as10 cm/d.Hydraulic conductivity，oxidation reduction potential（ORP），thequantity of nitrifying bacteria，and the pollutant removal rateof NH3-N increased with the decreaseof the RWD.For the pollutant removal rates of TP，BOD，and COD，therewere no significantdifference（p<0.05）under different RWDs.Thesuggested RWDwas1.0.Relative contributionof the pretreatmentand SW IS to the pollutant removalw as exam ined，and more than 80%removal of NH3-N，TN，TP，COD，and BOD occurred in the SW IS.

©2015 Hohai University.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http:// creativecommons.org/licenses/by-nc-nd/4.0/）.

Domestic sewage；Subsurface wastewater infiltration system；Interm ittent operationmode；Hydraulic loading rate；Pollutant removal rate

1.Introduction

In rural areas of Northeast China，conventional centralized sewer systemshave become impracticaldue to the topography and long distances between the connected facilities（Petter et al.，2010；Qian et al.，2007）.Water shortage in these areas has created a need for both higher quality and a greater quantity of reclaimed water.Conventional system s，such as activated sludge，biological contactors，and chemical precipitation，are alternatives，but previous studieshave shown that it is difficult for them to meet the discharge standard for phosphorus concentration（Arve et al.，2006；Fan et al.，2009）. Therefore，there isan urgentneed for simplemaintainable onsite system sw ith excellent treatment performance.

A subsurface wastewater infiltration system（SW IS）w ith pretreatm ent（e.g.，septic tank，biological contractor，and biological filtration）has been pioneered in Northeast China（Li et al.，2011；Qian et al.，2007）.Over the past 20 years，the SW IS has gained popularity as an effective and low-cost alternative forwastewater treatment，especially in villagesand small communities.The SW IS has shown excellent performance in organics，nitrogen，and phosphorus removal（Kadam et al.，2009）.Table 1 summarizes the treatment efficiency of SW ISs.Up to now，themain research on SWISs has focused on system design，treatm ent perform ance，and pollutant removalm echanisms（Pan et al.，2012；Zhang et al.，2011）.

Previous studieshave suggested that，although the SW IS is a treatment system with simple mechanisms，the treatment process of pollutant removal is intricate（Zhang et al.，2011）. The hydrology，microbiology，and water chemistry are complex and interconnected.Studies have presented relatively high removal efficiency for chem ical oxygen demand（COD），biochem ical oxygen demand（BOD），suspended solids（SS），and pathogens（Stewart and Louis，2010）.However，nutrient removal efficiency is low and variable.Moreover，the nutrient removal decreasesw ith the increase of service age of a SW IS（Robertson，2010）.In addition，compared to the conventional treatmentplants，SW ISsoccupy relatively largeareas resulting from the low hydraulic capacity.The cost of SW ISs is high due to the size requirement.If the hydraulic loading rate（HLR）could be higher，SWISs could be built smaller，and the initial costwould be lower，which would greatly enhance the market potential and application prospects of SW ISs.Therefore，the aims of this study were:（1）to examine the contribution of the interm ittent operation mode to the HLR encouragement in the SW IS；（2）to assess the pollutant removal contribution of both the SW IS and pretreatm ent；and（3）to evaluate the impact of the SW IS effluent on receivingwater quality under a high HLR.

Table 1 Treatment efficiency of SW ISs.

2.M aterials and m ethods

2.1.System description

Thewastewaterwas pretreated in a septic unitwith a hydraulic detention time of 4 h.The effluent flowed under the action of gravity through the distribution tank to four infiltration tanks.The dimension of each infiltration tank is 20 m long，15m w ide and 1.5m deep.Inflow ing pipeswere 0.5m underneath（100 mm in diameter w ith holes of 4 mm in diameter placed in the bottom side every 60mm）.Collecting pipeswere1.5m underneath（80mm in diameterw ith holesof 6 mm in diameter placed in the bottom side every 60 mm）. Thebedswereplantedw ith herbage（Poa annua and ryegrass），which wasmainly for landscape planting.A1，B1，C1，D1，E1，A2，B2，C2，D2，and E2（0.4 m intervals）were sampling positions for substrate samp les and bacteria numbering，as shown in Fig.1.The substrate samp les were taken tw ice a month from 0.2，0.4，0.6，0.8，and 1.0 m depths at each samp ling position，respectively.

Fig.1.Sketch of SW IS profile（units:m）.

2.2.Wastewater characteristics

Fieldexperimentswere carried outin Shenyang City，China. The influentto thepretreatmentunitwascombinedwastewater，from toilets，restaurants，etc.The rangesofmajorwaterquality indiceswere7.2-7.4 forthepH value，275-360mg/L forCOD，155-220mg/L forBOD，95-126mg/L forSS，30-45mg/L for total nitrogen（TN），3-4 mg/L for total phosphorus（TP），20-30mg/L forammonianitrogen（NH3-N），and0.2-0.3mg/L for nitrate nitrogen（NO-3-N）.

2.3.Substrate characteristics

The packed substrate in the SW ISwasa kind of novelbiosubstrate，which was composed of 5%activated sludge，65% meadow brown soil，and 30%coal slag m ixed evenly in volume ratios.The activated sludge was obtained from the aeration tanks in the Shenyang Northern Municipal Sewage Treatment Plant，China，and air-dried after being centrifuged for 15 min at 1 500 r/min.The meadow brown soil was samp led from the top 20 cm of soil at the Shenyang Ecological Station.Other materials（gravel and coal slag）were purchased from a local market（particle size: 10-25 mm of gravel and 4-8 mm of coal slag）.The infiltration rate，porosity，and surface area of the substratewere 0.37m3/（m2·d），59%，and 5.21m2/g，respectively.A previous study（Li et al.，2013）indicated that，in comparison w ith meadow brown soil，the bio-substrate provided a more favorable micro-environment for the pollutant removal.The m axim um adsorbing capacity for NH3-N was 0.724 mg/g，which was 0.253 mg/g higher than that of meadow brown soil.

2.4.Analysismethod

The ammonifying，nitrifying，and denitrifying bacteria in the substrate samples were counted using the most probable number（MPN）method tw icepermonth（Nie etal.，2011）.The medium components are shown in Table 2.Aliquots（1 m L）were diluted w ith 12-fold sterile distilledwaterand transferredto m icrotiter plates containing each type of medium，then incubated at 28°C for seven days for the ammonifying bacteria，14 days for the nitrifying bacteria，and 15 days for the denitrifying bacteria.Meanwhile，10 g of the substrate sampleswere oven-dried at105°C for 12 h to produce a constant weight.

During the study period，paired samplesof raw sewage and effluentwere collected every three days，stored at 4°C，and analyzed w ithin 24 h.The COD，BOD，SS，NO-3-N，NH3-N，TN，and TP concentrations of the water sam ples were analyzed according to the American Public Health Association（APHA，2003）guidelines.The potassium dichromatemethod wasused for COD determination，and the colorimetricmethod was used for NO-3-N and NH3-N measurements.Hydraulic conductivity wasmeasured according to themethod suggested by Lowe and Siegrist（2008）.Oxidation reduction potential（ORP）was analyzed through pre-buried sensors，and the readings were recorded every three days.All statistical analyses were carried out using the com puter software package Origin 7.5.W ith respect to surfacewater quality from different sampling points，a parametric analysis of variancewasused to determine the significant difference（p）.

Table 2 M edium components list for three kinds of bacteria.

2.5.Sampling and experimental operation

Thewhole system wasoperated from June 2009.Sampling for this study was conducted from April to September 2013. The water quality of the stream adjacent to the SW IS was monitored every three days to assess the impact of the SW IS effluent on receiving-water and verify that the SW ISwas not polluting the receiving-water under a high HLR.The stream was located next to the SW IS，as shown in Fig.2.The stream water was sampled at four locations，one upstream of the discharge point and the other three approximately 100，300，and 500m downstream of the discharge point；these locations are labeled a，b，c，and d，respectively.Mean concentrationsof BOD，SS，NH3-N，TN，TP，and COD in the upstream water during theexperimentalperiodwere 3.3，10.4，0.5，10，0.3，and 38.2 mg/L，respectively.During the experimental period，the intermittent operation mode was adopted as a method of encouraging the HLR.Each cycle of the interm ittent operation included a continuous flow period of 24 h（between 9:00 AM and 9:00 AM the next day）and a drying period of 0，24，48，72，and 96 h，indicating a wetting-drying ratio（RWD）of∞（termed a continuousoperationmode），1.0，0.5，0.33，and 0.25，respectively.All experimentswere repeated three times.

As for the HLR values，there is no exact information indicating what extent of the HLR is high.According to Sun and Li（2006），a value of the HLR lower than 8 cm/d is suggested for long lifespan of the SW IS.In this research，the intermittentoperationmodewasadopted to improve the HLR. Therefore，the HLR was gradually increased from 4 cm/d to 6 cm/d，8 cm/d，10 cm/d，and 12 cm/d.

Fig.2.Sketch of SWISand monitoring sites in stream.

3.Results and discussion

3.1.Effectsof intermittent operationmode and HLR on treatment efficiency of SWIS

Table 3 showspollutantconcentrationsand removal ratesat the discharge point under continuous and interm ittent operationmodes.The HLR was 8 cm/d for both operation modes.

When the RWDwas 0.25，the interm ittent operation mode improved pollutant removal ratesby（23.3±0.2）%for NH3-N，（10.7±0.3）%for TN，（19.5±1.1）%for TP，（12.7±0.3）%for BOD，（11.4±0.5）%for COD，and（37.8±2.5）%for SS，compared w ith pollutant removal rates under the continuous operation mode.As a general rule，the pollutant removal efficiency declines w ith the increase of the HLR.Experiments showed thatwhen the average HLR increased to 12 cm/d，the packed substrate in the SW IS clogged due to the excessive inflow of sewage and the permeability decreased quickly to a small value.The significantelevation of NH3-N concentration in effluent in this condition was attributed to deterioration of nitrification caused by substrate clogging.Under the interm ittent operation m ode，the NH3-N removal rate rose from（86.2±1.5）%at a RWDof 1.0 to（95.9±0.4）%at a RWDof 0.25，indicating that the oxidative condition of the substrate was improved through the drying period of the alteration operation，whichwas favorable for the nitrification process.In contrast，the TN removal rate decreased w ith the RWDdecline. Furthermore，Table 4 shows thathydraulic conductivity，ORP，and the number of nitrifying bacteria increased with the RWDdecline.On the other hand，the changes in water content andthe number of denitrifying bacteria showed an opposite trend. According to Liet al.（2013），therewas no significant difference in theammonifying bacteria numberwith the changing in R WD.

Table 3 Effect of continuous and interm ittent operation modes on pollutant removal.

Table 4 Effect of RWDon substrate characteristics.

According to previous studies（Candelaetal.，2007；Kadam et al.，2009），the biologicalmaterial attaches to the substrate surface as the influent passes through the SW IS.As the operation time goes on，the packed substrate tends to be choked by themicrobialmetabolites，which shorten the lifespan of the system.Thus，periodic resting is an effective method for removing themicrobialmetabolites and restoring the hydraulic capacity.The substrate surfaces are rested by removing them from service for an extended period of time（Moreno Escobaretal.，2005）.Second，themain reason for the low hydraulic conductivity in conventional systems is that，during the pollutant removal process，especially the nitrogen removal process，the produced gases，such as N2，CO2，and N2O，congest in and clog the packed substrate pores，thus reducing the hydraulic capacity.It is generally accepted（Nie et al.，2011；Wang et al.，2009）that the feasible hydraulic conductivity between 8.0×10-6and 7.2×10-5cm/s is essential in order for the SW IS to run smoothly and efficiently. Therefore，another function of the interm ittentoperationmode is to encourage the gases to escape from the system.According to the experimental resultsof pollutant removal（Table 3）and hydraulic conductivity（Table 4），the interm ittent operationmodew ith a RWDof 1.0wassuggested，and theHLR was chosen to be 10 cm/d.

Fig.3.Relative contribution of pretreatment and SW IS to pollutant removal.

3.2.Relative contribution to pollutant removal in pretreatment and SWIS

Fig.3 shows the overall pollutant removal efficiencies in the pretreatmentand SW ISwhen the RWDis1.0 and the HLR is 10 cm/d.The results show that the hydrolytic acidification cellwas relativelymore efficient in the removalof SS than the removal of other pollutants.The average SS removal ratewas（60.2±0.3）%in the pretreatment，but NH3-N，TN，and TP concentrations changed little during this procedure.

The overall removal rates of NH3-N and COD are 87.3% and 92.1%，respectively.More than 80%of the removal occurred in the SW IS，even though the HLR was 16.8%-50% higher than thatdescribed in previous reports（Nie etal.，2011；Wang etal.，2009）.Such a high rate of removalwasattributed to the intermittentoperationmode and the packed substrate in the SW IS.Periodic resting is conducive to the relatively high oxygen availability in the packed substrate surface and the NH3-N removal.The bio-substrate around the inflowing pipes was favorable to the grow th of microorganisms，as well as ammonia adsorption（Lietal.，2013）.After the sewage flowed in，the substrate first adsorbed refractory organic m atter，and then adsorbed organic matter was gradually converted intolow-molecularweightmatter by m icroorganisms，which were easily utilized by the denitrifier and other heterotrophs. Therefore，the removal rate of COD always gradually decreased because of adsorption saturation（Pavelic et al.，2011；Toze，2006）.In this experiment，under the intermittent operation mode，the mean removal rate for COD was（92.4±1.2）%，16.8%-21.8%higher than in the continuous operation mode，as reported in other studies（Petter et al.，2010；Zhang et al.，2007）.The SS removal process in the SWIS is based on sedimentation，adsorption，and biological processes.Previous studies have revealed that the permanent rem oval of SS usually occurs in the subsurfacew ith the effect of alum inum/iron compounds（Oladoja and Adem oroti，2006；Yang et al.，2007；Ye et al.，2008）.Furthermore，studies have shown that the SS reduction capacity decreases w ith time because them ineral sediments become fully saturated w ithin the infiltration system（Hand et al.，2008；Morkved et al.，2007；Zhang et al.，2005）.High influent SS concentration becomes the main reason for the SWIS clogging.Therefore，one encouraging method was to prolong the hydraulic detention time in pretreatment to remove more SS.Influent SS concentration less than 20 mg/L was suggested，as it can ensure a long lifespan of the SW IS（Sun and Li，2006）.

3.3.Receiving-water quality

The stream，which isadjacent to the SWIS，wasmonitored for over seven months.The average values of variables are shown in Table 5.

At the discharge point，mean concentrations of water quality indices in the effluent were 8.2 mg/L for BOD，25.3 mg/L for COD，1.2 mg/L for SS，4.8 mg/L for NH3-N，and 0.9 mg/L for TP.According to Table 5，although the pollutant concentrations of point b were slightly higher than thatof point a，the differenceswere not statistically significant（p<0.05）.Likewise，no significant differences were found between the pollutant concentrations in point a，point c，and point d.These results indicate that the stream has sufficient assim ilative capacity，and that the SW IS effluent had little negative influence on the water quality of the stream even though the HLR was ashigh as 10 cm/d.

Table 5 Water quality of receiving stream（mg/L）.

4.Conclusions

Thisstudy demonstrated theperformanceofa full-scaleSW IS in treating domestic sewage.The contribution of the interm ittent operationmodeto theencouragementfortheHLRwasexamined. Relative contribution of the pretreatment and SW IS to the pollutant removalwere exam ined.Finally，the impact of SW IS effluenton the receiving-water quality wasassessed.

The results indicate that interm ittent operation of the SW IS significantly encouraged the pollutant removal.When the RWDwas 1.0，the intermittent operation mode improved pollutant removal ratesby（13.6±0.3）%for NH3-N，（20.7±1.1）%for TN，（18.6±0.4）%for TP，（12.2±0.5）%for BOD，（10.1±0.3）%for COD，and（36.2±1.2）%for SS，compared w ith pollutant removal rates under the continuous operation mode.The pollutant removal efficiency declined w ith the increase of the HLR.With the drying days prolonged，hydraulic conductivity，ORP，the num ber of nitrifying bacteria，and the NH3-N removal rate increased，while water content，the num ber of denitrifying bacteria，and the TN rem oval rate decreased.There was no significant difference（p<0.05）between TP，BOD，and COD removal rates under different RWDs.The intermittentoperationmodew ith a RWDof 1.0was tested.Under this condition，the HLR increased to 10 cm/d. More than 80%of the removaloccurred in the SW IS for NH3-N，TN，TP，COD，and BOD.The main function of the pretreatmentwas to removemore SS to minimize the clogging risk of the SW IS.The analysis of receiving-water quality indicated that the SW IS had little negative influence on the nearby stream.

Acknow ledgem ents

Wew ish to givea special thanks to Dr.HongWang and Dr. XinWang from Shenyang University，who reviewed thispaper for its language quality.

References

American Public Health Association（APHA），2003.Standard Methods for the Examination ofWater and Wastewater.American Public Health Association，AmericanWaterWorks Association，and Water Environment Federation，Washington，D.C.

A rve，H.，Adam，M.P.，Lasse，V.，Kinga，.，Petter，D.J.，2006.A high-performance compact filter system treating domestic wastewater.Ecol.Eng. 28（4），374-379.http://dx.doi.org/10.1016/j.ecikebg.2006.06.011.

Candela，L.，Fabregat，S.，Josa，A.，Surilo，J.，Vigus，N.，Mas，J.，2007. Assessment of soil and groundwater impacts by treated urban wastewater reuse.A casestudy:application on agolf course（Girona，Spain）.Sci.Total Environ.374（1），26-35.http://dx.doi.org/10.1016/j.scitotenv.2006.12.028.

Fan，C.，Chang，F.C.，Ko，C.H.，Teng，C.J.，Chang，T.C.，Sheu，Y.S.，2009. Treatment of septic tank effluents by a full-scale capillary seepage soil biofiltration system.J.Environ.Health 71（7），56-60.

Greenan，C.M.，Moorman，T.B.，Kaspar，T.C.，Parkin，T.B.，Jaynes，D.B.，2006. Comparing carbon substrates fordenitrification ofsubsurfacedrainagewater. J.Environ.Qual.35，824-829.http://dx.doi.org/10.2134/jeq2005.0247.

Hand，V.L.，Lloyd，J.R.，Vaughan，D.J.，Wilkins，M.J.，Boult，S.，2008. Experimentalstudiesof the influenceofgrain size，oxygen availability and organic carbon availability on bio-clogging in porousmedia.Environ.Sci. Technol.42（5），1485-1491.http://dx.doi.org/10.1021/es072022s.

Howarth，R.W.，Sharpley，A.W.，Walker，D.，2002.Sources of nutrient pollution to coastal waters in the United States:implications for achieving coastal water quality goals.Estuaries 25，656-676.http://dx.doi.org/ BF02804898.

Kadam，A.M.，Nemade，P.D.，Oza，G.H.，Shankar，H.S.，2009.Treatment of municipal wastewater using laterite-based constructed soil filter.Ecol. Eng.35（7），1051-1061.http://dx.doi.org/10.1016/j.ecoleng.2009.03.008.

Li，Y.H.，Li，H.B.，Sun，T.H.，Wang，X.，2011.Study on nitrogen removal enhanced by shunt distributing wastewater in a constructed subsurface infiltration system under interm ittent operation mode.J.Hazard.M ater. 189（1-2），336-341.http://dx.doi.org/10.1016/j.hazmat.2011.02.039.

Li，Y.H.，Li，H.B.，Wang，H.，Wang，X.，Zou，Y.，Sun，T.H.，2013.Comparisonof the treatmentperformanceofbio-substrate based andmeadow brown soilbased subsurface infiltration systems for domesticwastewater treatment.Water Sci. Technol.67（3），506-513.http://dx.doi.org/10.2166/wat.2012.576.

Lowe，K.S.，Siegrist，R.L.，2008.Controlled field experiment for performance evaluation of septic tank effluent treatment during soil infiltration.J.Environ.Eng.134（2），93-101.http://dx.doi.org/10.1061/（ASCE）0733-9372（2008）134:2（93）.

Moreno Escobar，B.，Gomez Nieto，M.A.，Hontoria García，E.，2005.Simple tertiary treatmentsystems.WaterSci.Technol.WaterSupply 5（3-4），35-41. Morkved，P.T.，Dorsch，P.，Bakken，L.R.，2007.The N2O product ratio of nitrificationand itsdependenceon long-term changesinsoilpH.SoilBiol.Biochem. 39（8），2048-2057.http://dx.doi.org/10.1016/j.soilbio.2007.03.006.

Nie，J.Y.，Zhu，N.W.，Zhao，K.，Wu，L.，Hu，Y.H.，2011.Analysis of the bacterial community changes in soil for septic tank effluent treatment in response to bio-clogging.Water Sci.Technol.63（7），1412-1417.http:// dx.doi.org/10.2166/wst.2011.319.

Oladoja，N.A.，Ademoroti，C.M.A.，2006.The use of fortified soil-clay as onsite system for domestic wastewater purification.Water Res.40（3），613-620.http://dx.doi.org/10.1016/j.watres.2005.11.031.

Pan，J.，Yu，L.，Wang，Y.，2012.Clogging process caused by organic particle accumulation and biofilm grow in subsurfacewastewater infiltration system.J.Donghua Univ.2（29），187-191.http://dx.doi.org/1672-5220（2012）02-0187-06.

Pavelic，P.，Dillon，P.J.，Mucha，M.，Nakai，T.，Barry，K.E.，Bestland，E.，2011. Laboratory assessment of factors affecting soil clogging of soil aquifer treatment systems.Water Res.45（10），3153-3163.http://dx.doi.org/ 10.1016/j.watres.2011.03.027.

Petter，D.J.，Tore，K.，Adam，M.P.，Trond，M.，Kinga，A.，Carlos，A.A.，Arve，H.，Lena，J.，Daniel，H.，Hans，B.，et al.，2010.Filter bed systems treating domestic wastewater in the Nordic countries-Performance and reuse of filtermedia.Ecol.Eng.36（12），1651-1659.http://dx.doi.org/ 10.1016/j.ecoleng.2010.07.004.

Qian，Y.，Wen，X.H.，Huang，X.，2007.Developmentand application of some renovated technologies for municipal wastewater treatment in China. Front.Environ.Sci.Eng.China 1（1），1-12.http://dx.doi.org/10.1007/ s11783-007-0001-9.

Robertson，W.D.，2010.Ratesofnitrate removal inwoodchipmediaof varying age.Ecol.Eng.36，1581-1587.http://dx.doi.org/S0925857410000340.

Stewart，G.C.，Louis，A.S.，2010.Nitrate removal and hydraulic performance of organic carbon for use in denitrification beds.Ecol.Eng.36，1588-1595.http://dx.doi.org/S0925857410000704.

Sun，T.H.，Li，X.F.，2006.M ethods for Eco-treatmentand Resources Reuse of Domestic Wastewater.Chem ical Industry Press，Beijing（in Chinese）.

Toze，S.，2006.Reuseof effluentwater-benefitsand risks.Agric.WaterManag. 80（1-3），147-159.http://dx.doi.org/10.10.16/j.agwat.2005.07.010.

Wang，P.F.，Ding，Q.S.，Ding，W.M.，Pan，G.X.，2009.A measurementmethod of soil porosity.Exp.Technol.Manag.26（7），50-51（in Chinese）.

Yang，Q.，Peng，Y.Z.，Liu，X.H.，Zeng，W.，Mino，T.，Satoh，H.，2007. Nitrogen removal via nitrite from municipal wastewater at low temperatures using real-time control to opim ize nitrifying communities. Environ.Sci.Technol.41（23），8159-8164.http://dx.doi.org/10.1021/ es070850f.

Ye，C.，Hu，Z.B.，Kong，H.N.，Wang，X.Z.，He，S.B.，2008.A new soil infiltration technology for decentralized sewage treatment:two-stage anaerobic tank and soil trench system.Pedosphere 18（3），401-408.http:// dx.doi.org/10.1016/S1002-0160（08）60031-4.

Zhang，J.，Huang，X.，Liu，C.X.，Shi，H.C.，Hu，H.Y.，2005.Nitrogen removal enhanced by intermittentoperation in a subsurfacewastewater infiltration system.Ecol.Eng.25（4），419-428.http://dx.doi.org/10.1016/ j.ecoleng.2005.06.011.

Zhang，L.B.，Xing，M.Y.，Wu，Y.F.，Huang，Z.D.，Yang，J.，2011.Spatial distributions of biofilm properties and flow pattern in NiiM i process. Bioresour.Technol.102（2），1406-1414.http://dx.doi.org/10.1016/ j.biortech.2010.09.047.

Zhang，Z.Y.，Lei，Z.F.，Sugiura，N.，Xu，X.T.，Yin，D.D.，2007.Organics removal of combined wastewater through shallow soil infiltration treatment:a field and laboratory study.J.Hazard.M ater.149（3），657-665. http://dx.doi.org/10.1016/j.jhazmat.2007.04.026.

This work was supported by the National Natural Science Foundation of China（Grant No.51108275），the Program for Liaoning Excellent Talents in Universities（LNET）（Grant No.LJQ2012101），the Program for New Century Excellent Talents in Universities（GrantNo.NCET-11-1012），the Science and Technology Program of Liaoning Province（Grants No.2011229002 and 2013229012），and the Basic Science Research Fund in Northeastern University（Grants No.N130501001 and N140105003）.

*Corresponding author.

E-mail address:graceli_2003@163.com（Hai-bo Li）. Peer review under responsibility of HohaiUniversity.

http://dx.doi.org/10.1016/j.w se.2015.01.008

1674-2370/©2015 Hohai University.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http:// creativecommons.org/licenses/by-nc-nd/4.0/）.