Feasibility of using constructed treatment wetlands for municipal wastewater treatment in the Bogotá Savannah, Colombia

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  The main water bodies in the Bogotá Savannah have been seriously polluted due to the mismanagement of domestic, agricultural, and industrial wastewater. While there are a number of wastewater treatment facilities in the region, most do not function
  This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institutionand sharing with colleagues.Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:http://www.elsevier.com/copyright  Author's personal copy Ecological Engineering 35 (2009) 1070–1078 Contents lists available at ScienceDirect Ecological Engineering  journal homepage: www.elsevier.com/locate/ecoleng Feasibility of using constructed treatment wetlands for municipal wastewatertreatment in the Bogotá Savannah, Colombia Mauricio E. Arias ∗ , Mark T. Brown Howard T. Odum Center for Wetlands, Department of Environmental Engineering Sciences, University of Florida, United States a r t i c l e i n f o  Article history: Received 18 July 2008Received in revised form 27 February 2009Accepted 24 March 2009 Keywords: Municipal wastewaterEmergyDeveloping countriesAppropriate technology a b s t r a c t The main water bodies in the Bogotá Savannah have been seriously polluted due to the mismanagementof domestic, agricultural, and industrial wastewater. While there are a number of wastewater treatmentfacilitiesintheregion,mostdonotfunctionproperly.Thereisagreatneedforinexpensiveandsustainablewastewatertreatmentsystemsthatarenottechnologicallysophisticatedandthatdonotrequireintensivemanagement.Themaingoalofthisstudywastoquantifytheperformanceandsustainabilityoftreatmentwetlandsandexistingwastewatertreatmentsystemsinthisregion.Usingdatafromtheliterature,atreat-mentwetlandmodelwasdeveloped,whichfocusedonpollutantremoval.Themodeledperformancewascomparedtoasystemofwastestabilizationpondsandasequencingbatchreactor.Thethreesystemsweresubject to cost analysis and an emergy evaluation, leading to the assessment of indicators of cost-benefitfor comparison. The economic analysis suggested that the net annual cost of the treatment wetland wasUS$ 14,672, compared to US$ 14,201 for the stabilization ponds and US$ 54,887 for the batch reactor.The emergy evaluations show that the ponds have the lowest annual emergy flow (6.65+16sej/yr), fol-lowed by the constructed wetland (2.88E+17sej/yr) and the batch reactor (8.86E+17sej/yr). These resultswere combined to estimate treatment ratios (contaminants removed per lifetime cost, and contaminantsremovedpertotalemergy),costratios(costpervolumeofwater,annualcostpercapita,andconstructioncost per capita), and emergy ratios (treatment yield, renewable emergy, lifetime emprice, constructionemprice, non-renewable emergy, empower density, environmental loading, total emergy per volume of water, and emergy per capita).© 2009 Elsevier B.V. All rights reserved. 1. Introduction Colombia is in great need of low-cost, low-maintenancewastewater management strategies that take advantage of thiscountry’sclimaticconditionsandthatofferagoodinvestmentforacountry where capital for infrastructure is scarce. Bogotá, the cap-ital and largest city of Colombia, creates a large demand on theresourcesofthesurroundingruralarea,calledtheBogotáSavannah.Asoftheyear2005,therewere27municipalwastewatertreatmentplants (WWTPs) in the region, which served nearly 600,000 peo-ple. Of the 27 WWTPs, 16 were waste stabilization ponds (WSPs),7 were activated sludge systems, 3 were anaerobic reactors, and 1was a sequencing batch reactor (CAR, 2003).Mechanical treatment systems are maintenance- and energy-intensive (Tchobanoglous et al., 2003); consequently, their ∗ Corresponding author at: Howard T. Odum Center for Wetlands, University of Florida, Phelps Lab, P.O. Box 116350, Museum Road, Gainesville, FL 32611, UnitedStates. Tel.: +1 352 392 2429; fax: +1 352 392 3624. E-mail address:  moriche@ufl.edu (M.E. Arias). performance is affected when these requirements cannot beproperly provided. Thus, it should be clear that in regionswhere mechanical treatment technologies cannot be effectivelymaintained, promoting less energy-intensive wastewater (WW)technologiescouldresultinimprovedwaterquality,benefitingthehealth, economy, and aesthetics of the region.Theobjectiveofthisstudywastoevaluatetheperformanceandsustainability of a model constructed wetland treatment system(CWTS),andtocomparethisoptiontoconventionalsystems(WSPsandabatchreactor)currentlyusedformunicipalWWtreatmentinthe Bogotá Savannah. The hypothesis of this study is that a CWTSwouldyieldsignificantimprovementsinpollutantremovalincom-parisontoWSPs;likewise,theCWTSshouldresultinamuchlowermonetary and resources investment in comparison to a conven-tional treatment system. 1.1. Previous studies Few CWTS have been studied and documented in Colombia,despitetherecognizedbenefitsoflow-cost,year-longplantgrowthand bacterial activity in tropical climates (Okurut et al., 1999). 0925-8574/$ – see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.ecoleng.2009.03.017  Author's personal copy M.E. Arias, M.T. Brown / Ecological Engineering 35 (2009) 1070–1078  1071 Williams et al. (1999) published the first study on the design andinitial performance of a small CWTS in Chinchina, Province of Cal-das(designflowrate=45L/min,totalconstructionarea=4000m 2 ).In addition, several pilot studies have been conducted in Colom-bianuniversities(e.g.,LaraandVera,2005;Paredes,2005;Casta˜no,2005). Nonetheless, no large-scale, municipal WW treatment wet-land had been documented to the date of this study.Previous studies have evaluated WW treatment alternativesusingsimilartoolsastheonesusedinthisstudy(i.e.,treatmentper-formance,monetarycost,andemergy).Forinstance,Nelson(1998)designedandevaluatedtwosubsurfaceflow(SSF)wetlandsinMex-ico. He showed that these systems were efficient at removing anumber of wastewater constituents, including Biological OxygenDemand (BOD), Fecal coliform (FC), Total Nitrogen (TN), and TotalPhosphorus(TP).Thetotalemergyofthesetreatmentwetlandswasreported to be 3.54E+17 solar emjoules (sej) per year (see Section2.4 for explanation of this unit), which included 3.45E+17sej/yrof raw wastewater flow, and 8.7E+15sej/yr of all environmental,construction, and operation flows combined. The emergy requiredfor operation of a comparable conventional packaged plant was3.7E+15sej/yr, which was eighteen times greater than the opera-tional emergy of the treatment wetlands.In a different study, Geber and Björklund (2001) compared aconventional WWTP, a conventional plant combined with a CWfor tertiary treatment, and a natural wetland in Sweden. Thethree systems realized similar pollutants reduction, and similartotal emergy use per person equivalent (p.e.), ranging from 154to 180E+12sej/p.e. Also, a large variation in the ratio of purchasedgoods to environmental inputs (environmental loading ratio) wasfound, with a value of 9 for the natural wetland and 3056 for theconventionalWWTP.Despitethislargedifference,theauthorscon-cludedthatthenaturaltreatmentalternativesstudieddidnotfullysubstitute the purchased emergy with free environmental inputs.Furthermore,Behrend(2007)performedemergyevaluationsof seven wastewater treatment alternatives for the state of Georgia(USA). The environmental loading ratio varied greatly for the sys-tems evaluated, with septic tanks having the lowest (383) and awater reclamation facility with lime phosphorus removal havingthegreatest(101,176).Afacilitywithconventionalsecondarytreat-mentfollowedbyatreatmentwetlandfellinthemiddleoftherangeofenvironmentalloadingratioswithavalueof58,370(202withoutthe wastewater input). This facility treated an average of 1.4m 3 /s,and yielded a total emergy of 1.02E+22 (3.56E+17 if wastewaterinput is excluded from the calculation). 2. Methods  2.1. Study sites Two sites with municipal WWTPs were selected for the evalua-tions performed during this study. The first site was Tabio, located50kmnorthwestofBogotáwithapopulationof14,000,andanaver-age temperature of 14 ◦ C. This municipality has a WSP designed totreat 20L/s of WW. The 3.4ha treatment plant was constructed in1992, and consists of a screen, a sedimentation tank, an anaerobicbasin, and two series of facultative lagoons with two basins each(CAR, 2003).ThesecondsitewasthemunicipalityofLaCalera,located30kmeast of Bogotá. A Sequencing Batch Reactor (SBR) treatment plantwasconstructedin2002withadesignflowof36.5L/sandadesignpopulation of 16,000. The system consists of primary treatmentwith a manual screen and a sedimentation tank, followed by sec-ondary treatment with two reactor tanks, a sludge digestor, andsludge drying beds (CAR, 2003). Water quality and flow analysesfromthesefacilitieswereperformedandreportedbythelocalenvi-ronmental protection agency (Corporación Agrónoma Regional deCundinamarca, CAR).  2.2. Constructed wetland treatment system design TheCWTSmodelwasassumedtobeatthelocationoftheTabioWSPs. The hypothetical system consisted of screens, a sedimenta-tiontank,andanaerobicbasin,butinplaceoffacultativelagoons,acombinationofsubsurfaceflowandsurfaceflow(SF)wetlandunitswere modeled.TherawwastewaterqualitydataforTabioprovidedbyCARwereused to size the proposed wetland system. The area used in thedesign determined from the 3.4 hectares utilized by the currentplant, minus 2900m 2 occupied by the anaerobic basin, and minus30% of the extra area that accounted for pretreatment structures,open area, and others. A combination of SF and SSF wetland cellswas deemed necessary to provide sufficient treatment with min-imal costs. The CWTS model was subject to a sensitivity analysisto determine the effect of the system configuration on the overallfeasibilitystudy.Thisanalysiswasdonebyestimatingthepollutantremoval and the cost for different area distributions between SSFand SF wetlands; the configuration used in this study was found atthe point where maximum pollutant removal and minimum costintersected.Using the estimated wetland areas, the effluent concentrationat each wetland unit was calculated using the  k – C  * model (Kadlecand Knight, 1996): C  e  =  C   ∗+ ( C  i  − C  ∗ )exp   − kA 0 . 0365 Q    (1)where  C  e =outlet concentration (mg/L),  C  i =inlet concentration(mg/L),  C  *=background concentration (mg/L),  A =wetland area(ha), Q  =waterflowrate(m 3 /day),and k =first-orderarealratecon-stant (m/yr). Values of   k  used were the global averages proposedby Kadlec and Knight (1996), which are based on a large databaseofoperatingtreatmentwetlandsinthetemperateclimateofNorthAmerica and Europe. Even though treatment wetlands in the trop-ics could perform more efficiently, design parameters for CWTS inColombiahavenotbeendocumented,andthereforeusingthepub-lished constants proposed by Kadlec and Knight (1996) is the mostconservative approach for this evaluation.The pollutant removal efficiencies of the pretreatment andprimary treatment units were not modeled, but assumed fromtypical values found in the literature (Ramírez and Romero, 1997;Tchobanoglous et al., 2003).  2.3. Cost evaluations A cost evaluation was performed for each of the treatmentoptions studied. These evaluations included the major compo-nents of construction cost, and annual operation and maintenance(O&M). All costs were calculated for 2003 Colombian pesos, andthen converted to US dollars using the average 2003 exchange rateof 2877.55 Colombian pesos per US dollar. Unit costs for the CWTSand the WSPs came from a Colombian construction costing guide(Construdata,2003),CAR(2003),orotherwiseestimatedfromlocal market costs. The construction costs for the SBR were extractedfromabudgetrequestedbyCAR(Aquavip,2000).Onlycostsrelatedtotheconstructionofthetreatmentunitswereusedfromthisbud-get. Annual O&M costs for this facility were acquired from CAR documents or estimated from typical local market costs. The netannual cost of treatment was estimated as:Netannualcost  =  O&M+Constructionlifetime (2)  Author's personal copy 1072  M.E. Arias, M.T. Brown / Ecological Engineering 35 (2009) 1070–1078  Table 1 Description of indicators calculated.Indicator Units Calculation PreferenceTreatment ratiosContaminant (C) removed (e.g. BOD, SS, coliform, TP) per lifetime cost kg C/US$  [ C  in ] − [ C  out ]lifetimecost  ↑ Contaminant (C) removed (e.g. BOD, SS, coliform, TP) per total emergy kg C/sej  [ C  in ] − [ C  out ]totalemergy  ↑ Cost ratiosUnit costs (X) required (net annual, O&M, construction) per m 3 of water US$/m 3 /yr  X  designflow  ↓ Net annual cost per capita US$/p.e/yr  lifetimecostdesignpopulation  ↓ Construction cost per capita US$/p.e  constructioncostdesignflow  ↓ Emergy ratiosTreatment yield ratio Unitless  WW emergy − systemoutflowpurchasedgoodsandservices  ↑ Renewable emergy Unitless  environmentalresources + WW emergy lifetimecost  ↑ Lifetime emprice sej/US$  totalemergylifetimecost  ↑ Construction emprice sej/US$  totalemergyconstructioncost  ↑ Non-renewable emergy. Unitless  goodsandpurchasedservicestotalemergy  ↓ Empower density sej/yr/m 2 totalemergyarea  ↓ Environmental loading ratio Unitless  goodsandpurchasedservicesenvironmentalresources + WW  ↓ Total emergy per m 3 sej/m 3 totalemergyannualWWflow  ↓ Total emergy per inhabitant sej/p.e.  totalemergypopulation  ↓↑ : Largest ratio is preferred;  ↓ : smallest option is preferred. where net annual cost, construction, and annual O&M were calcu-lated in 2003 dollars. Lifetime refers to the facility design lifetime,assumed to be 25 years for the three systems.  2.4. Emergy evaluations For the purpose of this study, sustainability was addressed asthe appropriate use of resources to provide the service of wastew-ater treatment. In order to provide numerical quantities to thisconcept, emergy syntheses of the three wastewater systems wereperformed.Emergywasdefinedas“theavailabilityofenergyofonekind that is used up in transformations directly and indirectly tomake a product or service” (Odum, 1996). Emergy synthesis is an appropriate accounting method for environmental decision mak-ing,becauseitnotonlyaccountsfortheenergyflowsdrivingthesesystems,butitalsorecognizesthattheseenergyflowshaveaqualitycomponentthataccountsforthehierarchicaldistributionofenergyin systems. This accounting method has been recently used in anumber of environmental decision making applications, includ-ing regional sustainability (Lei et al., 2008) ecosystem restoration(Ton et al., 1998; Lu et al., 2006), solar power evaluation (Paoli et al., 2008a), small marinas sustainability (Paoli et al., 2008b), and wastewater technology assessment (Geber and Björklund, 2001;Siracusa and La Rosa, 2006).Emergy is typically measured in solar emjoules, which can beestimated from physical flows using three different unit emergyvalues (UEV):  Transformity ,  specific emergy , and  emergy per unit money  (Odum, 1996).  Transformity  is the energy used in the trans-formation of available energy, normally expressed in units of emjoulesperjoule(sej/J). Specificemergy istheenergyusedincon-centrating certain amount of mass, and it has units of emergy perunit mass (e.g., sej/g). The  emergy per unit money  is the amountof emergy represented by a unit of currency. This is measured insej/US$,anditisestimatedfromtheratioofemergyusetoeconomicproduct (typically gross domestic product (GDP)) at the nationalscale.Systems diagrams were created for the three treatment options(Appendix A). These diagrams show the main energy fluxes andstoragesforagivensystemandtheinteractionsamongthem. 1 Theflows in and out of the systems diagrams were the basic compo-nentsanalyzedintheemergyevaluationtables.Itemsinthetablesfor each system were organized in four categories: environmentalresources, wastewater, purchased goods and services, and systemoutflows.Thefirstcategoryincludessun,wind,andrain.Bothwaterinflows (rain and WW) were divided into four components: waterchemical potential, organic matter, total phosphorous, and totalnitrogen. The third category includes construction materials, andO&M components such as electricity and labor. The fourth cate-gory refers mainly to the treated water (including the same fourcomponents as for the environmental resources), and evapotran-spiration.Datausedfortheemergyevaluationscamefromseveralsources.Mass, energy, and money flows were converted to their respectivestandard units from data acquired from the CWTS performanceanalysis, the cost analysis, CAR reports, journal articles, or esti-matedfromtypicalvaluesobservedintheregion.UEVscamefromOdum(1996),GeberandBjörklund(2001),andemergyevaluations compiled at the Center for Environmental Policy at the Universityof Florida, Gainesville (Brandt-Williams, 2002; Brown and Bardi,2001; Buranakarn, 1998; McGrane, 1994; Odum et al., 1983, 1987;Vivas, 2004). All UEVs were corrected to the most recent emergybaseline as suggested by Brown and Bardi (2001).Eachtableincludesacalculationoftotalemergyofenvironmen-tal resources, total emergy of goods and purchased services, andtotal emergy. The maximum flow of environmental resources wasestimatedtobethetotalemergyforthiscategorytoavoiddoubled-counting. Emergy of goods and purchased services was calculatedas the sum of all the inflows from that category; the total systememergy was then calculated as the sum of the total environmental 1 For a description of the diagramming language see Odum (1994).  Author's personal copy M.E. Arias, M.T. Brown / Ecological Engineering 35 (2009) 1070–1078  1073  Table 2 Design parameters, pollutant concentrations, and percent removal at each stageof the CWTS system. All concentrations expressed as mg/L, except for  E. coli (CFU/100mL). resourcesandthetotalgoodsandpurchasedservices.Eventhoughthe WW emergy flows were quantified, these numbers were notincluded in the total emergy calculation. This exclusion was madebecause WW was considered a flow-through component, whichrather than being an input, it is the element that the entire systemis providing a service to.  2.5. Indicators Oncetheperformance,cost,andemergyevaluationsweredeter-mined, several indicators were calculated to compare the threeoptions (Table 1). These indicators were divided into three cate-goriesaccordingtotheunitintheirdenominator:Thefirstcategorywas treatment indicators, and their purpose is to show how effec-tive the systems are at removing the pollutants with respect tocost and emergy. Nitrogen was not included because no data wereavailable for the WSPs and the SBR. The second category was costindicators, which showed how economically expensive the treat-ment is per unit of water treated and per inhabitant served. Thethird category was emergy indicators, which show how energeti-cally expensive each alternative was. 3. Results and discussion  3.1. CWTS design and performance Findings from this study indicate that it is feasible to designa CWTS that satisfies size restrictions and pollutant removal effi-ciencies similar to existing WSPs in the region. The hypotheticalCWTS was composed of the existing pretreatment units and pri-maryanaerobiclagoon,followedbyoneseriesofSSFCWunitsandoneseriesofSFCWunits(Table2).Thetotalareawasmaintainedat3.4ha, with 0.4ha and 1.7ha for the SSF and the SF basins, respec-tively. One concern with the design parameters calculated is thatsome are out of the typical range of design values for these sys-tems (Kadlec and Knight, 1996). The detention times estimated forthe SSF and the SF CW were 0.6 and 4.5 days, respectively, andthe loading rates were 40 and 10cm/day. Detention times in SSFand SF are typically 2–4 and 7–10 days respectively, and loadingrates are typically 8–30 and 1.5–6.5cm/day. Nonetheless, there isevidence that CWTS in sub-/tropical weather have performed ade-quatelyevenwithlessconservativedesignparametersthatareoutof these recommended ranges (e.g., Yang et al., 1995; Greenwayand Woolley, 1999). Another concern with this model is the useofthearealrateconstants( k ),whichideally,shouldbedeterminedfromtreatmentwetlandsoperatingundersimilarconditions.How-ever,asithasbeenpointedoutpreviouslyinthispaper,large-scaletreatment wetlands (if existing) have been poorly documented inColombiaanditsneighboringcountries,anduntilsufficientdataaregenerated, treatment wetlands in this region should be designedusing information and tools established from databases of large-scale functional treatment wetlands (most of which are in NorthAmerica and Europe).A sensitivity analysis (Appendix A) revealed that a ratio of SSF to SF wetland area of 1–4 yielded appropriate removal whilemaintaining construction costs as low as possible. This sensitiv-ity analysis was a crucial step that had much impact in the otheranalyses performed in this study. The finding of this point of tradeoff between system performance (represented by BOD removal)andresourcesutilization(representedbygravelcost)wasthebasisfor the conceptual design of this CWTS, and without determiningthis point, it is likely that the area distribution of between SSF andSF cells would have been arbitrarily set, and hence, the potentialthat CWTS have to provide good performance with appropriateresources investment would have been underestimated.The CWTS treatment performance was compared with theremoval efficiencies of the WSPs and the SBR (Table 3), and it wasestimated that the proposed system has the potential of achiev-ing overall higher pollutant removal efficiency, particularly thoseof greatest concern, BOD and TSS, and to a lesser extent, nutrients.Removalof  E.coli wastheonlyparameterforwhichtheWSPswereshown to achieve a higher efficiency than CWTS. Although the dif-ferenceinperformanceisnotlarge,thisislikelytheresultoflongerdetentiontimesinWSPsduetogreaterstoragevolume.Itshouldbenoted that the CWTS performance estimates are based on resultsfrom an empirical model that has been proven accurate for manytreatment wetlands (Kadlec and Knight, 1996), but do not repre-  Table 3 Estimatedpercentremovalofpollutantsforthethreetreatmentsystemsevaluated.Constituent CWTS WSPs SBR SS 97 79 73BOD 92 86 79 E. coli  98 99.5 84TN 62 – –NH 4  62  − 8 –TP 47  − 35 30
Related Search
Similar documents
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!