Constructed Wetlands for Septage Treatment

8/6/2019 Constructed Wetlands for Septage Treatment

1/17

Constructed Wetlands for Septage Treatment Towards Effective faecal Sludge Management 1

Thammarat Koottatep*, Chongrak Polprasert*, Nguyen Thi Kim Oanh*, NarongSurinkul*, Agnes Montangero** and Martin Strauss**

* Urban Environmental Engineering and Management Program (UEEM),School of Environment, Resources and Development (SERD)Asian Institute of Technology (AIT),P.O. Box 4, Klong Laung, Pathumthani 12120, Thailand

** Department of Water and Sanitation in Developing Countries (SANDEC),Swiss Federal Institute for Environmental Science and Technology (EAWAG)P.O. Box 611, CH-8600 Dbendorf, Switzerland

Author to whom correspondence should be addressed: Thammarat Koottatep at [email protected] (Tel. +66-2-524 61 88 Fax +66-2-524 56 25) or: Martin Strauss at [email protected] (Tel. +41-1-823 50 20 Fax +41-1-823 53 99)

ABSTRACTThe majority of urban dwellers in Africa depend on on-site sanitation (OSS) systems for excreta disposal.There is a considerable gap-in-knowledge on strategies for sustainable management and on treatment optionsfor faecal sludges evacuated from OSS, which are appropriate for developing countries. The authors providean overview of the situation, challenges, and selected strategies in handling FS. Treatment options are

presented. The authors describe in detail pilot-scale investigations on using constructed wetlands (CW) for the treatment of septage. The experiments are being conducted at AIT, Bangkok since 1997. The plantconsists of three beds planted with cattail ( Typha ). A loading rate of 250 kg TS/m 2.yr was found optimum for the type of septage treated. Intermittent percolate impoundment is required to prevent cattail wilting duringdry weather. 65 % of the septage liquid passes through the underdrain and 35 % is evapotranspirated. The

beds have been operated for nearly four years with unimpaired bed permeability. Accumulated solids are lowin viable helminth eggs and satisfy expedient sludge quality standards for agricultural use. Compared withconventional sludge drying beds, CW require a much-reduced frequency at which dewatered biosolids needto be removed from the bed.

KEYWORDS

Cattails; constructed wetlands; faecal sludge; septage; management; treatment; on-site sanitation

LIST OF ABBREVIATIONS

1 Paper presented at the IWA 8 th Int. Conference on Weetlands Systems for Water Pollution Control, Arusha,Tanzania, Sept. 15-19.

BOD Biochemical oxygen demandCW Constructed wetlandsFS Faecal sludge

N Nitrogen NH 4 Ammonium NO 3 NitrateOSS On-site sanitation

SLR Solids loading rateSS Suspended solidsTCOD Total (unfiltered) chemical oxygen demand

TKN Total Kjeldahl nitrogenTS Total solidsTVS Total volatile solidsWWTP Wastewater treatment plant

FAECAL SLUDGE MANAGEMENT SITUATION, CHALLENGES ANDPLANNING

Situation and Challenges

In urban areas of developing countries, on-site sanitation (OSS) systems predominate over water- borne, sewered sanitation (Table 1). They comprise unsewered family and public toilets, aqua

privies and septic tanks. In sub-Saharan Africa, > 75 % of houses in large cities and up to 100 % in


2/17

Table 1 Proportion of urban populations served byon-site sanitation systems (OSS)

City or country % of inhabitants served by on-site sanitation systems

Ghana 85Bamako (Mali) 98

Tanzania > 85Manila 78Philippines (towns) 98Bangkok 65Latin America > 50

towns are served by on-site sanitationfacilities (Strauss et al. [1 ]). These willcontinue to play an important role for excreta disposal as area-wide sewerageis not affordable. In the USA, 25 % of

dwellings are served by septic tanks.The collection of the faecal sludges fromOSS installations and their haulage in largecities in Asia are faced with immense

problems. In the absence of long-termurban planning and/or failing

enforcement of existing zonal plans have lead to a lack of feasible landfilling or treatment sites atreasonable haulage distance. Emptying services are poorly managed and often hardly affordable tocustomers, due also to the lack of entrepreneurial competition or due to cartelization byentrepreneurs. A minor fraction (< 10 % ?), only, of the faecal sludges accumulating in on-sitesanitation installations are formally collected and discharged or treated. Where treatment schemesexist, charges are usually levied for each load of FS delivered to the plant by private collectors. As

a consequence, faecal sludge is often dumped at non-designated sites to avoid fee paying. This maygo unavenged due to lack of competition among collection entreprises, lack of adequateenforcement and corruption. Innovative incentive, fee levying and licensing procedures wouldcontribute considerably to avoiding such problems and to rendering FS management moresustainable.

FS Management Planning

Current FS management has resulted from the historical development in excreta handling (which,often, is linked to agricultural practices), general socio-economic developments and purposeful

planning efforts, which indirectly or directly affect the choice of sanitation systems. Improving onand finding appropriate strategies and solutions in FS management must, thus, be dealt with inconjunction with both unplanned and planned urban and peri-urban development, institutionalsettings, jurisdictional conditions, and expected future sanitation infrastructure and service

provision.

In short, an FS management concept should be based on the assessment of:

Existing sanitary infrastructure and trends Current FS management practices and their shortcomings Stakeholders customs, needs and perceptions regarding FS management and use Environmental sanitation strategy Prevailing socio-economic, institutional, legal and technical conditions, and The general urban development concept

(Klingel [2 ]; Klingel et al. [3 ]). Based on an FS management concept, FS treatment objectives maythen be formulated and, consequently, feasible treatment options be evaluated.

In most places, a large array of technical, economic and institutional/organizational measures arerequired to improve the FS management situation. Given the difficulties in collecting FS and inhauling it across cities to designated disposal and treatment sites, decentralizing FS management isall-important. The devising of modest-scale satellite treatment plants (Fig.1) and of neighbourhood or condominial septic tanks (Fig.2) may contribute significantly to reducingindiscriminate dumping of FS and, hence, to reducing health and pollution risks. However, everycity has to be taken at its own merits, given the great variability of spatial settings, sanitationinfrastructure and planning mechanisms.


3/17

Making sure that collected FS will actually be hauled to the designated treatment sites is one of thegreatest managerial and institutional challenges. It is now a widely accepted paradigm thatsustainable environmental sanitation may be achieved or enhanced only by applying appropriateincentive and sanctioning structures. Where fees are payable by private entrepreneurs for discharging FS to designated treatment or disposal sites, municipalities must devise an effectivesanctioning system in case FS is discharged at non-designated points, e.g., by not renewing servicecontracts with violators. Alternatively, such fee structures and cost recovery/financial proceduresshould be developed, by which entrepreneurs will receive reimbursement for FS delivered to thelegally designated treatment or disposal site. Jeuland (2002) [4 ] has proposed such alternative

procedures to secure sustainability of FS management and treatment in the planned manner for thecity of Bamako, Mali.

FAECAL SLUDGE PER-CAPITA QUANTITIES AND CHARACTERISTICS

Per-capita quantities

Table 2 contains the daily per capita volumes and loads of organic matter, solids and nutrients in faecal sludges collected from septic tanks and pit latrines, as well as from low or zero-flush,unsewered public toilets. Values for fresh excreta are given for comparative purposes. The figuresare overall averages and may be used for planning and preliminary design. Actual quantities may,however, vary from place to place.

Table 2 Daily per capita volumes; BOD, TS, and TKN quantities of different types of faecal sludges(Heinss et al. 1998) [5 ]

Where nematode infections are not endemic and, hence, eggs may be found at insignificantconcentrations only in excreta and FS, bacterial pathogens (e.g. Salmonellae spp. ) or

bacteriophages might be used as indicators-of-choice.

Communal instead of individual septic tanks

Allows to cater for better access to septic tanksand, hence, for more effective FS management

Fig. 2 The use of communal septic tanks astrategic tool to facilitate effective FS collection

What scale for FS treatment:

centralized or semi-centralized ?

Objective:

minimize overall cost for collection, haulage and treatmentwhile guaranteeing safety in FS handling, use or disposal

FSTP

Fig. 1 Satellite FS treatment a strategic tool tominimize cost, indiscriminate dumping, healthrisks and water pollution

Parameter Septage 1 Public toilet sludge 1 Pit latrine sludge 2 Fresh excreta

BOD g/capday 1 16 8 45

TS g/capday 14 100 90 110

TKN g/capday 0.8 8 5 10

Volume l/capday 1 2(incl. water for toilet cleansing)

0.15 - 0-20 1.5(faeces and urine)

1 Estimates are based on a faecal sludge collection survey conducted in Accra, Ghana.

2 Figures have been estimated on an assumed decomposition process occurring in pit latrines. According to the frequentlyobserved practice, only the top portions of pit latrines (~ 0.7 ... 1 m) are presumed to be removed by the suction tankers sincethe lower portions have often solidified to an extent which does not allow vacuum emptying. Hence, both per capita volumesand characteristics will range higher than in the material which has undergone more extensive decomposition.


4/17

Fig.3 Septage characteristics in Accra,Bangkok and Manila they are widelydifferin

0

20

40

60

80

CODtot [g/l] TS [g/l]

Accra

Bangkok

Manila

FS characteristics

In contrast to sludges from WWTP and to municipal wastewater, characteristics of faecal sludgediffer widely by locality (from household to household; from city district to city district; from cityto city).

Fig. 3 illustrates the vast differences of septagecharacteristics in the three cities of Accra,Bangkok and Manila (Montangero and Strauss2002) [6 ] and Fig. 4 depicts the differences insolids contents of faecal sludges, WWTPsludges and tropical wastewater (for wastewater characteristics, see Mara 1978) [7 ] .

A basic distinction can usually be made between sludges which, upon collection, are still relativelyfresh or contain a fair amount of recently deposited excreta (e.g. sludges from frequently emptied,unsewered public toilets) and sludges which have been retained in on-plot pits or vaults for monthsor years and which have undergone a biochemical degradation to a variable degree (e.g. sludgefrom septic tanks septage). Moreover, varying amounts of water or wastewater are collectedalongside with the solids, which have accumulated in vaults or pits. Based on numerous FS

monitoring studies, the authors found that FS can often be associated with one of two broadcategories, viz. high and low-strength sludge.

Total solids (TS) concentration

Faecal sludge

High-strength FS(e.g. from unsewered, low or zero-flush public toilets)

Low-strength FS(septage)

WWTP sludge

Wastewater in thetropics

mg TS/L 10,000 20,000 30,000 40,000 50,0000

Waste activated sludge Primary and anaerobicallydigested sludge

1,000-1,500

Fig. 4 Solids contents of faecal sludges, WWTP sludges and tropical wastewater


5/17

Table 3 Faecal sludges from on-site sanitation systems in tropical countries: characteristics, classificationand comparison with tropical sewage (after Strauss et al. 1997 [8] and Mara 1978 [7])

ItemType A

(high-strength)Type B

(low-strength)Sewage for

comparison purposes

ExamplePublic toilet or bucket latrine

sludgeSeptage Tropical sewage

CharacterisationHighly concentrated,

mostly fresh FS; stored for days or weeks only

FS of low concentration; usuallystored for several years; more

stabilised than Type A

COD mg/l 20, - 50,000 < 15,000 500 - 2,500

COD/BOD 5 : 1 .... 10 : 1 2.5 : 1

NH 4-N mg/l 2, - 5,000 < 1,000 30 - 70

TS mg/l 3.5 % < 3 % < 1 %

SS mg/l 30,000 7,000 200 - 700

Helm. eggs no./l 20, - 60,000 4,000 300 - 2,000

Table 3 shows typical FS characteristics. It is based on results of FS studies in Argentina,Accra/Ghana, Manila/Philippines and Bangkok/Thailand. The characteristics of typical municipalwastewater as may be encountered in tropical countries are also included for comparison purposes.

The hygienic quality of faecal sludge

In many areas of Africa, Asia and Latin America, helminth, notably nematode infections ( Ascaris,Trichuris, Ancylostoma, Strongyloides, etc. ) are highly prevalent. Ascaris eggs are particularly

persistent in the environment. The bulk of helminth eggs contained in faecal or in wastewater treatment plant sludges end up in the biosolids generated during treatment. Hence, in many places,nematode eggs are the indicators-of-choice to determine hygienic quality and safety where

biosolids are to be used as a soil conditioner and fertilizer. The concentration of helminth eggs inthe biosolids is largely dependent on the prevalence and intensity of infection in the populationfrom which FS or wastewater is collected. Where biosolids use in agriculture is a practice or beingaimed at, treatment must aim at reducing helminth egg counts and viability, or solids storage must

be long enough to achieve the desired reduction. Fresh FS may contain from 4, -60,000 helmintheggs /L, tropical wastewater from a few hundred to 2,000 /L., depending on the endemicity andintensity of infection.

FAECAL SLUDGE TREATMENT CRITERIA AND OPTIONS

Expedient treatment standards

In developing or newly industrialising countries, unduly strict standards at times uncriticallycopied from industrialised countries are often being stipulated. However, these are usually neither enforceable nor enforced. Institutional, economic and political conditions are often not consideredwhen formulating them. Hence, human wastes remain untreated, the performance of treatmentworks, where existing, are not controlled at all, or treatment options bringing about partialimprovements might be rejected at the planning stage already. In industrialised countries, treatmentstandards were made increasingly stringent in a process lasting for several decades, giving dueconsideration to the countries evolving technical, economic and institutional capacities (Johnstoneand Horan, 1996 [9 ]). A suitable strategy for economically developing countries would consist inselecting a phased approach, too, under the paradigm that something (e.g. 75 % instead of 95-99% helminth egg or COD removal) is better than nothing (the lack of any treatment at all or the

often totally inadequate operation of existing treatment systems) (Von Sperling and Fattal 2001[10]). As an example, Xanthoulis and Strauss (1991 [11 ]) proposed a guideline value for biosolids


6/17

as produced in FS or WW treatment schemes of 3-8 viable nematode eggs/ g TS (derived from theWHO guideline of 1 nematode egg/L of treated wastewater used for unrestricted vegetableirrigation (WHO 1989 [12 ]), and on an average manuring rate of 2-3 tons TS/hayear). In contrastto this, the standard to comply with in Western Europe and in the U.S., are 0 eggs/g TS and1 egg /4g TS, respectively ! Treatment or measures to comply with such standards would neither be

affordable nor epidemiologically necessary in the majority of developing countries.In industrialised countries, there is a change in paradigm from numerical limits stipulated for thequality of treatment products (such as for biosolids and liquid effluent) to the stipulation of specifictreatment options or specific soft measures (e.g. the restriction of selected crops to be irrigatedwith wastewater). Treatment works and soft measures are barriers against microorganisms or contaminants. They are designated as critical control points and the strategy is termed HACCP(hazard analysis and critical control point) (Matthews 2000 [13 ] and Deere et al. 2001 [14 ]).

Treatment Options Overview

Fig. 5 provides an overview of options for faecal sludge treatment, which can be implemented byusing modest to low-cost technology, and which therefore carries a high potential of sustainability.

Some of the options were or are currently being investigated upon by EAWAG/SANDEC and its partners in Argentina, Ghana, Thailand and The Philippines and will be presented in the following paragraphs.

Proper FS treatment, either in combination with wastewater or separately, has been practiced in afew countries only to date (e.g. China, Thailand, Indonesia, Argentina, Ghana, Benin, Botswana,South Africa). The authors are aware of several, very recent initiatives for improved FSmanagement and treatment, notably in West Africa. Treatment options used or proposed comprise

batch-operated settling-thickening units; unplanted and planted sludge drying beds; non-aeratedstabilization ponds; combined composting with municipal organic refuse; co-treatment of FS inwastewater treatment plants.

Faecal and WWTP sludges may, in principle, be treated by the same type of modest-cost treatmentoptions. The following unit operations or processes are deemed suitable for treating sludges ineconomically developing countries:

Solids-liquid separation in settling-thickening tanks or in primary sedimentation ponds Sludge drying beds (unplanted; planted) Pond systems for separate treatment of FS and combined treatment with wastewater Combined composting with organic solid waste (co-composting)

Anaerobic digestion with biogas utilization

Fig. 5 Overview of potential, modest-cost options for faecal sludge treatment


7/17

For complete treatment, processes and operations would have to combined or complemented by polishing processes (pond systems or constructed wetlands, e.g.) for liquid effluents from FStreatment. For these processes, reference is made to the ample literature available on the respectiveoptions. Biosolids resulting from the core processes may have to be subjected to further storage or drying to achieve more extensive pathogen inactivation and allowing for their unrestricted use inagriculture.

The fact that faecal sludges exhibit widely varying characteristics calls for a careful selection of appropriate treatment options, notably in primary treatment. Primary treatment would have to allowfor biochemical stabilization if the FS is still rather fresh and has undergone but partial degradationduring on-plot storage prior to collection.

The separating of the solids and liquids, which make up FS, is the process-of-choice in FStreatment unless it is decided to co-treat FS in an existing or planned WWTP and if the FS loadsare small compared to the flow of wastewater. Solids-liquid separation may be achieved throughsedimentation and thickening in ponds or tanks or filtration and drying in sludge drying beds.Resulting from this are a solids and a liquid fraction. The solids fraction, which may be designatedas biosolids, is of variable consistency. It may require post treatment, mainly to meet hygienerequirements for reuse in agriculture as a soil-conditioner and fertilizer. Additionaldewatering/drying might be required for landfilling. Polishing treatment might be necessary for theliquid fraction, too, to satisfy criteria for discharge into surface waters and/or to avoid long-termimpacts on groundwater quality, where effluents will be allowed to infiltrate.

CONSTRUCTED WETLANDS FOR SEPTAGE DEWATERING ANDSTABILIZATION PILOT EXPERIMENTS AT AIT, BANGKOK

The process and preliminaries

In pursuing R+D for appropriate options of faecal sludge treatment, EAWAG/SANDEC andAIT/UEEM initiated collaboration in 1996 to investigate upon the feasibility of using vertical-flowconstructed wetlands (CW; planted sand drying beds) as a hypothetically feasible option.

CW are man-made systems aiming at simulating the treatment processes in natural wetlands bycultivating emergent plants such as reeds ( Phragmites ), bulrushes ( Scirpus ), and cattails ( Typha ) onsand, gravel, or soil media. Based on investigations of pilot and field-scale systems, CWs have

been proven to be a feasible treatment alternative for wastewater and sludges from wastewater treatment plants characterized by low investment, operation and maintenance costs (Kadlec andKnight 1995 [15] and Cooper et al ., 1996 [16]). Therefore, utilization of CWs in waste treatmentand recycling is currently of interest, including their ancillary benefits such as supporting primary

production and enhancement of wildlife habitats. For several years, a number of CW systems have been employed to treat various kinds of liquid wastes including, more recently, sludge fromactivated sludge treatment plants (Heinss and Koottatep 1998 [17]).

For sludge dewatering, a vertical-flow mode of operation with a percolate-drainage system beneathCW beds is required. An advantage of CWs over conventional, unplanted sludge drying beds is themuch lower frequency of dewatered sludge removal from the bed, allowing for several years of sludge accumulation prior to bed emptying. Furthermore, the accumulating solids create a filteringmatrix allowing for the development of attached-growth biomass. As a consequence, mineralisation

processes are likely to lead to higher removal efficiencies than in unplanted sludge drying beds.Linard and Payrastre (1996 [18]) of CEMAGREF, the French national centre for agriculturalengineering, water and forestry (Lyon branch), have conducted pioneering laboratory-scaleexperiments using vertical-flow CW and reeds for treating septage. Their experience formed the

basis for initiating the pilot investigations at AIT.


8/17

Experimental set-up

Treatment scheme . The AIT pilot-scale septage treatment scheme comprises the following unitoperations and processes: screening balancing and mixing tank 3 constructed wetland (CW)

beds attached-growth WSP unit (1997-2000) / vertical-flow CW bed (> 2000) for percolate polishing. This report focuses on the investigations with the septage-loaded CW units treatingseptage for dewatering and biochemical stabilisation. Results of percolate treatment are reported insummarised form.

Dimensions of CW beds . The AIT pilot-scale CWs are square-shaped to allow uniform distributionof raw septage. The pilot works comprise three vertical-flow units, each with an area of 5 x 5 m atthe surface of the filter bed (6.2 x 6.2 m at the rim) and lined with ferro-cement as shown in Fig. 6.

Media Arrangement and Vegetation . The depth of the filter media (substrata) in these experimentswas designed to be 65 cm, because the length of the cattail roots and rhizomes is only 30-40 cm, asagainst 50-60 cm for reeds. A 10-cm layer of fine sand, 15-cm layer of small gravel, and 40-cmlayer of large gravel from top to bottom were used as substrata in each CW unit. A free board of 1m was allowed for accumulation of the dewatered sludge. On top of the sand layer, narrow-leavecattails ( Typha augustifolia ), collected from a nearby natural wetland, were planted in each CWunit at the initial density of 8 shoots/m 2. The current density is 40-50 shoots/m 2. Cattails wereselected because they are an indigenous species and evidently growing better than reeds in mostwetland areas of Thailand.

Given the fact that Typha is being used as a macrophyte in AITs CW plant, care would have to betaken in transferring the results of these investigations to CW schemes, in which reeds

( Phragmites ) would be used as a wetland plant. Bed operations and possibly designrecommendations may differ, as the two macrophytes may exhibit different growth patterns andwetting requirements. Also, their sensitivity towards anaerobicity and low redox potentials in theroot/rhizome zone and towards ammonia levels in the loaded sludge may be different.

Underdrain and Ventilation System . The bed support and drainage system consists of hollowconcrete blocks, each with a dimension of 20 x 40 x 16 cm (width x length x hollow space), and

perforated PVC pipes with a diameter of 20 cm at the bottom. Mounted on the drainage system areventilation pipes of the same diameter and extending approximately 1 m over top edge of the units.

Natural draught ventilation is required to avoid anaerobic conditions in the root zone and, hence, plant damage. The percolate of each CW unit was collected in a 3-m 3 concrete tank for samplingand analysis.

Fig. 6Pilot-scale constructed wetlandunits installed at AIT, Bangkok (schematic)

Filter substrata : 10-cmlayer of fine sand, 15-cmlayer of small gravel, and40-cm layer of largegravel from top to bottom


9/17

Septage characteristicsa

Parameter Range Mean pH 6.7 8.0 7.5TS (mg/L) 2,200 67,200 19,000TVS (mg/L) 900 52,500 13,500SS (mg/L) 1,000 44,000 15,000BOD (mg/L) 600 5,500 2,800TCOD (mg/L) 1,200 76,000 17,000TKN (mg/L) 300 5,000 1,000NH 4 (mg/L) 120 1,200 350

a. Based on 150 FS samples in this study during August 1997 Nov. 2000

Raw septage Characteristics

Table 4. Characteristics of septage collected from the city of Bangkok andtreated at the AIT pilot station

Table 4 lists the averagecharacteristics of Bang-

kok septage, which ishauled weekly to AIT bythe Bangkok Metropo-litan Authority and app-lied to the three CW pilot

beds. The wide range of values for each parameter is typical of the varia-

bility of faecal sludgecharacteristics world-over.

Acclimatisation

In order for cattails to acclimatize with FS, AIT sewage was fed into the CW units at theapplication rate of 5 cm/d for about 2 3 weeks until the cattails had grown to 1.5-1.8 m in height.After that, FS was gradually fed into the CW units in combination with AIT sewage for about 2months to minimize effects of ammonia toxicity towards the cattail plants. The young roots andstems of cattails began to grow, but some cattail plants could not adapt to the FS, causing their leaves to turn yellow and die. The CW units were then loaded with FS at the designed SLR andapplication frequency.

Operating Conditions

The authors hypothesized at the onset of the investigations that, in tropical regions, CWs can be

loaded with FS at higher solids loading rates (SLR) than those experimented with by Cooper et al. (1996 [16]) in temperate climate. There, SLR ranged from 30 to 80 kg total solids (TS)/m 2.yr withonce-a-week application.

For the study, septage was hauled from northern Bangkok by a contractual agreement. The raw FSwas passed through a bar screen to remove coarse objects, and then homogenised in two 4-m 3 mixing tanks before feeding to the CW units. The CW beds were operated as shown in Table 6.

In Runs I III, the CW percolates were allowed to flow freely into the percolate tanks soon after FS feeding. As from Run IV, percolate ponding was introduced by closing the outlet valves for aselected number of days. Instead of maintaining a constant SLR, which would result in variationsof the volumetric loading rate, FS was loaded at a constant volumetric rate of 8 m 3/week, allowingfor easier control of loading operations. This corresponded with a SLR range of 140 360 kgTS/m 2.yr.

Samples collected from the CW units at different operating conditions were analyzed for totalsolids (TS), total volatile solids (TVS), suspended solids (SS), biochemical oxygen demand (BOD),TCOD (total, unfiltered COD), TKN, ammonium+ammonia (NH 4+NH 3), NO 3, and helminth eggs.


10/17

Table 5. Operating conditions of the three pilot-scale CW units

Run SLR (kgTS/m 2.yr) Percolate Frequency of Period of CW-1 CW-2 CW-3 ponding * FS application operation

I 250 125 80 No Once + twice-a-week

Apr. 97 Dec. 97

II 500 250 160 No Twice-a-week Dec. 97 Jan. 98III 500 250 160 No Once-a week Feb. 98 Mar. 98

IV 250a

250 b

250c Yes Once-a week Apr. 98 Feb. 99

V+

140 360 140 360 140 360

Yes Once-a week Mar. 99 Sep. 99VI 175 450 - - Yes Once-a week Oct. 99 Nov. 00

VII 121 - 331 - - Yes Once-a-week June 00 Dec. 01

* Percolate was retained 10 15 cm below dewatered FS layers in CW units using a gate valvefitted to the drain pipe

+ To ease operations, FS was loaded at a constant volumetric rate of 8 and 10 m 3/week for Run V,VI and VII, respectively, resulting in varying SLR

a Ponding period = 6 days, b ponding period = 2 days, c no ponding

Except for helminth eggs, analyses were performed based on Standard Methods for theExamination of Water and Wastewater (APHA, AWWA, WPCF, 1992 [19]). For helminth eggs, amethod developed by USEPA and modified by Schwartzbrod (1998 [20]) was used.

Results and Discussion

Variation of SLR and application frequency . The results obtained from Runs I III (Table 5) showthat at SLR ranging from 80 500 kg TS/m 2.yr, removal efficiencies for TS, TCOD, TKN, and

NH 4, ranging from 76 - 88 %; 88 - 98 %; 82 - 99 %, and 52 - 98%, respectively, were attained inthe liquid percolating through the CW beds. As shown in Table 6, the application frequency of FS

did not have significant effect on removal efficiencies at the same ranges of SLR. The averageheight of dewatered solids was 10 - 15 cm after one year of operation, exhibiting TS contents from30-60 %.

Statistical analysis revealed that, similar to the application frequency parameter, the treatmentefficiencies of the CW units did not depend on variations in SLR in the 80 500 kg TS/m 2.yr range. This may have been due to the considerable fluctuations in raw septage characteristicsresulting in a wide range of effluent concentrations. Also, in the first two years of operation, theaccumulated sludge layers were not high enough to lead to enhanced filtration effects. In addition,mineralisation of carbonaceous compounds within the filtering matrix was unlikely to play asignificant role due to the relatively low biodegradability of Bangkok septage (COD:BOD ratio=6).Removal efficiencies decreased, though, upon doubling the SLR in each of the units.


11/17

Table 6 Average TS, TCOD, TKN, NH 4, and NO 3 contents in CW percolate and removal efficiencies as a

function of application frequency and solids loading rate (SLR) during runs I-III +

Sample Frequency SLR CWBed

Parameter*

, mg/L

No./week kg TS/m 2.yr No. TS TCOD TKN NH 4 NO 3

Raw FS ** 16,300 16,000 830 340 8

Percolate 1 80 3 3,640 (78)

210 (98)

45 (98) 32 (92) 2101 125 2 2,840 (80) 230 (98) 60 (96) 56 (80) 1801 160 3 4,320 (77) 1,250 (91) 150 (83) 110 (68) 2701 250 2 6,030 (77) 850 (96) 120 (87) 110 (70) 3201 250 1 2,640 (80) 300 (97) 62 (93) 46 (85) 1801 500 1 4,960 (82) 1,880 (94) 240 (82) 170 (52) 250

2 80 3 2,980 (83)

110 (98)

10 (99) 5 (98) 2002 80 3 2,670 (86) 910 (94) 95 (99) 49 (79) 2602 125 2 3,610 (80) 570 (96) 62 (98) 44 (93) 1902 125 2 2,700 (84) 460 (97) 100 (96) 36 (91) 1902 160 3 3,800 (80) 1,720 (88) 250 (79) 190 (52) 1902 250 1 3,340 (81) 810 (97) 110 (95) 100 (90) 2602 250 1 2,720 (88) 780 (95) 110 (94) 87 (88) 2002 250 2 3,600 (76) 800 (94) 140 (90) 100 (79) 2202 500 1 2,900 (81) 1,020 (93) 182 (87) 190 (69) 180

+ Removal efficiencies as shown in parentheses depended on the characteristics of raw FS used ineach experimental run.

* Average data were based on composite samples taken from each experimental run.** Raw FS data were averages of 72 samples of Run I to IV, during April 1997 March 1998

TKN and (NH 4+NH 3) removals in the CW units are most likely caused by accumulation of organic- N in the dewatered sludge layers, mainly. Additional sinks for TKN and (NH 4+NH 3) are ammonia

volatilization, plant uptake and nitrification reactions. It can be seen that NO 3 concentrationsincreased significantly from 8 mg/L in raw septage to 180 - 320 mg/L in the percolate due toaerobic conditions prevailing in the filter matrix and supporting nitrification. Dissolved oxygenconcentrations of 2 4 mg/L in the percolate are evidence of an oxidative environment. The resultsreveal the beneficial effects of the vertical-flow mode and of drainage system ventilation, bothenhancing nitrification.

It could be observed during the initial septage loading runs that the cattail plants in the CW unitsshowed signs of wilting with once-a-week loading and were shocked at SLR higher than 250 kgTS/m 2.yr. At twice-a-week application, the cattail plants grew slightly better because only half of the weekly septage load was loaded at each FS application. The cattail plants were entirely adaptedto FS and exhibited a healthy growth pattern even at once-a-week application after one year of operation. The results of the one-year parameter testing suggested that the optimum SLR is around250 kg TS/m 2.yr and the frequency of application once a week for the type of septage deliveredfrom northern Bangkok.


12/17

Variation of ponding periods

Inspite of the above results showing relatively high removals of TS, TCOD, TKN, and NH 4, theCW percolate remained relatively high in nitrate (NO 3). Also, the cattails showed signs of water deficiencies. In the experimental runs IV-VI, the percolate was therefore impounded in units no. 1and 2 by gate valves for a specified number of days prior to discharge. The respective results areshown in Table 7.Table 7 Average TS, TCOD, TKN, NH 4, and NO 3 contents in CW percolate and removal

efficiencies in runs IV-VI+

Unit SLR Ponding Parameter*

, mg/LSample

No. kg TS/m 2.yr days TS TCOD TKN NH 4 NO 3

Raw FS **

18,500 16,000 1,000 440 6

Percolate 1 250 6 2,000 (86) 270 (98) 100 (89) 80 (81) 20

Run IV 2 250 2 2,400 (84) 400 (97) 150 (85) 100 (77) 533 250 0 2,700 (81) 620 (96) 200 (80) 140 (69) 120

1 140 360 6 2,600 (82) 320 (98) 106 (89) 80 (81) 22Run V 2 140 360 2 2,700 (78) 450 (97) 150 (84) 100 (70) 55

3 140 360 0 3,300 (76) 780 (94) 220 (79) 140 (60) 130Run VI 1 175 450 6 3,550 (72) 250 (98) 73 (92) 50 (89) 35

+ Removal efficiencies as shown in parentheses depended on the characteristics of raw FS used in each expererimentalrun

++ Solid loading at the constant volumetric loading rate of 8 m 3/week * Average data were based on 15 composite samples taken from each experimental run.** Raw FS data were averages of 60 samples during April 1998 November 2000

The CWs were fed with septage at a solids loading rate of 250 kg TS/m 2.yr, while the CW percolates were withheld at ponding periods of 6, 2 and 0 days in CW units No. 1, 2, and 3,respectively. Table 7 shows that percolate ponding did not have significant effects on the TS andTCOD removal efficiencies, probably because the filtering capacity of the CWs remainedunaffected by percolate ponding and degradation of organic constituents was modest. Removalefficiencies for TKN and NH 4 of 89% and 81% were attained, respectively with ponding period of 6 days.

During runs No. IV-VI (April 1998-Nov. 2000), percolate TCOD concentrations of less than 800mg/L (median = 400 mg/L) were reliably achieved. This is in contrast to the preceding runs No. I-III (April 1997-March 1998) when effluent concentrations of up to 1,900 mg COD/L wereobserved. The increased process stability and reliability can largely be attributed to the increasedheight of accumulated biosolids and, hence, enhanced filtering effects.

The average NO 3 concentration in the percolate was only 20 mg/L, as compared with 53 and 120mg/L for the beds subjected to ponding of 2 and 0 days, respectively. This phenomenon was

probably due to denitrification reactions occurring in the CW beds with a longer percolate ponding. Nitrogen uptake by cattail plants could be another N removal mechanism in the CW units, asreported by Koottatep and Polprasert (1997 [21]).

After the operating conditions were amended to a constant FS loading rate of 8 m 3/week (equivalent to an SLR ranging from 140 360 kg TS/m 2.yr), it became apparent that no significantdifference in removal efficiencies of the CW beds occurred as compared to those fed at constantSLR. Hence, for reasons of operational easiness, the volumetric loading rate can be taken as theloading criteria, once the solids content of the particular septage has been ascertained withreasonable safety.


13/17

NH2O

Fig. 7 Water and nitrogen balance in CW units fed with septage

Bed performance in the 4 th and 5 th year of continuous septage loading

Table 8 shows contaminant removals in CW bed No.1 during run VII (June 00 - Dec. 01). Like inruns No. V-VI, septage was loaded at a constant volumetric rate of 8 m 3/week, which wasequivalent to a SLR of 235 kg TS/ m 2.yr. The loading frequency was once-per-week and percolate

ponding was 6 days/week.

Table 8 Avg. TS, COD, TKN, NH 4, and NO 3 contents in CW percolate and removal efficiencies in run VII+

CWbed

SLR Ponding Parameter*

, mg/LSample

No. kg TS/m 2.yr days TS TCOD TKN NH 4 NO 3

Raw FS **

12,300 16,000 1,000 500 4 Percolate

Run VII 1 235(121331)

6 3,200 (74) 270 (98) 61 (94) 46 (91) 101

+ Removal efficiencies as shown in parentheses depended on the characteristics of raw FS usedin each experimental run.

++ Septage loading at the constant volumetric loading rate of 8 m 3/week * Average data were based on 24 composite samples taken from each experimental run.** Raw FS data were averages of 60 samples during June 2000 December 2001

The data of Table 8 show that bed performance with respect to contaminant removals in the percolating liquid was essentially the same as in the preceding runs. This reveals the high degree of process stability. The increasing depth of accumulated solids allows for a sustained filtering, whilethe root and rhizome system of the macrophytes secures the continued permeability of the entire

bed.

Accumulation of biosolids and mass balances

Accumulation of biosolids. 70 cm of dewatered and stabilised solids had accumulated in the CW beds by the end of the 3.5 years experimental period, being equivalent to an average annualaccumulation of 20 cm. Given the freeboard of 1.0 m chosen for the AIT pilot plant, solidsaccumulation is likely to last for at least a total of five years before emptying becomes necessary.The 70 cm of solids accumulated so far are equivalent to a septage column of 60 m loaded onto the

beds. In spite of this extended loading without removal of accumulated solids, there was no bedclogging and percolate flow remained entirely unimpeded. This phenomenon was presumably dueto the continuous growth and distribution of the cattail roots and rhizomes in the dewatered sludgelayers and substrata, which helped to create and maintain porosity in the CW beds.

Mass balances .

Fig. 7 is a graphicalrepresentation of themass balances acrossthe CW beds for nitrogen and water. Nuptake by the cattailsaccounted for 5%,only, of the total Nloaded throughseptage. N lossesdue to ammoniavolatilization, micro-

bial uptake and Naccumulation in the


14/17

CW support media, amount to some 10 30%, whereas nitrogen leaving the system in the percolate accounts for 10-25 %. One third of the water brought onto the beds with septage, isevapotranspirated and two thirds are drained. Some 2 % only are retained in the accumulatingsolids.

At the end of Run V (300 days of operation), the TS mass balances in each CW bed were analyzedas shown in Table 9. The accumulated TS inputs to CW units 1, 2, and 3 were 187, 115 and 112 kgTS/m 2, respectively. The average TS mass in dewatered FS amounted to 38 52% of the TS inputs,while about 11 12% of the TS inputs were in the percolate portion. The unaccounted for TS of 36

50% can be attributed to the loss of organic matter through mineralisation to yield water andCO2, and TS accumulation in the CWs substrata.

Table 9 TS mass balance in CW units after 1 year of CW bed operation

Unit no. 1 Unit no. 2 Unit no. 3Balancekg TS/m 2 % kg TS/m 2 % kg TS/m 2 %

Accumulated TS loading 187 - 115 - 112 -Dewatered FS 93 50 60 52 43 38Percolate 20 11 14 12 13 12Unaccounted 74 39 41 36 56 50

Hygienic quality and fertilising value of accumulated solids

Helminth eggs, which constitute the hygiene criteria-of-choice for biosolids earmarked for agricultural use, become concentrated in on-site sanitation systems and, hence, during further treatment in the filter body of constructed wetland beds. Nematode egg concentrations in the rawseptage delivered to the AIT CW pilot plant ranged from 30-40 per g TS (495-660 /L of septage).This level is relatively low due probably to the limited endemicity and worm loads in the

population from which the particular septage is collected. Nematode egg levels found in the

dewatered and stabilised solids at the end of the reporting period, ranged from 100-270 per g TS,irrespective of the layer in the 70+ cm of accumulated solids. However, in none of the samples wasthe number of viable eggs higher than six (6) per g TS. The sludge is thus safe for agricultural useif measured against the expedient standard of 3-8 eggs per g TS as stipulated by Xanthoulis andStrauss (1991 [11]).

The nitrogen and phosphorus (as P) contents of the solids layer, which has accumulated during 4.5years of septage loading, amounted to 3 % and 1.2 %, respectively. This compares very favourablywith e.g. solid waste composts, due likely, to the fact that human excreta form the origin of thesolids accumulating in the CW beds.

Percolate polishing

The percolate of the septage-loaded CW beds was treated in a two-stage attached-growth wastestabilisation pond (AGWSP, equipped with plastic moulds) scheme (without and with precedinghorizontal gravel/roughing filtration) from April 97 until early 2001. CW bed No. 3 was plantedwith Canna, an ornamental plant indigenous of Thailand, and used for percolate treatment as fromJuly 2001 to test as an alternative polishing process. The AGWSP scheme comprises two ponds inseries, allowing for a total HRT of 10 days. The Canna-planted CW bed has a surface area of 25m2 at the media surface and is loaded at a hydraulic loading rate of 0.025 m/d. Fig. 8 shows theinfluent and effluent concentrations, respectively, of TCOD and NH4-N for the two AGWSPalternatives and for the Canna-planted CW bed. Both types of processes tend to produce effluentsof similar quality (100-180 mg/L) and of a quality, which may satisfy expedient dischargestandards (agricultural use of CW percolate and of FS liquids in general are unfit for agriculturaluse due to their excessive salinity).


15/17

Costing appraisalLand-free capital cost as extrapolated from the AIT pilot-scheme amount to approx. US $ 100/m 2 of net CW bed area, including screening, balancing tanks and piping, yet excluding polishingtreatment. The annual O+M costs are estimated to amount to 250 US $/1,000 inhabitants served.This excludes costs for dewatered sludge removal, which will become necessary after 5-6 years of operation, only. At the optimum SLR of 250 kg TS/m 2.yr, each CW unit is able to treat septagecollected from about 1,000 persons (1 L/cap.d and TS = 1820 g/L (U.S. EPA, 1995 [22]). Thus,capital and annual O+M cost amount to 2.5 and 0.25 $/cap, respectively. The gross landrequirement (= net land requirement x 1.3) for the CW plant at AIT is equivalent to approx. 32m2/1,000 persons. Additional 7 m 2 /1,000 persons would be needed for percolate polishing. For comparison purposes, the area required for pond treatment of faecal sludges, inclusive of polishingand dewatering/drying of separated solids has been estimated by Heinss et al. (1998 [5]) to amountto 40 m 2/1,000 persons if solids are treated by co-composting, and in 70 m 2/1,000 persons withsolids treatment on sludge drying beds.

FS dewatering in vertical-flow CW, which requires removal of dewatered biosolids every 2-5 yearsonly, results in significant savings in O+M cost as compared with unplanted sludge drying beds,which require emptying at weeks intervals.

CONCLUSIONS

The dramatic situation of uncontrolled FS management endangering public health and theenvironment calls for effective improvements in FS management strategies and for the devising of

treatment schemes making use of appropriate technology. Introducing sound cost recovery andmoney flux procedures, effective private-public partnerships, entrepreneurial competition andeffective sanctioning are elements required for sustainable FS management. Treatment options areat hand, which, when properly designed and operated, can satisfy expedient standards adapted todeveloping countries and cater for substantially reducing the shedding of viable intestinal

pathogens into the urban environment and, hence, public health risks. Also, water pollution can beconsiderably reduced.

The 4.5-year experimental results of treating septage in cattail ( Typha )-planted constructedwetlands suggest that suitable operating conditions for a vertical-flow CW system treating septageof the type collected in the selected suburban districts of Bangkok be as follows:

- SLR (solids loading rate) = 250 kg TS/m 2.yr or constant volume loading of 8 m 3/week - Once-a-week application- Percolate ponding period of 6 days

050

100150200250300350400450500

[ m

g / l ]

AGWSP HRF+AGWSP Canna Bed

COD

CW PercolateEffluent

0

20

40

60

80

100

[ m g / l ]

AGWSP HRF+AGWSP Canna Bed

NH4

CW PercolateEffluent

Fig. 8 Polishing treatment of percolate from septage-loaded CW in attached-growth waste stab. pondsand in constructed wetlands: influent and effluent concentrations of TCOD and NH4-N (AGWSP-attached-growth waste stabilisation ponds; HRF-Horizontal roughing filter; Canna Bed-Canna-

planted constr. wetland bed; results for the AGWSP systems based on 22 samples, for the Canna bed on 13 samples)


16/17

These operating conditions have resulted in optimum performance of the CWs with respect to FSdewatering and contaminant removal from the percolating liquid, healthy and reliable plant growth,and ease of operations in which removal of the dewatered FS was not required for a period of 5years. Percolate ponding, which was necessary to avoid Typha (cattail) wilting, significantlyenhanced nitrification/denitrification leading to a lowering of nitrogen loads in the percolate.Although more long-term data and full-scale operating experience are required, the resultsgenerated to date indicate that vertical-flow constructed wetlands constitute a promising, modest-cost technology for treating septage. Care must be exerted in transferring the experience presentedin this paper to localities where septage characteristics differ from those in Bangkok and wherereeds ( Phragmites ) or bulrushes ( Scirpus ) instead of cattails ( Typha ) are used as macrophytes.

ACKNOWLEDGEMENT

The authors would like to thank Mr. Surasak Limsuwan, Mr. Anawat Pinsakul, Mr. Roongrod Jittawattanarat,and Mr. Suprachoke Uakritdathikarn, for their valuable contributions in the operation and maintenance of theexperimental CW units as well as the parameter analyses. The generous support from the Swiss Agency for Development and Cooperation (SDC) through a grant to EAWAG is gratefully acknowledged.

REFERENCES

[1] Strauss, M., Heinss, U., Montangero, A. (2000). On-Site Sanitation: When the Pits are Full Planning for Resource Protection in Faecal Sludge Management. In: Proceedings, Int. Conference, Bad Elster, 20-24

Nov. 1998. Schriftenreihe des Vereins fuer Wasser-, Boden- und Lufthygiene, 105 : Water, Sanitation & Health Resolving Conflicts between Drinking Water Demands and Pressures from Societys Wastes(I.Chorus, U. Ringelband, G. Schlag, and O. Schmoll, eds.) . IWA Publishing House and WHO Water Series. ISBN No. 3-932816-34-X.

[2] Klingel, F. (2001). Nam Dinh Urban Development Project. Septage Management Study . Nam Dinh,Vietnam, Nov.. EAWAG/SANDEC and Colenco (Vietnam).

[3] Klingel, F., Montangero, A.., Strauss, M. (2002). Guide on Faecal Sludge Management Planning .EAWAG/SANDEC, drafted.

[4] Jeuland, M. (2002). Economic Management of Faecal Sludges in Bamako, Mali . Unpubl. report. March.[5] Heinss, U., Larmie, S.A., Strauss, M. (1998). Solids Separation and Pond Systems for the Treatment of

Septage and Public Toilet Sludges in Tropical Climate - Lessons Learnt and Recommendations for Preliminary Design. EAWAG/SANDEC Report No. 05/98.

[6] Montangero, A. and Strauss, M. (2002). Faecal Sludge Treatment. Lecture Notes, IHE Delft, Feb. 14.[7] Mara, D. D. (1978). Sewage Treatment in Hot Climates. Chichester: John Wiley & Sons.[8] Strauss, M., Larmie S. A., Heinss, U. (1997). Treatment of Sludges from On-Site Sanitation: Low-Cost

Options. Water Science and Technology, 35, 6.[9] Johnstone, D.W.M., Horan, N.J. (1996). Institutional Developments, Standards and River Quality: An UK

History and Some Lessons for Industrialising Countries. Water Sci. & Techn ., 33 , No. 3, pp. 211-222.[10] Von Sperling, M. and Fattal, B. (2001). Implementation of Guidelines Practical Aspects for

Developing Countries. In: Water Quality Guidelines, Standards and Health. Assessment of risk and management of water-related infectious disease . Eds.: Fewtrell, L. and Bartram, J.. IWA Publ. and WHO.

[11] Xanthoulis, D. and Strauss, M. (1991). Reuse of Wastewater in Agriculture at Ouarzazate, Morocco(Project UNDP/FAO/WHO MOR 86/018). Unpublished mission/consultancy reports.

[12] WHO (1989). Health Guidelines for the Use of Wastewater in Agriculture and Aquaculture. Report of aScientific Group. World Health Organisation Technical Report Series 778.

[13] Matthews, P. (2000). Regulations for Biosolids Utilization. In: Proceedings, Forum on BiosolidsManagement and Utilization , Hongkong, Sept. 8, and Nanjing, P.R. China, Sept. 10-13. Hongkong BaptistUniversity, Inst. for Natural Resource & Waste Management.

[14] Deere, D., Stevens, M., Davison, A., Helm, G., Dufour A. (2001). Management Strategies. In: Water Quality Guidelines, Standards and Health. Assessment of risk and management of water-related infectious disease . Eds.: Fewtrell, L. and Bartram, J.. IWA Publishing and World Health Organization..

[15] Kadlec, R.H. and Knight, R.L. (1995). Treatment Wetlands , Lewis Publisher, New York.[16] Cooper, P.F., Job, G.D., Green, M.B., and Shutes, R.B.E. (1996). Reed Beds and Constructed Wetlands

for Wastewater Treatment , WRc Swindon, Wiltshire.[17] Heinss, U. and Koottatep, T. (1998). Use of Reed Beds for Faecal Sludge Dewatering . EAWAG/AIT.


17/17

[18] Linard, A. and Payrastre, F. (1996). Treatment of sludge from septic tanks in a reed-bed filters pilot plants, Pre Prints of the 5 th Int. Conference on Wetland Systems for Water Pollution Control , Vienna, 8 p.

[19] APHA, AWWA, WPCF (1992). Standard Methods for the Examination of Water and Wastewater , 17 th Edition, Washington, D. C..

[20] Schwartzbrod, J. (1998). Quantification and Viability Determination for Helminth Eggs in Sludge(Modified USEPA Method). Fac. of Pharmacy, Henri Poincar University, Box 403, Nancy Cedex France.

[21] Koottatep, T. and Polprasert, C., (1997). Role of plant uptake on nitrogen removal in constructedwetlands located in tropics, Wat. Sci. & Tech. , Vol. 36, No. 12, pp. 1-8.[22] United States Environmental Protection Agency (U.S. EPA) (1995). Process Design Manual Surface

Disposal of Sewage Sludge and Domestic Septage, Washington DC.

Constructed Wetlands for Septage Treatment

Documents

Transcript of Constructed Wetlands for Septage Treatment