A constructed wetland to treat sewage from family Subandi house has been built in Bandung, Indonesia, in February 1999 to serve as a pilot project.  Water samples from both influent and effluent were taken twice a month and analysed for COD, BOD5, NO3-N, NO2-N, NH4-N, PO4-P, pH, fecal coliforms bacteria, electrical conductivity, and settleable solids. The objective of this study was to install one constructed wetland with a vertical flow system to treat sewage from private households by using an aquatic macrophyte (Phragmites karka).  The treatment efficiency of this constructed wetland was relatively high, although it was in operation for only eleven months.  The average treatment efficiencies during the period from March 1999 to January 2000 for NH4-N, PO4-P, fecal coliform bacteria, BOD5 and COD were 90.54%, 68.59%, 99.99%, 85.58% and 81.08%, respectively.  These results are very promising with respect to the use of constructed wetland to be installed and developed in tropical countries, especially in Indonesia as a viable alternative to conventional wastewater technology, because this system is cost effective and simple.

Introduction

Environmental pollution in Indonesia, especially at rivers, lakes and other public water bodies, has been increasing considerably over the past few years. The main sources of water pollution in Indonesia are domestic wastewater (40%) and industrial wastewater (30%), and the rest is contributed from agricultural wastewater, animal husbandry wastewater or others.  In Indonesia only around 25% of wastewater is being treated mostly at the primary level prior to disposal, and the remaining 75% of untreated wastewater is discharged into the rivers or other public waters.  This has created severe environmental pollution problems such as eutrophication and transmission of waterborne diseases (cholera, typhoid, dysentery and hepatitis).  Conventional systems of sewage treatment can be very effective.  However, they do have limitations in Indonesia.  Lack of local technical ability combined with high repair and maintenance costs often requires expensive foreign currency and can cause system failures.
In recent years there has been increased interest in an alternative technology with the aims of developing low cost, low maintenance and energy efficient methods of treating sewage.  Such a kind of system (Constructed wetland for wastewater treatment) was first established in Germany through the work of Dr. Seidel and Prof. Kickuth in the 1960s.  This system offers a simple and effective process design and performs well not only for municipal sewage treatment but also for agricultural and industrial wastewater (REDDY and SMITH, 1987).  It is highly appropriate for use in third world countries, due to their low capital and operational costs compared to the conventional system.
At present, there are thousands of constructed wetland for wastewater treatment systems in Germany (KUNST and FLASCHE, 1995).  The efficiency of constructed wetland varies according to the type of substrate, plant species, hydraulic conductivity, influent characteristics, loading cycles (continuous or intermittent), and type of flow (vertical or horizontal).  Constructed wetlands with horizontal flow system are an appropriate technology for high removal of BOD5 and COD, but nutrient removal is only 30-50% (SCHIERUP et al., 1990).  To achieve higher removal rates, constructed wetlands with vertical flow system have been developed.  The advantage of constructed wetland with vertical flow system compared to horizontal flow system lies in its  smaller land requirement (PLATZER and MAUCH, 1997) and higher removal of nutrients (FLASCHE, 1995).  Vertical flow systems require only 2.5 m2/population equivalent (p.e) compared to 5 m2/p.e for horizontal flow systems (ATV, 1997).
Although the use of constructed wetlands for wastewater treatment has received international attention, performance data of constructed wetlands operating under tropical conditions are only scarce.  In addition, there is no information stating that constructed wetlands for wastewater treatment have already been constructed in Indonesia.  Tropical climatic conditions like in Indonesia are conducive to the rapid establishment of aquatic macrophytes and all biological activities will be more efficient.  These conditions, together with the availability of materials locally and the results of a previous research project of the author in Germany, imply a great potential of vertical flow system of constructed wetland for water pollution control to be installed in Indonesia.
The objective of this study was to install a constructed wetland with vertical flow system to treat sewage from a private household to serve as a pilot project.

Materials and Methods
A constructed wetland (5m long, 3m wide and 1.1 m deep) to treat sewage from family Subandi house has been built in Bandung, Indonesia in February 1999.  The constructed wetland is a subsurface flow constructed wetland (6 p.e., 2.50 m2 surface/p.e., 15 m2, vertical flow, discontinuous feeding by timing device and drainage system spread over the whole bed area), planted with Phragmites karka at a density of seven plants per m2.  The wastewater was mechanically pre-treated in a sedimentation tank (3 m3) and pumped onto the sand filter via a polyethylene pipe.  The sewage was pumped onto the filter bed once a day.  A polyethylene membrane served to seal the bed.  The filter bed was built from a multi-layer with sand as the main media (Figure 1).  Small size gravel (8-16 mm) was used in the first top layer (10 cm), followed by 15 cm of bigger size gravel (8-32 mm) and another 5 cm of small size of gravel (8-16 mm). Sand with a hydraulic conductivity (Kf value) of 6.2 x 10-4 m/s, d10 (0.25 mm) and uniformity (U) 4.0 was used as the main layer (60 cm deep), followed by 5 cm of small size gravel (8-16 mm) and finally, at the bottom, 15 cm larger sized gravel (16-32 mm).  The treated water was collected in a drain at the bottom of the filter bed and used again as irrigation water for gardening or directed to the nearest public waters.
The water samples from both influent and effluent were taken twice a month for a period of eleven months (March 1999 till January 2000) and analysed in the Wastewater Laboratory, Faculty of Agriculture, Padjadjaran University, Bandung, Indonesia and Wastewater Laboratory of PDAM Bandung, Indonesia, for COD, BOD5, NO3-N, NO2-N, NH4-N, PO4-P, pH, fecal coliform bacteria, electrical conductivity and settleable solids.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Results
The treatment efficiency of constructed wetland Subandi in Bandung, Indonesia, was already relatively high, although this constructed wetland was in operation for only eleven months (Table1).  During this study period (March 1999-January 2000), the treatment efficiency for NH4-N varied from 70.58% to 99.3%, PO4-P (6.25% to 92.51%), BOD5 (66.67% to 93.47%), COD (50% to 94%), total-N (17.56% to 50.31%), and fecal coliform bacteria (90% to 99%).  These results were still highly variable.  This was probably because of the wetland and the sedimentation tank was newly constructed.  The nitrification processes in a newly wetland system usually occur slowly and are not constant due to limited amounts of bacteria.
The average concentrations of BOD5 and COD in effluent during the period of March 1999-January 2000 were 28.86 mg l-1 for BOD5 and 68.50 mg l-1 for COD (Table 2 and Figure 2).  These values are relatively low and fall considerably short of the German effluent standards for BOD5 (40 mg l-1) and COD (150 mg l-1) (ATV, 1998).
The average concentration of NH4-N in effluent during the period of March 1999-January 2000 was 6.59 mgl-1 NH4-N, that of NO3-N 22.55 mg l-1, of NO2-N 2.02 mg l-1 and of PO4-P 5.88 mg l-1 (Table 2 and Figure 2).  At present, there are no effluent standards for BOD5, COD and nutrients in Indonesia.  The average concentration of PO4-P in effluent was quite high (5.88 mg -1l).  This was probably due to a high consume of detergents for washing and house cleaning.
 
 

Tab. 1. Monthly treatment efficiency of constructed wetland for wastewater treatment at
            family Subandi house in Bandung, Indonesia, during the period of March 1999 – January 2000
 
Parameter Treatment Efficiency (Percent)
 Mar April May Jun July Aug Sept Oct Nov Dec Jan
SS  100 100 100 100 100 100 100 100 100 100 100
NH4-N 96.77 88.96 98.72 98.39 99.30 70.58 97.29 87.45 89.86 84.60 84.08
NO2-N -99.50 -66.40 -91.00 -86.90 -75.90 -65.6 33.33 52.68 32.09 45.56 24.13
BOD5  71.43 90.91 90.90 92.85 66.67 90.90 92.86 93.47 85.71 88.88 76.78
COD 55.55 94.00 91.38 88.57 50.00 83.33 88.23 83.69 81.55 89.73 85.91
PO4-P  92.51 74.85 6.25 73.83 92.00 90.22 59.37 56.49 46.30 89.20 73.53
Total-N n.a n.a n.a n.a n.a n.a 25.45 20.22 17.56 50.31 42.14
Fecal Coli 99.99 99.99 99.95 99.95 99.99 99.99 99.97 99.96 99.00 97.76 99.21
Note : n.a : not available

The average concentration of fecal coliforms bacteria in the influent of this wetland was 6.2 x 108 fecal coliforms bacteria per 100 ml.  This concentration was still in the range commonly found in  settled domestic sewage (GERSBERG et al., 1989).  The final concentration of  fecal coliform bacteria in the effluent was 9.3 x 103, but this concentration tended to decrease drastically (to 650-180 fecal coliforms bacteria per 100 ml) during the period from November 1999 to January 2000.  These effluent concentrations were clearly still below the WHO (1989) guideline value (1000 fecal coliforms per 100 ml) for unrestricted irrigation.
The average concentration of oxygen in effluent (6.58 mg l-1 O2) of constructed wetland Subandi was higher than in influent (1.10 mg l-1 O2).  The average values for pH, temperature and electrical conductivity were also somewhat higher in influent than in effluent (Table 2).
Tab. 2. Average concentrations of SS, NH4-N, NO2-N, NO3-N, BOD5, COD, PO4-P, Total-N, pH, Temperature, O2,
            electrical conductivity and fecal coliform bacteria from influent and effluent of constructed wetland
            Subandi in the period from March 1999 to January 2000.

Parameter Influent Effluent
Settleable  Solids (ml/l) 0.19 0.0
NH4-N   (mg/l) 37.40 6.59
NO2-N   (mg/l) 0.67 2.02
NO3-N   (mg/l) 11.08 22.55
BOD5       (mg/l) 229.54 28.86
COD      (mg/l) 460.82 68.50
PO4-P    (mg/l) 18.70 5.88
Total-N  (mg/l) 93.90 65.60
PH 7.64 6.59
Temperature oC 22.58 22.55
Oxygen   (mg/l) 1.10 6.58
Elec. Conductivity (µS/cm) 1750 1465
Fecal Coliform (MPN/100 ml) 6.2 x 108 9.3 x 103

Discussion
 This study demonstrates that constructed wetland Subandi (vertical flow system) in Bandung, Indonesia, had high a treatment efficiency in terms of the biochemical oxygen demand (BOD5), chemical oxygen demand (COD), NH4-N, PO4-P and fecal coliform bacteria.  COOPER and GREEN (1995) stated that constructed wetland with vertical flow system can achieve full BOD5 and COD removal because of high amounts of oxygen transfer through the reed bed.  The average concentration of oxygen in constructed wetland Subandi in Bandung, Indonesia, was raised from 1.10 mg l-1 O2 in influent to 6.58 mg l-1 O2 in effluent.  Higher oxygen content in this wetland is required for bacteria to remove both organic and nutrient pollutants.  PLATZER and MAUCH (1997) reported that removal efficiency of constructed wetland with vertical flow system is mainly based on very efficient soil aeration and therefore, BOD5, COD and NH4-N removal is high, but total-N elimination is limited.  According to WOOD (1990) the colloidal and soluble BOD5 and COD remaining in solution is removed as a result of the metabolic activity of microorganisms and physico-chemical interaction within the root zone.
 The overall results are of great promise concerning the possibility of constructed wetland to be installed and developed in tropical countries, especially in Indonesia, as a viable alternative to conventional wastewater technology, because this system is cost effective and simple and can be regarded as the appropriate technology.  This system is also suitable for small to medium sized communities in sparsely populated areas and in agricultural areas (e.g. transmigration areas).
Zusammenfassung
Pflanzenkläranlage zur Reinigung von Hausabwässern in Bandung, Indonesien.
Eine Pflanzenkläranlage zur Reinigung der Abwasser des Hauses von Familie Subandi wurde in Bandung, Indonesien, im Februar 1999 fertig gebaut (als Pilotprojekt). Von Zulauf und Ablauf  wurden alle 15 Tage Wasserproben genommen und auf CSB, BSB5, NO3-N, NO2-N, NH4-N, PO4-P, pH, fäkale coliforme Bakterien, Leitfähigkeit und Absetzbare Stoffe analysiert.  Das Ziel diese Studie war es, eine Pflanzenkläranlage mit vertikalem Flußsystem unter Benutzung von Schilfpflanzen (Phragmites karka) an einem Privathaus zu bauen  und zu erproben. Die Leistungsfähigkeit dieser Anlage war relativ hoch, obwohl sie nur 11 Monate in Betrieb war.  Die durchschnittliche Reinigungsleistung der Anlage in der Zeit von März 1999 bis Januar 2000 betrug: NH4-N = 90.54%, PO4-P = 68.59%, fäkale coliforme Bakterien = 99.61%, BSB5 = 85.58%, und CSB = 81.08%.  Diese Ergebnisse sind sehr vielversprechend und lassen darauf schließen, daß solche Anlage in tropischen Ländern, besonders in Indonesien, aufgrund ihrer Effektivität und ihrer günstigen Kosten als Alternative zu gewöhnlichen Kläranlagen gebaut werden können.
References
ABWASSERTECHNISCHE VEREINIGUNG/ATV, 1997:  Grundsätze für Bemessung, Bau und Betrieb von Pflanzenbeeten für kommunales Abwasser bei Ausbau größen bis 1000 Einwohnerwerte. Regelwerk Abwasser-Abfall. Hinweis A. 262.
ABWASSERTECHNISCHE VEREINIGUNG/ATV, 1998:  Behandlung von häuslichem Abwasser in Pflanzenbeeten.  Regelwerk Abwasser-Abfall. Hinweis H. 262.
COOPER, P.F. and GREEN, B., 1995: Reed bed treatment systems for sewage treatment in the United Kingdom-the first 10 years experiences.  Water Science Technology 32, 317-327.
FLASCHE, K. (1995).  Untersuchung der mikrobiellen Abbauvorgänge in Vertikalfiltern. ÖWAF-Seminar, Wien ??Seitenzahlen??
GERSBERG, R.M., GEARHEARTH, R.A., and IVES, M., 1989: Pathogen removal in constructed wetlands. In: D. Hammer (ed.), Constructed wetlands for Wastewater Treatment: Municipal, Industrial and Agricultural,  431-445. Lewis Publishers Inc., Chelsea, Michigan.
KUNST, S., and FLASCHE, K., 1995: Untersuchungen zur Betriebssicherheit und Reinigungsleistung von Kleinkläranlagen mit bewachsenen  Bodenfilter. Abschlußbericht, Universität Hannover.
PLATZER, C. and MAUCH, K., 1997:  Soil clogging in vertical flow reed beds-mechanism, parameters, consequences and solutions ?.  Water Science and Technology 35, 175-181.
SCHIERUP, H.H., BRIX, H. and LORENZEN, B., 1990: Wastewater treatment in constructed reed beds in Denmark-state of art.  In: Cooper, P.F. and Findlater, B.C. (eds.), Constructed Wetlands in Water Pollution Control, 495-504. Pergamon Press.
REDDY, K.R.. and SMITH, W.H., 1987: Aquatic plants for water treatment and resource recovery.  Magnolia Publ., Inc. Orlando, Fl.
WHO, 1989: Health Guidelines for the use of wastewater in agriculture and aquaculture.  Technical Report Series No. 778.  World Health Organization, Geneva.
WOOD, A., 1990: Constructed wetlands for wastewater treatment-Engineering and design consideration. In: Cooper, P.F. and Findlater, B.C. (eds.), Constructed Wetlands in Water Pollution Control, 481-494. Pergamon Press.
 

Address of authors:
D. Kurniadie, Department of Agronomy, Faculty of Agriculture, Padjadjaran University, Jatinangor, Bandung 40600 Indonesia.
Chr. Kunze, Institut für Pflanzenökologie, der Justus Liebig Universität, Lehrstuhl für Botanik 2.  Heinrich-Buff-Ring 26-32 Giessen, 35392, Germany.