Performance of a Pilot-Scale Biofilters and Constructed ... - CiteSeerX

World Applied Sciences Journal 6 (11): 1555-1562, 2009 ISSN 1818-4952 © IDOSI Publications, 2009

Performance of a Pilot-Scale Biofilters and Constructed Wetland with Ornamental Plants in Greywater Treatment Teck-Yee Ling, Kasing Apun and Siti-Rubiah Zainuddin Faculty of Resource Science and Technology, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysia Abstract: Partially treated black water from septic tanks and grey water from households in Kuching City were polluting the Sarawak River. A pilot scale ecological sanitation was implemented where blackwater was held in septic tanks and greywater was channeled to a grease trap, biofilters and a constructed wetland before discharge. The objectives of this study were to evaluate the efficiency of the biofilters and the constructed wetland with two species of terrestrial ornamental plants in greywater treatment. For the combined system, results indicated that the influent dissolved oxygen of below 1 mg/L improved to 3.4-4.6 mg/L. Removal of biochemical oxygen demand (BOD5) and chemical oxygen demand (COD) were the highest (99 and 95%). Influent fecal coliform (FC) of 4.0x105 CFU/100mL dropped to 6.3x102 CFU/100mL at the effluent. More than 80% of ammonia-nitrogen (NH4-N), total nitrogen (TN) and total suspended solids (TSS) were removed. Reactive phosphorus (RP) and total phosphorus (TP) removals were 64 and 61% respectively. The biofilters contributed most of the total removal of BOD5, COD, FC and NH4-N. However, the wetland and biofillters were equally efficient in TSS removal. The removal efficiency of the wetland in decreasing order was NO 3-N>FC> TSS>BOD5>TN>COD>NH4-N>RP>TP and all removals exceeded 55% with the exception of P (38-39%). Both species of plants grown on the wetland contained significantly higher weight and P content than the control. Tissue P content of F. microcarpa was significantly higher than S. campanulatum indicating F. microcarpa as a better accumulator of P. This indicates that constructed wetland with F. microcarpa could be potentially implemented in urban housing areas to reduce river pollution. Key words: Household wastewater

Constructed wetland

INTRODUCTION With 21% of the 2.2 million population of Sarawak State concentrated in Kuching City, treatment of wastewater from households is a challenge. Partially treated black water from septic tanks and grey water were discharged into storm water drains and subsequently into the rivers. It was reported that the main pollution source of the Sarawak River, running through the city, was the discharge from households [1]. Implementing a centralized wastewater treatment system in our cities would be expensive due to construction and maintenance. On the other hand, biological methods are regarded as economical in terms of construction and running and less polluting [2]. Therefore, the option of an urban ecological sanitation (EcoSan) was explored by the Sarawak State Government.

Biofilter

Grey water

Ornamental plants

EcoSan is a cycle-a sustainable, closed-loop system, which closes the gap between sanitation and agriculture and thus it involves separating excreta at source and recycling of the nutrients [3]. A comparative life cycle assessment of an EcoSan was carried out for an office building by comparing it with conventional systems and it was reported to reduce contribution to ecosystem quality damage by more than 60% and also found to be a promising alternative small scale wastewater treatment system [4]. In this type of sanitation, human excreta are treated as a resource and are sanitized before being recycled as fertilizer. The grey water from kitchens, baths and laundries, though not mixed with the toilet water, has to be treated before being reused. Since most residential areas have a park area allocated for recreation, the treatment system could be constructed in the park. Thus, a pilot project of EcoSan was implemented at Hui

Corresponding Author: Teck-Yee Ling, Faculty of Resource Science and Technology, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysia

1555

World Appl. Sci. J., 6 (11): 1555-1562, 2009

Sing Garden in Kuching, Malaysia [5]. Grey water from those households were channeled to a grease trap and then pumped to biofilters before flowing through a subsurface constructed wetland (SSFCW) prior to discharge. A biofilter consists of filter media with attached-growth microorganisms to carry out biological treatment processes to remove organic matter and nutrients from wastewater [6]. A wetland is an inundated land area with water depth typically less than 0.6 m that supports the growth of emergent plants [6]. SSFCW is becoming increasingly popular for wastewater treatment worldwide due to its low cost in construction and maintenance in addition to its use of natural processes in pollutants removal [6]. Studies showed that SSFCW was effective in removing pollutants such as suspended solids, organic matter and faecal coliforms and nutrients [7, 8 and 9]. However, for SSFCW, aquatic plants were usually used to enhance nutrients removal. Most of the studies used plants such as cattail (Phragmites sp.) and reed (Typha sp.) [6, 10]. Knowledge on the potential use of tropical ornamental terrestrial plants is lacking in literature. Since ornamental terrestrial plants are highly demanded in landscaping, they could be planted on the wetland where they could serve the same purpose as the aquatic plants in terms of uptake of nutrients and their roots providing media for filtration and adsorption of solids and also available surfaces on which bacteria which are responsible for degradation of organic matter could grow [11]. A pilot-scale treatment of domestic wastewater in a subsurface flow constructed wetland planted with two ornamental species (Canna and Heliconia) in Thailand was recently reported [12]. However, other ornamental plants in the tropics have not been investigated. Therefore, this study aimed to investigate the efficiency of the biofilters combined with the constructed wetland with ornamental plants in treating greywater. Specifically, the objectives were to determine the efficiencies of biofilters, constructed wetland with ornamental plants and the combined system in improving water quality and also to compare the suitability of two species of ornamental plants in the removal of P.

EcoSan was implemented in 2003 [5]. The septic tanks of nine households were converted into holding tanks and low-flush toilets were installed. The content of the holding tank was emptied periodically and transported to Kuching Sludge Treatment Center. The greywater flows through an oil and grease separator before pretreatment biofilters or trickling filters. With the help of a pump, wastewater was transported to four biofilters of two meters in diameter where it was sprayed intermittently on top of a lightweighted material (expanded clay). Outflow from the biofilter entered a horizontal subsurface flow constructed wetland (SFW) filled with 1.5-2.5 cm sized crushed limestone. The biofilters and wetland were totally covered with soil of about 10 cm thick. The sizing of the wetland was based on population equivalent of 45. The hydraulic retention time was 20 hours and the discharge was 6-8 m3/d. Treated effluent flowed into the storm water drain. The schematic diagram of the setup is shown in Fig. 1. Two species of ornamental plant (Syzygium campanulatum and Ficus microcarpa) were planted on the wetland with each row having a mix of the two species. For each species, eighteen seedlings of a year old were planted. Another plot was setup in the same park where the plants were not grown on the wetland as a control. The plants were planted 3 months prior to the first sampling and they were not fertilized throughout the experimental period.

MATERIALS AND METHODS Pilot Project: Hui Sing Garden is a residential area located 4.5 km from downtown of Kuching City. Prior to the implementation of EcoSan, black water from the toilets was partially treated in the septic tanks and the outflow was discharged into the storm drains. The pilot project of

Sample Collection and Analysis: Sample collection was carried out five times between October 2005 and March 2006. Water samples were collected at around 9:00 am at the pump-sump (1), four rows of sampling ports (2, 3, 4 and 5) along the SSFCW and the discharge point (6) at the storm water drain. On each row of sampling ports, equal volumes were collected from the four ports to form a composite sample. Samples collected were transported to the laboratory at 4°C for immediate analysis. Dissolved oxygen (DO), pH and temperature were determined in situ. DO was determined using a DO meter (Jenway), pH and temperature by a pH meter (Cyberscan Water proof 300). The meters were calibrated in the laboratory prior to in situ measurement. One to two plants of each species were randomly sampled for dry matter and P content analysis. Water quality parameters analyzed were total suspended solids (TSS), biochemical oxygen demand (BOD5), chemical oxygen demand (COD), ammonianitrogen (NH4-N), nitrate-nitrogen (NO 3-N), total nitrogen

1556

World Appl. Sci. J., 6 (11): 1555-1562, 2009

(TN), reactive phosphorus (RP), total phosphorus (TP) and fecal coliform (FC). TSS, BOD5 and COD were analyzed according to the Standard Methods [13]. For COD, the closed reflux colorimetric method was used. The method for analysis of NH4-N, NO 3-N, TN, RP and TP followed that of [14]. Nessler method was used to determine the concentration of NH4-N in the distillate. TNT Persulfate digestion method was used to analyze TN. RP was determined by ascorbic acid method and TP was analyzed according to the acid persulfate digestion method. For P content in plant tissue, the whole plant including roots was cleaned, dried and ground before analysis. Parameters analyzed were dry matter and P. The analyses followed methods of [15]. Dry ashing followed by ascorbic acid methods were used to determine the concentration of P in the sample. FC counts were quantified following standard microbiological method using membrane filter cultured on m-FC medium incubated at 37°C for 24 hours [16]. All analyses were conducted in triplicates.

equally effective in removing solids where the effluent TSS decreased to 6.7 mg/L. As water flowed through the wetland, further entrapment and sedimentation of the solids occurred due to the presence of granular medium which reduced the hydraulic surface loading rates [9].

Statistical Analysis: Significant difference of each parameter between the influent and effluent concentration was analyzed using paired t-test. Significant difference in weight and P content in the tissue between plants grown on wetland and control plot was analyzed using Univariate ANOVA. All data analyses were conducted using SPSS version 12.0 package. RESULTS AND DISCUSSION Temperature, pH and TSS: The temperature, pH and mean TSS during the experimental period are shown in Fig. 2. The mean temperature in the outflow (29.6°C) was significantly lower than the inflow (30.2°C) (P=0.035). The mean temperature was high as this is a tropical country with high ambient temperature. Mean pH of 6.3 at the inflow to the biofilters increased significantly to 6.9 at the effluent (P=0.010). Similar observation of small increase in pH was reported by Kaseva [6], who studied SSFCW treatment of hostel wastewater using aquatic plants in a tropical climate. TSS decreased from 42.9 mg/L at the influent of biofilter to 26.1 mg/L at the effluent (Fig. 2). The influent concentration at the biofilter was not very high possibly due to some sedimentation at the sump. The large decrease in TSS in the biofilters indicates sedimentation and filtration of particles. The wetland was found to be

DO, RP and TP: The influent DO level improved significantly after passing through the biofilters and the wetland (PBOD5>COD>NH4-N>TSS>TN>RP>TP. The proportion of total reduction of each parameter due to biofilters is shown in Fig. 7. Reduction of 39%, 96%, 88%, 67%, 50%, 37%, 43% and 96% of TSS, BOD5, COD, NH4-N, TN, TP, RP and FC respectively was observed after passing through biofilters. Biofilters were more efficient in reducing those parameters except for TSS where no significant difference in efficiency between biofilters and wetland was observed (P=0.75). However, the presence of wetland resulted in further reduction of 43% of TSS, 19 % of NH 4-N, 31% of TN, 22% of RP and 24% of TP. For the efficiency of the

1560

World Appl. Sci. J., 6 (11): 1555-1562, 2009

wetland itself, the removal in decreasing order was NO3N>FC>TSS>BOD5>TN>COD>NH4-N>RP>TP asshownin Fig. 8. The removal of NO3-N formed in the biofilters from the oxidation of NH4-N was the highest. NO3-N, NH4-N and FC removals in the present study (87%, 55.9% and 75.1%) were much more than that reported by Kaseva [6] in a tropical climate with wetland plants P. mauritianus and T. latifolia (40.3%-44.3% for NO3-N, 23-25% for NH4-N and 43-72% for FC). For COD removal, the result of the present study (57.9%) falls within the range of that reported by Kaseva [6] (56.3-60.7%). Wetland with ornamental plants played an important role in the removal of NO3-N that was formed by the oxidation of NH4-N in the biofilters. Constructed wetland planted with ornamental plants Canna and Heliconia was also reported to show TSS removal of more than 88% and COD of 4-83% [12]. Ornamental Plants: The mean weights of the two ornamental plants S. campanulatum and F. microcapa grown on the constructed wetland were higher compared to those of the control plot (Fig. 9). According to ANOVA test, the differences were significant (P=0.019 for S. campanulatum; P=0.003 for F. microcapa). The P content in the plant tissue of F. microcarpa was higher than that in S. campanulatum (Fig. 10). Furthermore, the P content of both species of plants grown on the wetland were consistently higher than those from the control plot and the difference was significant (P