Nutrient removal from polluted river water by using constructed wetlands

Nutrient removal from polluted river water by using constructed wetlands

Bioresource Technology 76 (2001) 131±135 Nutrient removal from polluted river water by using constructed wetlands Shuh-Ren Jing a,*, Yin-Feng Lin a, ...

118KB Sizes 0 Downloads 124 Views

Bioresource Technology 76 (2001) 131±135

Nutrient removal from polluted river water by using constructed wetlands Shuh-Ren Jing a,*, Yin-Feng Lin a, Der-Yuan Lee a, Tze-Wen Wang b a

Department of Environmental Engineering and Health, Chia-Nan University of Pharmacy and Science, Tainan, Taiwan, ROC b Department of Pharmacy, Chia-Nan University of Pharmacy and Science, Tainan, Taiwan, ROC Received 19 April 2000; received in revised form 27 June 2000; accepted 17 July 2000

Abstract The Erh-Ren River is one of the most polluted rivers in Taiwan. Although its ¯ow rate is relatively low, the rate is still beyond the capacity of any traditional water treatment facility. A pilot-scale constructed wetland (CW) is the attempt used to purify the highly polluted river water and to collect data for the construction and operation of a full-scale system in the future. This article reports the results from this initial stage of our research work. During the study, the most ecient nutrient removal occurred between April and October. The monthly average removal rates of chemical oxygen demand (COD) ranged from 13±51% of ammonia-N (AN) from 78±100%, and of orthophosphate (OP) from 52±85%. After November, input COD levels increased, and the monthly average removal rates of ammonia-N dropped to 16% and of orthophosphate to 13%. The dramatic changes in removal eciency suggest that the macrophytes in the CW had a direct in¯uence on the water treatment and that the change of seasons and the quality of the river water inhibited the growth of the macrophytes. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Constructed wetlands; Nutrient removal; River water puri®cation; Treatment; Water resource reservation; Macrophytes

1. Introduction Wetlands are among the most important links in the natural ecosystem. The major functions of wetlands include holding and recycling nutrients, providing wildlife habitats, stabilizing shorelands, controlling and bu€ering natural ¯oods, recharging groundwater, providing treatment for pollutants in water, and so on (Hammer and Bastian, 1989; Day et al., 1998; Hughes et al., 1998; Mitsch and Cronk, 1995; Hamilton et al., 1997). To date, the most frequent application of wetlands to a river system is for ¯ow management (Zalidis, 1998; Chauvelon, 1998), and little is known about using wetlands to treat polluted river water (Green et al., 1996). The Erh-Ren River is located in southern Taiwan and is highly polluted by discharges such as untreated municipal wastewater, wastewater and runo€ from swine farms, and toxic water from metal-processing factories along the upland regions. According to our long-term monitoring in 1998 on the quality of the river water, the annual average concentration of chemical oxygen de-


Corresponding author. Fax: +886-6-2667 323. E-mail address: [email protected] (S.-R. Jing).

mand (COD) was 96 mg/l, ammonia-N (AN) 8 mg N/l, orthophosphate (OP) 2 mg P/l, and suspended solids (SS) 92 mg/l. During the periods from March to April and from November to January, autumn overturn in the river occurred, and the average concentrations of the above constituents rose to 191 mg/l (COD), 14 mg N/l (AN), 3 mg P/l (OP) and 98 mg/l (SS). As a result of these high levels of pollutants, water from the Erh-Ren River (especially downstream) cannot be used for any practical application, which is a serious problem for a place with limited water resources such as Taiwan. Although the Erh-Ren River does not have a high ¯ow rate (an average of 499 million m3 per year according to data from the Taiwan Province EPA, 1999), it still would be prohibitively expensive to apply any traditional wastewater treatment system to purify the river water suciently for it to be suitable as a source of water for daily usage. Under such conditions, the effective and essentially free treatment of polluted water through a wetlands system becomes a reasonable option. A pilot-scale constructed wetlands (CWs) system containing a free surface ¯ow (FSF) unit and a subsurface ¯ow (SSF) unit was used to treat the highly polluted water in order to collect data needed to evaluate the possibility of using a large-scale CW system to purify a

0960-8524/01/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 8 5 2 4 ( 0 0 ) 0 0 1 0 0 - 0


S.-R. Jing et al. / Bioresource Technology 76 (2001) 131±135

natural water system. As the CW system requires such a long period to reach a stable condition, this work shows only the initial results of our research. 2. Methods 2.1. The CW systems In order to treat the water of the Erh-Ren River directly, the CW systems were set up by the riverbank about 8 km from the coast of the Taiwan Strait. This pilot-scale CW system is combined with a FSF unit and a SSF unit (Fig. 1). The FSF unit was 3.9 m long, 1.9 m wide, and 1.0 m high. A soil layer 41 cm thick lined the bottom of the FSF unit, and the surface water depth was 24 cm. This unit was separated into two cells in parallel. One cell was planted with emergent plants (Pennisetum alopecuroides L.) with a density of two plants per m2 . The other cell was planted with ¯oating vegetation (Lpomoea reptans) with 50% surface coverage. The SSF unit was 2.9 m long, 1.4 m wide, and 0.85 m high. A 57 cm thick layer of gravel (20 mm average in size) was placed inside, providing a void volume of 40%, and the water level within the gravel was 30 cm deep. Phragmites communis L. was planted in the gravel with a density of two plants per m2 . A control experiment (CE) system was operated under the same conditions as the CW system in order to compare the treatment of polluted water in a system without

macrophytes. The CE system was 1.4 m long, 0.7 m wide, and 0.95 m high. This system was separated equally in length. The front part was a FSF cell with a 27 cm thick layer of soil and with surface water 26 cm deep, followed by a SSF cell with a 31 cm thick layer of gravel. All the systems were made of cast iron, and the inside walls were covered with 0.1 cm thick impermeable plastic liner. 2.2. Operation of the CW systems River water was pumped into a water tank. In¯uent to the FSF unit of the CW system was fed by using a constant ¯ow pump, with equal ¯ow rates in both the emergent- and ¯oating-plant cells. The ¯ow rate was set at four days hydraulic retention time (HRT), based on the volume of free surface water in the FSF unit. Ef¯uent from the FSF unit then ¯owed into the SSF unit gravitationally for further treatment. The e‚uent from the SSF unit was the ®nal e‚uent and was discharged back to the river. Flow to the CE system was set at the same HRT as the CW system. A rainfall cylinder placed by the system recorded the rainfall and evaporation rate in order to adjust the concentrations of measured constituents in the water samples. 2.3. Sampling and measurements The water was sampled twice a week under normal operating conditions. All the sampling points are shown

Fig. 1. Schematic diagram of the Erh-Ren River CW system. FSF unit: free-surface ¯ow wetland system, SSF unit: SSF wetland system, CE: control experiment wetland system (without macrophytes), Sampling points: A: composite river water, B: e‚uent of the emergent plant unit, C: e‚uent of the ¯oating plant unit, D: combined e‚uent of the FSF units or in¯uent of SSF unit, E: e‚uent of the SSF unit and the ®nal treatment result, F: e‚uent of the CE system.

S.-R. Jing et al. / Bioresource Technology 76 (2001) 131±135

in Fig. 1. Water samples were measured for COD, AN, OP, nitrate-N, nitrite-N, SS, E. coli, chlorophyll, and some metals. The methods used for the measurements were those speci®ed by the ``Standard Methods for the Examination of Water and Wastewater'' (APHA, 1995). 3. Results and discussions 3.1. The characteristics of Erh-Ren River water Several physical properties of the Erh-Ren River water were measured throughout the prolonged observation period (from March 1998 to April 1999). The pH value of the water ranged from 7 to 8.2, with an average of 7.6. Dissolved oxygen (DO) concentration was highly a€ected by the spring and autumn overturns of the river. In the strati®cation period (in November), DO was reduced as a result of an oxygen de®cit caused by decaying organic compounds in the sediment. In addition, the spring and autumn overturns caused toxins, such as metals, in the river sediment to be released to the water. As the testing location was only about 8 km from the coast, salinity was detected in the in¯uent during high tide hours in the winter season when the river ¯ow rate was signi®cantly low (Table 1). The highest concentration of salinity found in the composite water in the water tank was 2.3%. Water temperatures also changed with the seasons, ranging from 21°C to 35:4°C, with an annual average of 27:7°C. This temperature range is normally very conducive to the growth of macrophytes in a CW system. Due to the increased salinity, levels of toxins, and decreased DO, water quality in the Ehr-Ren River (and consequently in the CW system) deteriorated signi®cantly from November through March (Table 2). During this period, we found that the macrophytes in the CW system stopped growing, and some even started dying o€. In addition, we found that the treatment ef®ciency was directly a€ected by the growing conditions of the macrophytes in the system. As discussed below, all measures of treatment eciency decreased during this period. 3.2. COD removal The most ecient COD removal occurred between April and August in this ®rst year of research on the CW


system. The COD removal rate was as high as 51% between April and October and was lowest between November and March, with COD concentrations sometimes actually higher in the e‚uent than in the in¯uent during the winter season (Table 2). As the oxygen supply from the macrophyte root zone can enhance organic removal by microorganisms in the sediment or on the surface of stems and roots of the macrophytes (Guntenspergen et al., 1989; Reddy and D'Angelo, 1997) the existence of macrophytes in the CW system is important in removing organic materials from polluted water. As vegetation dies in the system, the dead tissue acts as COD to consume oxygen and therefore causes the negative COD removal rate. 3.3. Ammonia removal Generally speaking, the CW system was highly ecient in removing ammonia-N from polluted water during the macrophytes active growing season. Virtually no ammonia-N was detected from August to October in the e‚uent. The monthly average removal eciencies in this period ranged from 94% to 100% and were all clearly higher in the CW system than that in the CE system (Table 2). This result showed that plant uptake was highly responsible for ammonia-N removal. In measurements taken for the entire year, little nitrate and nitrite nitrogen remained in the CW system (Table 3), indicating that the nitri®cation±denitri®cation was sucient in the system. Meanwhile, ammonia-N was found to be the limiting in the CW system during the active treatment period, as compared to phosphate. 3.4. Orthophosphate removal P-removal rates in the CW system gradually increased from August to October, reaching a maximum value of 85.1% in October. Rates gradually decreased afterward to a minimum value of 19.6% in March (Table 2). By contrast, P-removal rates in the CE system were low year-round. However, the limited P-removal (50±80%) rates, compared with high ammonia-N removal (94± 100%), suggest that insucient ammonia-N could be associated with phosphates to be taken up by macrophytes or microorganisms for biomass production. Besides the biotic uptake, other important mechanisms for P-removal in a wetlands system are adsorption

Table 1 Average ¯ow rate of Erh-Ren River in the pasta















Flow rate …m3 =s†













From Wen, 1999, The Review, Discussion and Update of the River Classi®cation in Taiwan ± Attachment 3: Southern Rivers, The Environmental Protection Agent of Taiwan Province.


S.-R. Jing et al. / Bioresource Technology 76 (2001) 131±135

Table 2 Summary of monthly average overall removal eciencies of constituents in the CW and the CE systemsa Constituents

Month System

Mar-98 Unit









Average in¯ow concentration (mg/l) 100.1








40.1 3.1 42.0

32.4 27.5 51.0

31.1 12.5 39.7

)9.8 21.0 13.2

18.6 16.0 31.7

18.4 )10.7 9.7













Removal eciencies (%) CW


4.4 )52.4 )45.7

CE System


Average in¯ow concentration (mg/l)






Removal eciencies (%) CW


70.9 64.6 89.7

CE System


53.4 )11.6 48.0

98.3 100 100.0

72.4 78.3 94.0

92.8 83.6 98.8

83.0 42.1 90.2

41.4 16.4 51.0







Average in¯ow concentration (mg/l)










Removal eciencies (%) CW


13.3 46.2 53.3

CE a

68.0 47.4 52.8

8.3 49.8 54.0

44.7 47.5 71.0

44.8 73.0 85.1

30.1 51.6 66.1

)1.60 39.2 38.3








AN ± ammonia-N, OP ± orthophosphate.

Table 3 Monthly average concentrations of nitrate and nitrite in the CW and the CE systems Month Nitrate (mg N/l) In¯uent FSFa SSFb CEc Nitrite (mg N/l) In¯uent FSF SSF CE










1.109 0.327 0.259 1.204

0.829 0.360 0.171 0.622

0.288 0.234 0.197 0.233

0.212 0.272 0.071 0.086

0.276 0.387 0.273 0.279

0.387 0.407 0.378 0.429

0.360 0.272 0.238 0.215


0.094 0.033 0.002 0.121

0.106 0.002 0.003 0.013

0.127 0.002 0.002 0.007

0.017 0.228 0.001 0.004

0.007 0.014 0.003 0.049


0.063 0.355 0.030 0.437



E‚uent from the FSF unit of the CW system. E‚uent from the SSF unit of the CW system. c E‚uent from the CE system. b

and sedimentation on the soil and gravel (Reddy and D'Angelo, 1997). The P-removal rates varied in the FSF unit with the change of seasons; however, the SSF unit maintained relatively consistent removal rates throughout the year (Table 2). It may be concluded that media sorption played an important role in the SSF unit. Nevertheless, the sorption capacity of media is limited (Sakadevan and Bavor, 1998) and the stability of

phosphate compounds is a€ected by the redox potential and the pH of the media environment (Kadlec and Knight, 1996). Therefore, the best ways to ensure longterm and continuous P-removal in a CW system are to plant a species of macrophytes that can tolerate the highly polluted water conditions and to harvest the vegetation regularly to remove excess phosphorus from the system.

S.-R. Jing et al. / Bioresource Technology 76 (2001) 131±135

4. Conclusion A CWs system clearly can be an e€ective treatment facility for polluted water in natural reservoirs such as rivers. The initial result of this research work, however, found that the ability of a wetlands system to treat such polluted water is dominated by the growth of the macrophytes planted in it. When treating an unstable in¯ow such as river water, the growth of macrophytes was affected not only by the change of seasons, but also by the variation in water quality. During the growing season, organics, nitrogen, and phosphate were all removed from the river water eciently. However, in the autumn months, vegetation naturally began to grow more slowly. Together, spring and autumn overturns occurred in the river, resulting in the release of large amounts of toxic materials, such as metals, from the river sediment. The dual e€ects created very poor growing conditions for the macrophytes, with some plants even dying as a result. Therefore, the ability of the CW system to treat the in¯uent properly was greatly reduced, and at times the water quality was even made worse by the decaying vegetation in the system. Although this preliminary research suggests that CW systems are e€ective in removing pollutants from water, maintaining adequate treatment e€ectiveness continuously throughout the year is clearly the ®rst goal to reach when applying this method to purify river water. To reach this goal, we need to understand in more detail the nature of the CW and its operating parameters. Therefore, the next step of research work would be to cultivate species of domestic macrophytes that can tolerate local seasonal water quality and river strati®cation changes to achieve stable conditions in the CW system and thus maintain treatment eciency. After that, di€erent operating strategies can be investigated, including hydraulic loadings and harvesting frequencies, to collect information towards the evaluation of using full-scale systems to purify polluted river water. Acknowledgements We would like to express our appreciation to the National Science Council of the Republic of China (Project Number: NSC-87-2621-B-041-002) for funding


support to this project, and to Dr. James P. Kaetz, Associate Professor, Department of English, Auburn University and Editor of National Forum, USA, for helping in editing this paper. References APHA, 1995. Standard Methods for the Examination of Water and Wastewater, 19th ed., American Public Health ssociation/American Water Works Association/Water Environmental Federation, Washington, DC, USA. Chauvelon, P.A., 1998. Wetland managed for agriculture as an interface between the Rhone river and the Vaccares lagoon (Camargue, France): transfer of water and nutrients. Hydrobiologia 373&374, 181±191. Day Jr, J.W., Rismondo, A., Scarton, F., Are, A., Cecconi, G., 1998. Relative sea level rise and Venice lagoon wetlands. Journal of Coastal Conservation 4 (1), 27±34. Green, M., Safray, I., Agami, M., 1996. Constructed wetlands for river reclamation: experimental design, start-up and preliminary results. Bioresource Technology 55, 157±162. Guntenspergen, G.R., Stearns, F.F., Kadlec, J.A., 1989. Wetland vegetation. In: Hammer, D.A. (Ed.), Constructed Wetlands for Wastewater Treatment ± Municipal, Industrial and Agricultural. Lewis Publishers, Michigan, pp. 73±89. Hamilton, S.K., Sippel, S.J., Calheiros, D.F., Melack, J.M., 1997. An anoxic event and other biogeochemical e€ects of the Pantanal wetland on the paraguay river. Limnology and Oceanography 42 (2), 257±272. Hammer, D.A., Bastian, R.K., 1989. Wetland ecosystems: natural water puri®er. In: Hammer, D.A. (Ed.), Constructed wetlands for wastewater treatment ± municipal, industrial and agricultural. Lewis Publishers, Michigan, pp. 6±20. Hughes, C.E., Binning, P., Willgoose, G.R., 1998. Characterisation of the hydrology of an estuarine wetland. Journal of Hydrology Amsterdam 211 (1±4), 34±49. Kadlec, R.H., Knight, R.L., 1996. Phosphorus. Treatment Wetlands. Lewis Publishers, Michigan, pp. 443±480 (Chapter 14). Mitsch, W.J., Cronk, J.K., 1995. In¯uence of hydrologic loading rate on phosphorus retention and ecosystem productivity in created wetland, Final Report WES/TR/WRP RE-6. Reddy, K.R., D'Angelo, E.M., 1997. Biogeochemical indicators to evaluate pollutant removal eciency in constructed wetlands. Water Science and Technology 35 (5), 1±10. Sakadevan, K., Bavor, H.J., 1998. Phosphate adsorption characteristics of soils, slags and zeolite to be used as substrates in constructed wetland systems. Water Research 32 (2), 393±399. Wen, C.G., 1999. The review, discussion and update of the river classi®cation in Taiwan ± attachment 3: southern rivers, The Environmental Protection Agent of Taiwan Province. Zalidis, G., 1998. Management of river water for irrigation to mitigate soil salinization on a coastal wetland. Journal of Environmental Management 52 (2), 161±167.