Constructed wetlands in China

Constructed wetlands in China

Ecological Engineering 35 (2009) 1367–1378 Contents lists available at ScienceDirect Ecological Engineering journal homepage: www.elsevier.com/locat...

306KB Sizes 7 Downloads 63 Views

Ecological Engineering 35 (2009) 1367–1378

Contents lists available at ScienceDirect

Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng

Review

Constructed wetlands in China Dongqing Zhang a,∗ , Richard M. Gersberg b , Tan Soon Keat c a

DHI-NTU Centre, Nanyang Environment & Water Research Institute, N1-B3b-29, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore Graduate School of Public Health, San Diego State University, Hardy Tower 119, 5500 Campanile, San Diego, CA 92182-4162, USA c Maritime Research Centre, School of Civil and Environmental Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore b

a r t i c l e

i n f o

Article history: Received 29 May 2009 Received in revised form 27 June 2009 Accepted 20 July 2009

Keywords: China Constructed wetlands Role of wetland plants Design of constructed wetlands

a b s t r a c t Large-scale centralized wastewater treatment systems often prevail in industrial countries and have been regarded as a successful approach during the last century. However, to solve the multifold water-related problems in China with its rapid growth of urbanization and industrialization, complete replication of this centralized, cost- and energy-intensive technology has proved to be extremely limited in scope and success. As one of the most important applications of ecological engineering, constructed wetland (CW) systems for wastewater treatment can offer an optimal alternative and result in beneficial conservation of natural resources with low capital costs and energy consumption, as well as minimal operation and maintenance expenditures. CW technology is particularly suitable for rapidly growing small- and medium-size cities in China. This paper aims at examining the mechanisms of pollutant removal efficiency in these systems and investigating the merits, status and feasibility of using constructed wetland systems to treatment wastewater in China. Additionally, it investigates existing impediments to application and implementation of CWs in China, as well as challenges to future development. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Rapid urbanization and industrialization, and highly accelerated economic development in China have resulted in excessive water consumption and degradation of water resources. Historically, traditional centralized sewer systems have been regarded as the optimal solution for water pollution control and have prevailed in many industrial countries. To a large degree, this centralized approach did and does solve the problems of sanitation very efficiently. However, at the end of 2002, the official rate of municipal wastewater treatment was approximately 36.5%, which is far from adequate given China’s serious water pollution (U.S. Department of Commerce, 2005; Wang et al., 2005b). It follows then, that in order to solve the multifold water-related problems in China, complete replication of centralized water-, energy- and cost-intensive technology has proven to be rather limited and not entirely feasible. Amongst 660 cities in China, more than half of which are of medium- (population between 200,000 and 500,000) and small-size (population less than 200,000) (China Daily Report, 2005), and this is reflected by the fact that 50% of the population of China still resides in these small- to medium-size

∗ Corresponding author. Tel.: +65 8165 6212; fax: +65 6790 6620. E-mail address: [email protected] (D. Zhang). 0925-8574/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ecoleng.2009.07.007

urban areas. While big cities are predominantly served by sewage treatment plants based on conventional intensive technologies (physical–chemical–biological treatment), there is increasing doubt that whether these intensive technologies for sewage treatment systems are appropriate for medium- and small-size municipalities (Brissaud, 2007). Constructed wetlands (CWs) for wastewater treatment have great potential as an optimal alternative and would be ideal for China’s small- to moderate-size cities. Indeed, in other places worldwide, CWs have proved to be an attractive and stable alternative because of their low cost, and energy savings. In addition, there is the advantage of multi-purpose re-use of the high quality effluent, self-remediation and self-adaptation to the surrounding conditions and environment (Song et al., 2006; Brissaud, 2007; Kivaisi, 2001). It is the objective of this paper to review the progress of CWs for wastewater treatment in China with the aim of delineating some of the key treatment efficiency and performance issues which may be elucidated by the China experience, including the following: • • • • •

Design; Specific role of the plant; Effect of climate; Cost/energy/space efficiency; Sustainability.

1368

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

Through a review of the Chinese CW experience, we may better define the scope and issues at hand, and by doing so, overcome certain key challenges for the future development.

ment designed by Yanshan Petrochemical Company in Beijing (Li and Jiang, 1995), and an infiltration wetland system for domestic sewage treatment located on the costal saline–alkali soil on Dagang Oil Field near Tianjin City (Li and Jiang, 1995). These are considered to be milestones of CW application in China (Chen et al., 2008).

2. Ecological engineering in China 3. The design of constructed wetlands As a relatively new branch of ecology and an interdisciplinary science, ecological engineering was initially formulated in the 1960s. Ecological engineering as described by Mitsch and Jørgensen (1989) “is engineering in the sense that it involves the design of this natural environment using quantitative approaches and basing our approaches on basic science.” The term “ecological engineering” was in the 1960s first independently proposed by Prof. Ma Shijun, known as “the father of ecological engineering in China” (Mitsch and Jørgensen, 2003a; Yan et al., 1993; Ma, 1988). He argued in 1978 that in recognition of the interdependency of social, economic and natural systems, a cross between social and natural sciences could form the theoretical basis for treating the ecological crises the world was facing (Ma, 1988; Yan et al., 1993; Ma, 1978). Ma (1988) defined ecological engineering as “. . .a specially designed system of production process in which the principles of the species symbiosis and the cycling and regeneration of substances in an ecological system are applied with adopting the system engineering technology and introducing new technologies and excellent traditional production measures to make a multi-step use of substance” (Mitsch et al., 1993; Mitsch and Jørgensen, 2003a). Yan et al. (1993) reported that during the period of 1970 to 1990, the dissemination of the knowledge and techniques of ecological engineering resulted in a rapid growth of its popularity all over China. Over 2000 experimental sites for the ecological engineering of agriculture and environmental protection have been founded in all the provinces of mainland China. Over 100 sites for the ecological engineering of wastewater treatment and utilization were created as well. Indeed, the special historical background, ancient Chinese philosophy and rich traditions of Chinese agricultural practices, as well as the social–economical settings in China give ecological engineering in China many rich and distinct characteristics. Mitsch et al. (1993) concluded that differences between Western and Chinese systems related to design principles, objectives, human manipulation of ecosystem structure, and recognized values and economics. The emphasis of ecological engineering in the West has been a partnership with nature and research has been carried our primarily in experimental ecosystems rather than in full-scale applications. Ecological Engineering, as pioneered by Ma in China, has been applied to a wide variety of natural resource and environmental problems, ranging from fisheries and agriculture, to wastewater control and coastline protection (Mitsch et al., 1993; Mitsch and Jørgensen, 2003a,b; Mitsch, 1997). In addition, in the west, the goal of ecological engineering projects is usually environmental protection, while in China it is not only environmental protection, but also economic and social benefits (Mitsch et al., 1993). As one of the most important applications of ecological engineering, constructed wetlands (CWs) systems for wastewater treatment could offer an optimal alternative and result in beneficially natural resources conversation with low capital costs, low energy consumption, and minimal operation and maintenance. The first full-scale CW (SSF system) for wastewater treatment from small or medium scale towns in the sub-tropics in China – Bainikeng Constructed Wetland was put in operation in July 1990 at Longgang, Shenzhen Special Economic Zone (Yang et al., 1995). Other early established CWs in China include: the FWS system built in Changping District, Beijing for municipal sewage treatment (Li and Jiang, 1995), the in-series FWS for petrochemical effluent treat-

In general, two types of constructed wetlands systems are most commonly designed and used: the Free Water Surface (FWS) systems, the Subsurface Flow (SSF) systems including horizontal- or vertical flow. SFW systems are similar to natural marshes as they tend to occupy shallow channels and basins through which water flows at low velocities above and within the substrate. In SSF systems, wastewater flows horizontally or vertically through the substrate, which is composed of soil, sand, rock or artificial media. 3.1. Subsurface flow vs. free water surface wetlands Table 1 summarizes several applications of the FWS system in China and presents relative treatment efficiencies. Compared to the discharge standards set by the Chinese Government (Environment Bureau of the State, 1997) (COD < 60 mg/l, BOD5 < 20 mg/l, TN < 15 mg/l, TP < 0.5 mg/l), the majority of effluent values of TSS, BOD5 and COD are generally lower than that required by discharge standards in China. According to Li and Jiang (1995), in all seasons except winter, the levels of BOD5 and TSS in the effluent of every system could reach the standards of biological secondary treatment. However, for reasons that are unclear, perhaps due to lack of information and design/operation expertise, in the last 15 years, there are have been very few FWS reported in the literature. 3.2. Horizontal subsurface flow vs. vertical subsurface flow wetlands In the horizontal subsurface flow (HSSF) system, the influent flow is under the surface of the bed following a horizontal path until it reaches the outlet zone. In the vertical subsurface flow (VSSF) system, however, the wastewater is fed onto the whole surface area through a distribution system and passes through the filter in a vertical path. Due to long retention time, HSSF is more effective for the removal of BOD5 , COD, TSS (Kadlec, 2009; Mander and Mitsch, 2009). The experience in China (see Tables 2 and 3) showed that the mean removal efficiencies TSS and COD of HSSF systems (75.5% and 70.09%) are higher than that of VSSF systems (74.7% and 62.09%), while the mean removal efficiencies of BOD5 of both systems are similar (82.22% and 82.95%, respectively). Usually nitrification is limited due to the lack of oxygen that is characteristic for this kind of system. Whereas in a VSSF system, mostly aerobic conditions provide the ideal environment for oxygen-requiring nitrifying bacteria and nitrification can be achieved in these systems. In a study on high-rate nitrogen removal in a two-stage VSSF system, Langergraber et al. (2008) indicated that nitrogen elimination in this two-stage VSSF system occurred because nitrification of about 80% in the first-stage guaranteed the presence of nitrate for denitrification in the impounded drainage layer. However, denitrification may not take place to a large extent. Table 2 shows the treatment efficiencies of HSSF systems in China. Compared to the discharge standards set by the Chinese Government (Environment Bureau of the State, 1997), the majority of the effluent values of TSS, BOD5 and COD are generally lower than that required by discharge standard in China. Meanwhile, in comparison to the average removal efficiencies of TSS (73.5%), BOD5 (80.3%) and COD (65.5%) from 78 of HSSF systems

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

1369

Table 1 A summary of the treatment efficiency of FWS systems in China. TSS

BOD5

COD

Changping, Beijinga Effluent value (mg/l) Removal efficiency (%)

17 93.8

17.8 85.8

– –

Qinghe, Beijinga Effluent value (mg/l) Removal efficiency (%)

6.1 83.8

5.6 37.7

CW1, Tianjina Effluent value (mg/l) Removal efficiency (%)

19.5 79.9

CW2, Tianjin (infiltration type)a Effluent value (mg/l) Removal efficiency (%)

Hydraulic retention time (day)

TN

TP

Hydraulic loading rate (m3 /day)

– –

5.1 64.6

0.42 55.1

500

7.3

0.34 53.9

5.6



3.08 29.2

120



18.1 84.9

– –

– –

19.54 50.6

0.98 70.3

200

1.5–3.0

11.4 90

10.3 85

– –

– –

6.15 83.4

0.32 86

1800

Public Park, Shanghaib Effluent value (mg/l) Removal efficiency (%)

30 70

7.7 15.4

32 17.9

– –

9.7 10.2

0.53 18.5



Liaohec Effluent value (mg/l) Removal efficiency (%)

– –

3.9 88

77 80

– –

1.6 86

Taihu, Zhejiang Provinced Effluent value (mg/l) Removal efficiency (%)

– –

– –

Average efficiency in China (%) a b c d

83.50

66.13

5.93 16.5 38.13

NH4 -N

1.37 22.8 –

– –

18.75

10

1.2

15

3.97 19.8

0.103 35.1

0.64 m/day



49.11

53.15





Li and Jiang (1995). Li et al. (2009). Ji et al. (2007). Li et al. (2008).

in Europe (Haberl et al., 1995), even under relatively high loading rates, the removal efficiencies of TSS (75.50%), BOD5 (82.22%) and COD (70.09%) in China are slightly better. In particular, the average removal efficiencies of TN (56.09%) and TP (59.0%) in HSSF systems in China are much higher than that in Europe (39.8% and 31.7% respectively). Table 3 summarizes the application of VSSF systems in China. Compared with HSSF systems, VSSF systems usually require smaller foot print (Lüderitz et al., 2001), and it is therefore an attractive alternative for southern China where land is scarce and population density is high. In a pilot VSSF system near Longdao River in Beijing, CW occupied less than half of the area of conventional CW following adoption of an improved design. The authors compared Longdao VSSF system with other HSSF systems in other countries, the removal efficiencies of BOD5 (87.2%), COD (81.8%) and TSS (85.1%) of the VSSF CWs are verified comparable to the highest performers, while the removal efficiencies of TP (98.8%) and NH3 -N (77.4%) are much higher than that in other countries. Additionally, the effluent concentrations of all substances were stable even during the winter (Chen et al., 2008). Experiences showed that in China (see Tables 2 and 3), although VSSF systems are efficient at BOD5 and TP removal because of good oxygen supply, the removal rate of TN (43.66%) and NH4 -N (56.17%) remains lower in comparison of that in HSSF systems (56.09% and 64.59%, respectively), probably due to the lack of carbon source during denitrification. Also, despite the smaller foot print, the technical demand and cost of VSSF systems are relatively higher than that of HSSF (Yin et al., 2008). 3.3. Hybrid wetland systems As wastewater from various sources are generally difficult to treat in a single-stage wetland system, hybrid wetland systems which consist of various types of natural systems staged in series

have gained increasing interest in Europe (Vymazal, 2005). For example, single-stage constructed wetlands cannot achieve high removal of total nitrogen due to their inability to provide both aerobic and anaerobic conditions at the same time. While there may be other better alternatives, combining ponds and vertical flow constructed wetlands, as well as infiltration percolation and horizontal flow CWs have proven to be effective (Brissaud, 2007). In addition, CWs systems combining horizontal- and vertical flow were shown to be more efficient than non-hybrid system, and various types of constructed wetlands maybe combined to complement each other and to achieve higher treatment effect, especially for nitrogen (Vymazal, 2005; Lüderitz et al., 2001; Brissaud, 2007). There has been good experience on application of hybrid systems in China. Table 4 presents several applications of hybrid systems in China. Zhai et al. (2006) reported a new type of hybrid constructed wetland: an innovative design of vertical-baffled flow wetland and horizontal subsurface flow wetland (HSFW) has been introduced in Chongqing University (CQU), China. The experimental results exhibited a dramatic reduction in the land requirement and the system was found suitable for waste treatment for a small township. In his report the author indicated that high hydraulic retention time (HRT) and internal circulation had very positive effect on pollutant removal – the removal rate of TN could double through an internal circulation, with a flow rate that 1–2 times of the influent. The highest pollutants removal rate of the hybrid CW with internal circulation occurred HRT of 70 h. Another successful example is Shatian hybrid CW which consists of two-stage-SSF systems. According to Shi et al. (2004), first-stage wetland designed in horizontal flow pattern and the total area of wetland is 4800 m2 with bed depth of 1m and HRT of 11.5 h. The second-stage wetland takes the form of verticaldownwards flow, a total of 4 trains arranged in parallel. The total

1370

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

Table 2 A summary of the treatment efficiency of HSSF systems in China. TSS

BOD5

Rongcheng, Shangdong Provincea Effluent value (mg/l) Removal efficiency (%)

27.8 71.8

23.8 70.4

91 62.2

11.3 40.6

Dongying, Shandong Provinceb Effluent value (mg/l) Removal efficiency (%)

8.53 88.2

4.61 90

41.6 75.8

7.12 67.31

Jiaonan, Shandong Provincec Effluent value (mg/l) Removal efficiency (%)

30 57.1

11 66.7

Miyun, Beijingd Effluent value (mg/l) Removal efficiency (%)

– –

– –

Futian, Shenzhen Provincee Effluent value (mg/l) Removal efficiency (%)

– –

CIW-TS, Tianjinf Effluent value (mg/l) Removal efficiency (%)

13 84.9

Taihu, Zhejiang Provinceg Effluent value (mg/l) Removal efficiency (%)

– –

– –

Baptist University, Hong Kongh Effluent value (mg/l) Removal efficiency (%)

– –

– –

Average efficiency in China (%)

75.5

a b c d e f g h

COD

NH4 -N

Hydraulic loading rate (m3 /day)

Hydraulic retention time (day)

– –

2 29.6

20,000

120

– –

0.86 59.23

50,000

21.6

2.98 –

30,000



– –

63.8 11.1

50 95.1

63 87.1

61.8 85.3

17.6 73.6

270

72

8.37 90

25.31 70

6.28 50

8.27 46

0.65 60

5

72

9.04 94.01

44 87.02

4.49 80.13

5.7 80.04

0.25 72.96





4.23 39.6

1.16 32

2.29 52.1

0.052 65.7

0.64 m/day



– 95

– 62

– 52



120

64.59

56.09

59.01





82.22

125 60.9

TP

TN

– – 70.09

Song et al. (2006). Wang et al. (2005b). Song et al. (2008). Wang et al. (2008). Yang et al. (2008). Yin and Shen (1995). Li et al. (2008). Chung et al. (2008).

surface area of the secondary stage wetland is 4640 m2 with bed depth of 0.75 m and HRT of 8 h. A total of 7 species of plants have been chosen for this wetland system. And the average removal efficiency of TSS, BOD5 , COD, TN and TP are 86.78%, 86.4%, 76.72%, 44.93%, and 81.7%, respectively. Wang et al. (1994) investigated a hybrid system for industrial wastewater at Yantian industry area in Shenzhen City, Guangdong Province, which consists of anaerobic lagoon and three water hyacinth ponds and two HSSF beds planted with Phragmites australis. Despite the very high hydraulic loading (36 cm/day for the HSSF stage), the authors reported that the removal efficiencies of TSS (99%), COD (81%), BOD5 (69%), and TP (62%) were very good and steady in hybrid systems, but the removal efficiency of TN was relatively low. Recently, hybrid constructed wetlands comprise more than two types of CWs and quite often include a FWS stage in Europe (Vymazal, 2005). In China, Liu et al. (2007) investigated the water quality variation in a hybrid CWs in purifying the Yongding River, Beijing. There were altogether 7 parallel-connected wetland units, and the influents flowed from the emerging plant pond (1. Surface flow); the first-stage plant-gravel bed (2. Subsurface flow); the floating plant pond (3. Surface flow); the second-stage plant-gravel bed (4. Subsurface flow); the sand-filtration tank (5. Subsurface flow). The removal ratios of the main pollutants in Yongding River by this hybrid system were TSS (99.1%), COD (52.8%), BOD5 (77.0%), TN (59.4%), NH4 -N (52.8%), NO3 -N (60.3%), PO4-P (92.7%), respectively. The purification effect was remarkable.

3.4. Design and performance In sum, when comparing FWS, HSSF, VSSF and hybrid systems in China (see Tables 1–4), experience has showed that hybrid systems perform best in the removal of TSS (94.96%), COD (78.52%), and TP (79.68%). Compared to VSSF systems, HSSF systems showed better removal efficiency for TSS and COD (75.5% and 70.09%, respectively). As for nitrogen removal, the TN removal efficiency of HSSF systems was significantly higher than that for VSSF systems. Apparently, despite experience in China pointing to the superior oxygen supply of VSSF systems, the removal rate of TN (43.66%) remains lower in comparison of that in HSSF systems (56.09%), probably due to the lack of carbon source during denitrification. However, surprisingly, even the ammonia removal efficiency of HSSF systems in China (64.59%) was higher than for the VSSF systems (56.17%), indicating that at least in these systems reviewed here, that the HSSF systems are probably better at nitrification that are the VSSF systems in China. From Chinese CW experience (see Tables 1–4), a comparison of removal efficiencies by FWS, HSSF, VSSF and hybrid systems can be made between China and Europe, the latter for which Haberl et al. (1995) reported on the efficiencies of 268 wetlands in operation. The mean BOD5 removal efficiencies were 66.13%, 82.22%, 82.95%, and 80.10% for FWS, HSSF, VSSF, and hybrid systems, respectively in China. In comparison with the value in Europe (79.1%) (Haberl et al., 1995), most systems in China seem to perform in the same range as those in Europe.

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

1371

Table 3 A summary of the treatment efficiency of VSSF systems in China. TSS

BOD5

COD

NH4 -N

TN

TP

Hydraulic loading rate (m3 /day)

Hydraulic retention time (day)

Bainikeng, Shenzhen, Guangdong Provincea Effluent value (mg/l) 10.9 Removal efficiency (%) 92.6

6.9 90.5

38.3 73.5

18.5 10.5

18.5 10.6

1.59 30.6

3100



Longdao, Beijingb Effluent value (mg/l) Removal efficiency (%)

7.08 85.1

6 87.2

19.7 81.8

5.2 77.4

– –

0.061 98.8

2000



CIW-TS, Tianjinc Effluent value (mg/l) Removal efficiency (%)

48 44

69.4 54

15.3 32

23.6 30

1.09 44





4.25 40.4

0.89 45.9

2.37 51.6

0.056 64.3

0.64 m/d



Laboratory, HongKong; Pilot Project, Guangzhoue Effluent value (mg/l) – 8.37 Removal efficiency (%) – 90

25.31 70

6.28 50

8.27 46

0.65 60

0.45 m3 /(m2 day)

Jinan, Shandong Provincef Effluent value (mg/l) Removal efficiency (%)

– –

19.5 90.05

1.3 94.8

38.15 26.66

0.25 92.25

1000



Wuxi, Zhejiang Provinceg Effluent value (mg/l) Removal efficiency (%)

96 77.1

61.8 81.3

– –

32.9 61.7

41.3 66.6

– 48.9

0.4



Jinhe River, Tianjinh Effluent value (mg/l) Removal efficiency (%)

– –

– –

68.9 35

1.65 71.25

2.58 64.85

0.2 61.24

0.8 m/day



Chongming, Shanghaii Effluent value (mg/l) Removal efficiency (%)

– –

– –

13.8 67

2.7 62

3.7 53

1.9 33





62.09

56.17

43.66

59.23





Taihu, Zhejiang Provinced Effluent value (mg/l) Removal efficiency (%)

Average efficiency in China (%) a b c d e f g h i

– –

74.7

– –

2.66 94.68

82.95

207 39

18

Yang et al. (1995). Chen et al. (2008). Yin and Shen (1995). Li et al. (2008). Chan et al. (2008). Yin et al. (2008). He et al. (2006). Tang et al. (2009). Wang et al. (2006b).

The mean NH4 removal efficiencies were 64.59%, 56.17%, and 37.37% for HSSF, VSSF, and hybrid systems, respectively in China. And the mean TN removal efficiencies were 49.11%, 56.09%, 43.66%, and 46.76% for FWS, HSSF, VSSF, and hybrid systems, respectively in China. In Europe, the average NH4 removal rate was 30%, and the average of TN removal rate is 39.6% (Haberl et al., 1995). Apparently, both of the removal rates for NH4 -N and TN in these CW systems in China are higher than that in Europe. The mean phosphorus removal efficiencies were 53.15%, 59.01%, 59.23%, and 79.68% for FWS, HSSF, VSSF, and hybrid systems, respectively in China. In comparison with the value reported for Europe (47.1%) (Haberl et al., 1995), the average removal efficiencies of TP in China are generally higher than that for Europe. 4. The role of the plant in constructed wetland treatment There has been some debate on the importance of plants in pollutant removal by constructed treatment systems (Wu et al., 2008). Experience has shown that a wetland system with vegetation has a higher efficiency of pollutant removal than that without plants, and the significance of the plants used for wastewater purification has been emphasized by previous researchers (Brix, 1997; Peterson and Teal, 1995; Gersberg et al., 1983). However, the quantitative

role that the plant plays in wastewater purification is still a subject of some debate. The removal capabilities of a well-developed vegetation could be explained by: (i) the rhizosphere connected to a plant with active oxygenic photosynthesis will allow the transfer of a certain amount of oxygen to the vicinity of the roots; (ii) in the root system, where there exists a large number of bacteria whose oxidation–reduction potential and nitrification rate are both higher than the area without plants, and each plant root system is regarded as a mini aerobic/anoxic biological treatment system, and (iii) uptake into the plants. Generally, the amounts of nutrients removed by vegetation harvesting are insignificant compared to the load brought into the system with wastewater (Brix, 1994; Merlin et al., 2002). If the wetland vegetation is not harvested, most of nutrients could be temporarily stored in the litter compartment. According to Verhoeven and Meuleman (1999), during the autumn and winter, a large part of nutrients will be gradually released again through leaching and organic matter mineralization. Only a small part of the nutrients stays in the vegetation as additional long-term storage in aggrading wood or rhizome material. Vymazal (2005) also concluded that the removal of nitrogen and phosphorus through plant harvesting is negligible and forms only a small fraction of the removed amount.

1372

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

Table 4 A summary of the treatment efficiency of hybrid systems in China. TSS Chongqing University, Sichuan Provincea Effluent value (mg/l) – Removal efficiency (%) –

BOD5

– –

COD

NH4 -N

TN

TP

Hydraulic loading rate (m3 /day)

Hydraulic retention time (day)

51.5 78.5

17.8 42.3

20.1 51.7

1.12 65.9

1555 m3 /(m2 day)

70

9.11 44.93

0.56 81.7

5300

11.5 (Stage 1) 8 (Stage 2)

6.38 59.4

0.1 91.8

0.58 m3 /(m2 day)

34.26

– 97



72

Shatian, Shenzhen, Guangdong Provinceb Effluent value (mg/l) 7.92 Removal efficiency (%) 86.78

7.68 86.4

33.9 76.72

Yongding River, Beijingc Effluent value (mg/l) Removal efficiency (%)

12.3 99.1

5.91 77

5.47 67.4

Guangdong Provinced Effluent value (mg/l) Removal efficiency (%)

– –

– 88

– 89

– –

– –

Yantian, Shenzhen, Guangdonge Effluent value (mg/l) Removal efficiency (%)

3.2 99

58 69

88 81

12.2 17

15.5 31

1.8 62

36 cm/day



Average efficiency in China (%)

94.96

80.10

78.52

37.37

46.76

79.68





a b c d e

– – 4.27 52.8

Zhai et al. (2006). Shi et al. (2004). Liu et al. (2007). Cui et al. (2006). Wang et al. (1994).

4.1. The capacity of plants for supplying oxygen The role of plants in supplying oxygen is still being debated. In CW systems, an important function of macrophytes is to transport oxygen and release them from its root system into the wetland, influencing the biochemical cycles in the substrate, and supplying oxygen to bacteria growing on plant roots to improve the decomposition of organic matter and convert ammonium to nitrate (Gersberg et al., 1986; Barko et al., 1991; U.S. EPA, 2000). However, such capacity of oxygen transfer is limited. Brix et al. (1996) found a negligible oxygen input of 0.02 g/(m2 day). And Zhu and Silora (1994) pointed out that no obvious nitrification could be observed when dissolved oxygen concentration is lower than 0.5 mg/l. Furthermore, in anaerobic soils, oxygen is transferred to the roots primarily for plant respiration and only excess oxygen is leaked to the micro-zone at the rhizosphere (Brix, 1990). In China, the data on the ability of plant in translocation oxygen is also rare. Yin et al. (2004) reported that the ability of oxygen translocation are 0.02–12 g/(m2 day), 0.5–5.2 g/(m2 day) and 0.25–9.6 g/(m2 day) by reed, submerged- and floating plant respectively, these values showing the wide range of translocation abilities of these aquatic plants. However, despite this fact, in research on nitrogen removal and microorganism in a SSF system in Sihong County, Xia et al. (2006) reported that compared with the removal of COD and BOD5 , the nitrifying process was slow. Because oxygen is mainly used for removal of organic matter and the nitrifying reaction begins only if BOD5 is reduced to a significant extent, the small amount of oxygen (0.2–0.4 mg/l) in this CW system limited the activity of the nitrifier Nitrosomonas, which limited any further removal of nitrogen (by sequential nitrification–denitrification) from CWs. 4.2. Role of the plant in nitrogen removal The nitrogen removal mechanisms in wetland systems are very complex. The processes that affect nitrogen removal during wastewater treatment in CWs are manifold. Basically, it includes NH3 volatilization, nitrification, denitrification, nitrogen fixation, plant and microbial uptake, mineralization (ammonifi-

cation), nitrate reduction to ammonium (nitrate-ammonification), fragmentation, sorption, desorption and leaching (Vymazal, 2006). Of the many kinds of removal mechanisms, however, only a small subset of these processes ultimately play an important role in total nitrogen removal while most processes just convert nitrogen to its various forms. Two major processes have been identified and they are: (i) storage, which is achieved by assimilation into biomass (e.g., plant and microbial uptake) or adsorption to the substrate (e.g., soil); and (ii) removal through the N cycle: nitrification and denitrification (Jamieson et al., 2003). In a FWS in Dianchi Valley (Kunming City), Lu et al. (2009) investigating the N distribution pathway and removal efficiency, concluded that plants were important for the wetland, as the plants provided good growth conditions for microbes, which removed the majority of N from the CWs. Over a 5-year period, the wetland received slightly more than 2000 kg/ha of nitrogen, mostly from farmland drainage. The nitrogen removal was mostly due to plant uptake (1110 kg/ha) and soil accumulation (570 kg/ha), with the contribution of denitrification being estimated at around 7%. The authors concluded that this was because Zizania caduciflora and Ph. agmites had large biomass and thus had good N and P absorption capability. However, much experience at higher hydraulic application rates (high nitrogen loading rates), show that the processes of sequential nitrification–denitrification play an increasingly major role as compared to plant uptake (Gersberg et al., 1986). For example, in a controlled comparison of ammonia-N removal efficiencies in vegetated vs. unvegetated contracted wetland beds, Gersberg et al. (1986) showed that the presence of plants did indeed make a significant (p < 0.05) difference in removal efficiency (although not via N-incorporation into plant biomass. These authors found that ammonia removal efficiencies were 28–94% for various types of wetland plants versus only 11% for unvegetated wetlands. They concluded that clearly sequential nitrification–denitrification was responsible for the higher rates of ammonia removal in these planted wetlands. Similarly, Tang et al. (2009) in the study of seven experimental pilot-scale VSSF systems in Tianjing, also reported that, with respect to NH4 -N removal, the planted wetland showed higher

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

removal performances than that of the unplanted wetlands. The improvement was significant and accounted for a 17.18% increase as compared to that in unplanted wetland in NH4 -N mean removal efficiency (p < 0.05). The authors also indicated that insufficient microbial activity in unplanted wetland substrate is likely to limit NH4 -N removal. Meanwhile, the planted wetland showed a better TN removal than the unplanted wetland, and the presence of Typha latifolia in a significant (p < 0.05) additional TN removal of 21.78%. This observation verified that wetland plants can make significant contribution to TN removal. In a study of the HSSF system at Baptist University in Hong Kong, Chung et al. (2008) indicated that plant uptake only removed 2.6–3.1% N in the microcosms, and denitrification was the main removal pathway. Loss of N through denitrification was 34 and 50% in 10-day HRT and 5-day HRT unplanted treatment respectively. In planted treatment, loss of N through denitrification was 20% and 32% in 10-day HRT and 5-day HRT treatment. The average removal efficiencies in NH4 -N were 95–97% and 92–94% planted and unplanted treatment, respectively. Meanwhile, the average removal efficiencies in TKN were 63–66% and 40–52% for planted and unplanted treatment, respectively, and removal of TKN was comparatively lower in unplanted treatments, because anoxic conditions were found in unplanted treatments due to ponding of wastewater, and this limited the rate of removal. In this study, loss of N through denitrification was 19–41% of total N input for all treatment. Additionally, the plants could reduce total N to significantly lower levels than unplanted treatments. The high removal rate of NH4 -N in planted treatments showed that nitrification was very active but the high NH4 -N removal was enlarged by the increased rate of evapotranspiration during plant growth. Wu et al. (2008) investigated the capabilities of mangrove SSF microcosms in treating primary settled municipal wastewater collected from a local sewage treatment work in Hong Kong. The removal efficiencies in the planted systems were 76.16–91.83% for ammonium-N, 47.89–63.37% for inorganic-N, and 75.15–79.06% for total Kjeldahl nitrogen. And the authors also indicated that for total nitrogen, the planted system had significantly higher removal (55.56–83.33%) than the unplanted treatments (22.22–33.33%). Nitrification and denitrification process are believed to be an important mechanism for nitrogen removal, and in this study, decreases of ammonia in effluent were followed by increases in nitrate. Additionally, the planted systems had significantly lower effluent nitrate concentrations than the unplanted ones. The authors concluded that apparently the mangrove plants not only absorb nitrate for their growth, but they also enhance the efficiency of both nitrification and denitrification processes. Furthermore, even though the plants could take up nitrate in the soil pore water and their roots could provide a large surface area for microbial growth, in this study, the amount of nitrogen accumulated in the plants was only 2.88–3.28% of total nitrogen. 4.3. Role of the plant in phosphorus removal Jiang et al. (2008) investigated a two-stage SSF (combining with HSSF and VSSF) in Longgang district of Shenzhen city (Guangdong Province), and reported that in the first-stage CW, results of TP concentration indicated that this was superior for TP removal, and the main area for TP accumulation. In the 15 cm and 36 cm depth layer, 87.92% and 86.24% of TP was accumulated in the superior half part, and this was totally in agreement with the removal situation of wastewater. The authors further examined P transfer into different plants and reported in the first-stage of the CW, the order of TP concentration in organs of Ph. australis Trin, Miscanthus sacchariflorus, Thalia dealbata was flowers, roots, leaves and caudex, while that of

1373

Scirpus validus was roots, caudex and flowers. In the second-stage CW of Canna generalis and Cyperus papyrus, phosphorus transfer was coincident with phosphorus distribution in plants in the order of seeds, followed by leaves, roots and caudex. The unregularity of phosphorus transfer in the CW system demonstrated that it was greatly influenced by plant growth environment and species. Similarly, Chung et al. (2008) reported that removal of PO4 P was at least twofold higher in the planted treatment and the presence of plants could effectively remove PO4 -P because it is readily available for plant uptake. Low removal of TP in unplanted treatments was expected, removal was sixfold lower than planted treatments. The authors also indicated that vegetation, detritus, fauna and microorganisms are an important sink for P in the short term, but substrate is the main sink for P in long term. However, in the long term, TP removal will be decreased in the vegetated treatment due to the saturation of P adsorption in the substrate. In a mass balance of P, the uptake of P by plants was only 1% if the total amount of P, the presence of plants has increased the P removal rate and improved the treatment efficiencies. Despite the finding above, since phosphorus removal is not mediated by a microbial transformation process (as in the case with nitrogen), plants would not be expected to play a major role in phosphorus removal at higher hydraulic application rates. Indeed, Wang et al. (2005a) reported that even in the growing season, the vegetation did not show significant uptake capacity for phosphate removal, and the phosphate was taken up by vegetation roots mostly from the sediments. In a similar study, Yang et al. (2008) investigated the treatment efficiency in a pilot-scale mangrove wetland in Futian, Shenzhen for municipal sewage treatment and the removal efficiency data indicated that plant growth had played a minor role in phosphate removal which was confirmed by an insignificant correlation between phosphate removal and the increase in plant height. 4.4. Role of plant in COD, BOD5 and TSS removal The removal COD and BOD5 rely largely on the good combination between physical and microbial mechanisms. Due to a physical separation mechanism and low porosity of the soil media, the organic solids could be filtered and trapped in the bed of CWs for a long time, thereby allowing for better biodegradation of organic solids. The high removal rates for COD and BOD5 are caused by sedimentation of SS and by rapid decomposition processes in the water and upper soil layers. Organic matter is consumed and reduced by bacteria and other microbes both aerobically and anaerobically (U.S. EPA, 1993). Yang et al. (2007) presented a comparative study of the efficiency of contaminant removal between several emergent plants species and between vegetated and unvegetated wetlands conducted in Shenzhen, Guangdong Province for domestic wastewater treatment. The authors reported that there were no significant differences in the removal of organic matter between vegetated and unvegetated wetlands. Similarly, Tang et al. (2009) investigated and assessed the effect of plants [T. latifolia L. (cattail)] through severs experimental pilotscale SSF CWs in Tianjin. A statistical analysis indicated that there was not a significant difference in COD removal rate between the planted wetland and the unplanted wetland, and the presence of T. latifolia only led to an insignificant (p < 0.05) increase of 2.94% with respect to the mean COD removal efficiency. Therefore, plants played a negligible role in chemical oxygen demand (COD) removal. Apparently, despite the fact that BOD5 and COD removal in CWs are mediated through biological degradation of the organic matter, it would appear that in most wetland systems either anaerobic decomposition plays a major role, or alternatively, aeration of the

1374

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

substrate along (without plants) is sufficient oxygen demand of organic removal. 4.5. Role of plant species on removal efficiency In the report of investigation on twelve small gravel-based SSF CWs systems and larger SSF CWs systems were installed at the Virginia Tech’s Kentland Research Farm and at the Powell River Projects, Huang et al. (2000) indicated that plant species had no impact on TKN or NH4 -N concentrations in the wetland effluent or removal of these N species from the wetland. However, other researchers indicated that the removal efficiency of pollutant is varied by the plant species (Gersberg et al., 1986; Peterson and Teal, 1995). In China, many studies revealed that different wetland systems performed differently with plant species and productivity varied. Yang et al. (2007) concluded that there was a significant difference in the removal of total nitrogen (TN) and total phosphorus (TP). Wetlands plants with Canna indica Linn., Pennisetum purpureum Schum., and Phragmites communis Trin. had generally higher removal rate for TN and TP than wetlands planted with other species. The authors also indicated that fine root (root diameter ≤3) biomass rather than the mass of the entire root system played an import role. Moreover, removal efficiency varied with season and plant growth, e.g., wetlands vegetated by P. purpureum significantly outperformed wetlands with other plants in May and June, whereas wetlands vegetated by Ph. communis and C. indica demonstrated higher removal efficiency from August to December. In a similar study, Yang et al. (1995) investigated a CW system at Bainikeng, Shenzhen and indicated that different plant species resulted in great differences in removing efficiency: effluent BOD5 was 17.1 mg/l for Cyperus malaccensis, 18.2 mg/l for Ph. communis (sampling at secondary gravel bed); and 5.3 mg/l for Cyperus malaccensis, and 7.78 mg/l for Lepironia articata (sampling at the fourth gravel bed). The authors concluded that C. malaccensis is the most efficient one at removing BOD5 while L. articata is the least. Even using same plant type, the treatment efficiency varies largely by species. Yang et al. (2008) invested the treatment efficiency in a pilot-scale mangrove wetland in Futian, Shenzhen for municipal sewage treatment and also indicated that although 70% of the organic matter, 50% of TN, 60% of NH3 -N, 60% of the TP, and 90% of the coliforms were removed, Sonneratia caseolaris was the most efficient one with all the effluent samples below the discharge standards for COD, BOD5 , and NH3 -N, whereas the percentage of samples meeting the discharge standards varied from 71.43% to 85.71% for A. corniculatum and Kandelia candel. The removal of TP was the lowest among the nutrients with 42.86% (K. candel) to 74.43% (A. corniculatum) of the samples meeting the discharge standards. Investigating the growth vitality and their removal ability of the pollutants in domestic sewage, nine aquatic plant species commonly used in northern China and transplanted in a HSSF CWs in Beijing region, Wang et al. (2008) reported that among the tested plant species, Iris pseudacorus (with the capacity of high N & P removal efficiency) ranked first in setting up the constructed wetland, followed by Typha angustifolia, Acorus calamus, and Triarrhena sacchariflora, whereas Alisma plantago and Arundo donax were not recommended due to their sensitivity in cold winter in northern China. 5. Climate effects Treatment performance in constructed wetlands may be less consistent than in conventional treatment since they are strongly

influenced by climate and weather (U.S. EPA, 2000). The best prospects for successful wetland treatment should be in the warmer regions. However, in cold weather, wetlands continue to function, but rates of microbial decomposition may be slow if the wetland either freezes solid or under a cover of ice. Maehlum et al. (1995) and Jenssen et al. (1996) stated that nitrogen cycling was inhibited in colder months due to the decrease of oxygen availability. Besides lower winter temperatures, low oxygen availability which is already a common limiting factor in FSSF systems during the growing season, may be even more severe in winter. Similar results were also obtained by Maehlum and Stalnacke (1999), in which they found that the differences in efficiency between cold and warm periods were less than 10% for all parameters, and the temperature effects were partially compensated for by longer hydraulic retention time. Wang et al. (2006a,b) reported that the removal efficiency of ammonia nitrogen in October (71.6%) was much higher than that in May (32.9%), although the water pH and temperature, which are the most important factors affecting the volatilization rate of NH3 -N, in May were similar to that in October. That means the volatilization was not a major removal mechanism for ammonia nitrogen. Song et al. (2006) investigated the seasonal and annual performance of a full-scaled CW in Rongcheng, Shandong Province. He concluded that there was a significantly seasonal component to this wetland for BOD5 , COD, ammonia nitrogen and total phosphorus, when measured on a percentage reduction basis: and (i) the mean BOD5 and COD percent reduction were approximately 10% less efficient in winter compared to spring and summer, as physical processes such as sedimentation are important in organic matter removal and are unaffected by winter conditions; (ii) ammonia nitrogen removal was about 40% less efficient in winter than in summer and was associated with an increase in temperature and plant growth; and (iii) there was less variability in seasonal phosphorus removal (around 20% less efficient winter) when compared to ammonia nitrogen, due to sedimentary binding of phosphorus. Peng et al. (2005) reported for a multi-stage pond–wetlands ecosystem located in Dongying, Shandong Province, that in cold season the removal efficiency of BOD5 , COD, and NH3 -N was about 84.5%, 40%, 19.6% respectively, whereas in warm season, that increased to 91.8%, 73% and 71.4% respectively. Yin and Shen (1995) reported that a CW with reed beds for industrial and municipal wastewater treatment located in Tanjin, North China, could successfully operate under ice layers when the average temperature was lower than −4 ◦ C and the lowest temperature ranges from 21.2 ◦ C to 26.3 ◦ C. And effluent quality are 9.04 mg/l, 13 mg/l, 5.5 mg/l and 0.25 mg/l for BOD5 , SS, TN and TP, respectively, which are better than secondary treatment level, e.g., BOD5 < 20 mg/l, SS < 20 mg/l, TN < 15 mg/l, TP < 0.6 mg/l. The high removal rate of BOD5 indicated that soil microbes in winter still have the capacity to decompose organic contaminants. Also, reeds in winter play an important role although they stop growing. They give the ice layer a strong support and partly transfer oxygen from air to water. Zhang et al. (2006) studied the treatment performance of an SSF system treating polluted river water in Shandong Province and reported that water temperature had great influence on ammonium nitrogen removal and plastic film mulch on the wetland could improve pollutant removal efficiently. In his investigation, the average removal rates of ammonium nitrogen and COD could rise up from 29.4% and 29% to 67.6% and 46.6% respectively. Microorganism enzyme activity experiments showed that increase of microorganism activity caused by the overlay of plastic film mulch contributed very much to pollutant removal enhancement. Wang

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

1375

Table 5 A comparison of the cost of a conventional wastewater treatment processes and CW system.

Conventional WWTP in Chinaa Conventional activated sludge process in Chinab Constructed wetland in Chinab Dongying, Shangdong Provincec Longdao River, Beijing Cityd Dagong Oil Field, Tianjin Citye Wei Fang, Shangdong Provincee Hong Kongb a b c d e

Design capacity (m3 /day)

Total capital cost (US$)

Unit capital cost (US$/m3 )

Treatment cost (US$/m3 )

O/M cost (US$/m3 )

– – – 100,000 200 2000 180,000 0.45 m3 /(m2 day)

– – – 8.2 million 29,191 41,176 – –

220 115 28.82 82 146 20 102 37.64

0.15 – – – 0.03 0.025 0.021 –

0.13 0.116 0.022 0.012 0.014 0.031 – 0.019

Li and Wang (2006). Chan et al. (2008). Wang et al. (2005a). Chen et al. (2008). Li and Jiang (1995), Yin and Shen (1995).

et al. (2005a) reported that the BOD5 removal rate ranged from 75.6% to 90.7% in winter and 85.5% to 83.0% in summer respectively, with effluent BOD5 of 3.29–16.7 mg/l in winter and 1.50–5.91 mg/l in summer. Similarly, Yang et al. (2007) also indicated that the concentration of pollutant in the effluent was significantly higher in October and December compared to summer. During these months, the mangrove plants had slower growth and the microbial activities were also lower due to the low temperature. However, Lu et al. (2009) reported that the removal rate of N in winter was not far lower than in other seasons and the contaminant removal rate of the CWs had less than 10% difference between the warm and cold period. The authors concluded that the good performance of the CWs during winter was mainly due to three reasons: (i) the initial harvest prevented N release caused by the decomposition of plant matter and strengthened oxygen diffusion from the atmosphere; (ii) the free water surface CWs was built in Kunming City, which is in the north-subtropical zone, and the average water temperature in winter was higher than the minimum required temperatures of nitrification and denitrification; and (iii) the intermittent inflow was beneficial to the processes of nitrification and denitrification. 6. Cost/energy/land requirements and limitations 6.1. Cost In China, wastewater in most small- and medium cities as well as rural areas has not been properly treated, because of the invariability of wastewater treatment facilities. The use of CWs system for the treatment of polluted water has attracted increasing attention in the last decades due to its minimal costs for construction, operation and maintenance. Table 5 compares the investment and operation cost for a traditional wastewater treatment plant (WWTP) and CWs in China. Although conventional WWTP and activated sludge processes are efficient for wastewater treatment, their cost-effectiveness can only be achieved in densely populated urban areas. In contrast, the application of CW is more affordable for the wastewater treatment demands of small communities. Table 5 shows that although CW systems do not present apparent advantage in construction cost, the treatment and O/M cost of CW systems is much lower than that of conventional WWTP and activated sludge processes. Wang et al. (2006a) reported that the total capital cost of an ecosystem consisting of integrated ponds and constructed wetland system (located in Dongying City, Shandong Province) was US$ 82/(m3 day), which is about half of the conventional systems based on activated sludge process. The O/M cost is US$ 0.012/m3 ,

only one fifth that of conventional treatment systems. Li and Jiang (1995) reported that the capital investment and operation cost of a large-scale reed bed FWS in Weifang City, Shandong Province were 35% and 14% of that of A2O (anaerobic/anoxic/oxic) treatment systems. Similarly, the investment cost for a CWs with reed beds for wastewater treatment in Tianjin, North China was summed up to US$ 20/(m3 day) and the operation cost was US$ 0.025/(m3 day) (Yin and Shen, 1995). Chen et al. (2008) also reported that with the treatment capacity of 200 m3 /day, the construction cost in the Longdao River CW (located in Beijing) was calculated to be US$ 0.02/m3 , and average treatment cost was summed up to US$ 0.03/m3 , which is equal one-fifth of that in traditional WWTP. 6.2. Energy Constructed wetlands are an attractive and promising alternative (both for industrialized and developing countries) to conventional technologies to treat wastewater due to their low energy consumption. Lüderitz et al. (2001) compared three different strategies for wastewater treatment and disposal for three villages, namely discharge to a large-scale sewage treatment plant 20 km away, construction of a central mechanical–biological wastewater system for the three villages, or construction of a wetland for every local community. The authors concluded that a semi-centralized constructed wetland needed 83% less energy than that of a central technical system and 72% less energy than the discharge to a central treatment located 20 km away. In addition, in the case of energy, the advantage of the CW in operational efficiency dominates. In China, data and reports on energy consumption for constructed wetland are very rare. From 1985 to 1990, National Planning Committee, National Science and Technology Committee and National Environmental Protection Agency organized a nationwide study on sewage wetland treatment systems. The research projects were installed in different climatic zones—Northwest: Xinjiang Autonomous Region (arid area, the north temperate zone); Northeast: Shenyang City (the north temperate zone); North China: Beijing and Tianin (the medium temperate zone) and Southwest: Kunming City (the north-subtropical zone). At the treated wastewater included municipal sewage, paper industry effluent, petrochemical processing wastewater and beer brewery effluent. And the treated sewage capacity ranges from 120 m3 /day to 500 m3 /day, the resultant technical-economic comparative analysis indicated that energy consumption for the different CW systems was only 15–25% of that of conventional activated sludge technology (Li and Jiang, 1995).

1376

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

6.3. Space and land requirements CWs systems for wastewater treatment are usually land intensive and may require more space than conventional wastewater treatment systems (Kivaisi, 2001; Wittgren and Maehlum, 1997; Brissaud, 2007). The high land requirement for CWs is the main barrier for expanding the application of CWs in China. Also, CW cannot be applied in densely populated areas where land prices are often too high. The land requirements of CWs for wastewater treatment vary widely. China’s distinct problem is that 81% of its water resources are in the country’s southern part but the largest part of arable land (64%), is in the north, where the nation’s political and economic centre is located (Varis and Vakkilainen, 2006). Since there is considerable diversity of geography, climate, land- and water resources distribution between northern- and southern China, the availability of land use for CW construction varies correspondingly. Therefore, although Zhai et al. (2006) indicated that the land requirement of traditional CWs for wastewater treatment is from 10 m2 /(m3 day) to 70 m2 /(m3 day), experience of CW construction in southern China indicated that the land requirement is much smaller than that in the northern region. For instance, according to Li and Wang (2006), in the constructed wetland systems of Shatian (Shenzhen, Guangdong Province), the land requirement was approximately 1.88 m2 /m3 and 1.2 m2 /capita, in the case of assumed consumption of 300 l/(capita day). Additionally, with the consideration of preliminary treatment, the total land requirement is 4 m2 /m3 . Also, in the constructed wetland of Bainikeng (Shenzhen, Guangdong Province) the land requirement for wastewater treatment was around 2.7 m2 /m3 . Southern China belongs to a subtropical climate zone, with relatively high temperatures and a humid climate, which is favourable for water plant’s growth. However, in Southern China, the average population density is 210 persons/km2 (Zhai et al., 2006). In this region, land resources are scarce with a high population density. Furthermore, the price of land is so high that land cost forms a high percentage of total investment for CWs. Therefore, CW treatment may be economical relative to other options only where land is available and affordable. As the available land possessed per capita in China is much lower than that of international standard, wetland systems with small land requirement and good effluent performance are more suitable for application. Great effort should be made therefore towards improving the treatment efficiency of CWs and decreasing the land requirement.

Emergy analysis at the scale of biosphere and society is an evaluation system free of human bias, which can represent both the environmental and economic values of a given system (Odum, 1988; Brown and Ulgiati, 1999). Emergy accounting as an ecological approach came out of creative combination of thermodynamics and systems ecology (Odum, 1996; Geber and Bjoerklund, 2001), and this approach represents a measure for comparison of environmental good, energy quality, and economic valuation (Odum, 1988; Brown and Herendeen, 1996). In China, Zhou et al. (2009) measured the energy and resource consumption and conducted a comparative study on a constructed wetland (Longdao River, Beijing) and conventional wastewater treatment with cyclic activated sludge system (CAAS) (Hangtiancheng, Beijing). In this study, emergy-based indices such as the ratio of purchased/fee, local/imported and the ratio of electricity emergy used were chosen to characterize the two treatment systems in self-sufficiency and environmental effect, respectively. The report revealed that the ratio of purchased inputs to free inputs for CWs were 3.4, compared with the ratio of 1450 for CAAS. The ratio of local inputs to imported for CWs was 0.35, almost three times more than those for CASS, revealing that the system of CASS depended more on external resources and were driven mainly by the imported emergy. Similarly, the ratio of the electricity consumption for CWs is 3.9% while 37.1% for CASS. This indicates that more local renewable resources and less ecological cost are involved, thus promoting the economic benefit due to less energy consumption and the lowering of environmental stress. Zuo et al. (2004) initialled a comparative study on the sustainability of original and constructed wetlands in Yancheng Biosphere Reserve (YBR) located in Jiangsu Province. The authors employed two new emergy indices, base emergy change (Bec) and the net profit (Np) in order to compare the ecological-economic benefits of different kinds of wetlands. Results indicated that a water fowl pond, constructed for ecological reasons at the edge of the core zone of YBR, has much more Bec than the original wetlands and fishponds, while its Np is negative and much lower than the other sites. Fishponds built for economic reasons in the buffer zone have negative Bec while the Np is the highest. However, the emergy yield ratio (Yr) of the original wetlands is the highest. In some way, it could be said that the negative Bec in the fishponds may mean a purely exploitation activity searching for economic benefits by exhausting natural resources, and fishpond creation should be stopped to ensure better conservation of the original wetlands and rare bird species. The positive Bec and negative Np of the waterfowl pond indicated an effective way forward for biodiversity conservation, which was proved by the increasing numbers of birds and bird species observed.

7. Sustainability In 1987, the concept of sustainable development was defined at Brundtland Commission as: “development that meets the needs of the present without compromising the ability of future generation to meet their own needs” (Brundtland Commission, 1987). In the last decade, cost–benefit analysis has been considered as the major evaluation system for sustainable development activities. However, such monetary cost–benefit evaluation procedures do not consider all of the resources involved. Although it is difficult, to give an “adequate” definition of sustainability, the measurement and assessment on sustainability of at least three aspects, embracing the economic cost that determines the operation and maintenance of the system, input/output efficiency that is necessary for scarce resource allocation, and the “ecological cost” of restoration that is important to deal with the interaction between the biosphere and societal environment, have to be thoroughly included (Chen et al., 2009).

8. Conclusions To solve the multifold water-related problems in China, completely replication of centralized water-, energy- and costintensive technology has proved to be extremely limited and not feasible, especially in fast growing small- to medium-sized urban area in China. Constructed wetlands have gained increasing attention and been implemented as wastewater treatment facilities in many parts of the world because of their low-cost and energysavings. This paper reviews the progress of CWs for wastewater treatment in China, and delineate some of the key treatment efficiency and performance issues which may be elucidated by the China experience. Comparison on the existing of FWS, HSSF, VSSF and hybrid systems in China that we have data for indicates that hybrid systems perform best in the removal of TSS, BOD5 , COD, and TP. Compared

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

to VSSF systems, HSSF systems showed better removal efficiency for BOD5 and TP (82.22% and 59.01%, respectively), although for TSS removal the VSSF showed much better removal efficiency (75.52%). As for nitrogen removal, the TN removal efficiency of HSSF systems was significantly higher than that for VSSF systems. And surprisingly, even the ammonia removal efficiency of HSSF systems in China (56.2%) was higher than for the VSSF systems (43.3%). Additionally, this comparison of removal efficiencies by CWs in China to CW treatment of 268 systems throughout Europe (Haberl et al., 1995) indicated that the removal rates for nearly all the parameters, were higher in China than Europe. Experience in China show that plants can play a key role and make a significant difference in treatment efficiency. Numerous comparative studies have verified that the planted wetlands show higher removal efficiency of TN and NH4 -N than that unplanted wetlands. However, plants play a much lesser role in the removal of TP, COD and BOD5 . For COD and BOD5 , it would appear that in most wetland systems either anaerobic decomposition plays a major role or, alternatively, aeration of the substrate (without plants) is sufficient to satisfy the oxygen demand of organics removal. Although CW systems in China do not have an apparent advantage in construction costs, the costs for treatment and operation/maintenance of CW systems are much lower than those of conventional WWTP and activated sludge processes. Similarly, results of technical-economic comparative analysis of various CW systems in China indicate that energy consumption for different CW systems was far less than that of conventional activated sludge technology. Land requirements for CWs present one of the factors most limiting their broader use, especially in southern China, where land resources are scarce and population density is high. References Barko, J.Wo., Gunnison, D., Carpenter, S.R., 1991. Sediment interactions with submerged macrophyte growth and community dynamics. Aquat. Bot. 41, 41–65. Brissaud, F., 2007. Low technology systems for wastewater perspectives. Water Sci. Technol. 55 (7), 1–9. Brix, H., 1990. Gas exchange through the soil–atmosphere interphase and through dead culms of Phragmites australis in a constructed reed bed receiving domestic sewage. Water Res. 24, 259–266. Brix, H., 1994. Functions of macrophytes in constructed wetlands. Water Sci. Technol. 29 (4), 71–78. Brix, H., Sorrel, B.K., Schierup, H.-H., 1996. Gas fluxes achieved by in situ convective flow in Phragmites australis. Aquat. Bot. 54, 151–163. Brix, H., 1997. Do macrophytes play a role in constructed treatment wetland? Water Sci. Technol. 35, 11–17. Brown, M.T., Herendeen, R.A., 1996. Embodied energy analysis and emergy analysis: a comparative review. Ecol. Econom. 19, 219–235. Brown, M.T., Ulgiati, S., 1999. Energy evaluation of the biosphere and natural capital. AMBIO 28, 486–493. Brundtland Commission, 1987. Our Common Future. Oxford University Press, New York. Chan, S., Tsang, Y.F., Chua, H., Sin, S.N., Cui, L.H., 2008. Performance study of vegetated sequencing batch coal slag bed treating domestic wastewater in suburban area. Bioresource Technol. 99, 3774–3781. Chen, B., Chen, Z.M., Zhou, Y., Zhou, J.B., Chen, G.Q., 2009. Emergy as embodied energy based assessment for local sustainability of a constructed wetland in Beijing. Commun. Nonlinear Sci. Numeric. Simul. 14, 633–635. Chen, Z.M., Chen, B., Zhou, J.B., Li, Z., Zhou, Y., 2008. A vertical subsurface-flow constructed wetland in Beijing. Commun. Nonlinear Sci. Numeric. Simul. 13, 1986–1997. China Daily Report, 2005. China’s urban population to reach 560 Million. December 17, 2005. Available http://english.peopledaily.com.cn/200512/17/ eng20051217 228778.html (accessed May 21, 2009). Chung, A.K.C., Wu, Y., Tam, N.F.Y., Wong, M.H., 2008. Nitrogen and phosphate mass balance in a sub-surface flow constructed wetland for treating municipal wastewater. Ecol. Eng. 32, 81–89. Cui, L.H., Liu, W., Zhu, X.Z., Ma, M., Huang, X., Xia, Y.Y., 2006. Performance of hybrid constructed wetlands systems for treating septic tank effluent. J. Environ. Sci. 18 (4). Environment Bureau of the State, 1997. National standard for sewage discharge in China. Environment Science Press, Beijing (in Chinese). Geber, U., Bjoerklund, J., 2001. The relationship between ecosystem services and purchased input in Swedish wastewater treatment systems – a case study. Ecol. Eng. 19, 97–117.

1377

Gersberg, R.M., Elkins, B.V., Goldman, C.R., 1983. Nitrogen removal in artificial wetlands. Water Res. 17 (9), 1009–1041. Gersberg, R.M., Elkins, B.V., Lyon, S.R., Goldman, C.R., 1986. Roles of aquatic plants in wastewater treatment by artificial wetland. Water Res. 20 (3), 363–368. Haberl, R., Perfler, R., Mayer, H., 1995. Constructed wetland in Europe. Water Sci. Technol. 32 (3), 305–315. He, L.S., liu, H.L., Xi, B.D., Zhu, Y.B., 2006. Effects of effluent recirculation in verticalflow constructed wetland on treatment efficiency of livestock wastewater. Water Sci. Technol. 54 (11–12), 137–146. Huang, J., Reneau, R.B., Hagedorn, J.R.C., 2000. Nitrogen removal in constructed wetlands employed to treat domestic wastewater. Water Res. 34 (9), 2582–2588. Jamieson, T.S., Stratton, G.W., Gordon, R., Madani, A., 2003. The use of aeration to enhance ammonia nitrogen removal in constructed wetlands. Can. Biosyst. Eng. 45, pp. 1.9–1.14. Jenssen, P.D., Maehlum, T., Zhu, T., 1996. Construction and performance of subsurface flow constructed wetlands in Norway. Paper presented at the symposium on constructed wetlands in cold climates. Niagra-on-the-Lake, Orttario, June 4–5, 1996. Ji, C.D., Sun, T.H., Ni, J.R., 2007. Surface flow constructed wetland for heavy oilproduced water treatment. Ecol. Eng. 98, 436–441. Jiang, T., He, J., Yang, X., Lv, B., 2008. Nutrients transfer in subsurface-flow constructed wetland. In: Bioinformatics and Biomedical Engineering, 2008. ICBBE 2008. Proceedings of the Second International Conference. Kadlec, R.H., 2009. Comparison of free water and horizontal subsurface treatment wetland. Ecol. Eng. 35, 159–174. Kivaisi, A.K., 2001. The potential for constructed wetlands for wastewater treatment and reuse in developing countries: a review. Ecol. Eng. 16, 545–560. Langergraber, G., Leroch, K., Pressl, A., Rohrhofer, R., Haberl, R., 2008. A two-stage subsurface vertical flow constructed wetland for high-rate nitrogen removal. Water Sci. Technol. 57 (12), 1881–1887. Li, L., Wang, Q.Q., 2006. The development of constructed wetlands in China. Available www.Chinacitywater.org (accessed October 12, 2008) (in Chinese). Li, L.F., Li, Y.H., Biswas, D.K., Nian, Y.g., Jiang, G.M., 2008. Potential of constructed wetlands in treating the eutrophic water: evidence from Taihu Lake of China. Bioresource Technol. 99, 1656–1663. Li, X., Jiang, C., 1995. Constructed wetland systems for water pollution control in north China. Water Sci. Technol. 32 (3), 349–356. Li, X., Chen, M., Anderson, B.C., 2009. Design and performance of a water quality treatment wetland in a public park in Shanghai, China. Ecol. Eng. 35, 18–24. Liu, C., Du, G., Huang, B., Meng, Q., Li, H., Wang, Z., Song, F., 2007. Biodiversity and water quality variations in constructed wetland system. Acta Ecol. Sin. 27 (9), 3670–3677. Lu, s., Zhang, P., Jin, X., Xiang, C., Gui, M., Zhang, J., Li, F., 2009. Nitrogen removal from agricultural runoff by full-scale constructed wetland in China. Hydrobiologia 621 (1), 115–126. Lüderitz, V., Eckert, E., Lange-Weber, M., Lange, A., Gersberg, R., 2001. Nutrient removal efficiency and resource economics of vertical flow and horizontal flow constructed wetlands. Ecol. Eng. 18, 157–171. Ma, S., 1978. The development of environmental system theory and its significance. The Report in Inaugural Meeting of Environmental Science Society, China (in Chinese). Ma, S.J., 1988. Development of agro-ecological engineering in China. In: Ma, S.J., Jiang, A., Xu, R., Li, D. (Eds.), Proceedings of International Symposium on AgroEcological Engineering. Ecological Society of Beijing, August 1988, pp. 1–13. Maehlum, T., Stalnacke, P., 1999. Removal efficiency of three cold climate constructed wetlands treating domestic wastewater: effects of temperature, seasons, loading rates and input concentrations. Water Sci. Technol. 40, 273–281. Maehlum, T., Jenssen, P.D., Warner, W.S., 1995. Cold-climate constructed wetlands. Water Sci. Technol. 32, 95–101. Mander, U., Mitsch, W.J., 2009. Pollution control by wetlands. Ecol. Eng. 35, 153– 158. Merlin, G., Pajean, J., Lissolo, T., 2002. Performances of constructed wetlands for municipal wastewater treatment in rural mountainous area. Hydrobiologia (469), 87–98. Mitsch, W.J., Jørgensen, S.E., 1989. Ecological Engineering: An Introduction to Ecotechnology. John Wiley & Sons, Inc, New York, 472 pp. Mitsch, W.J., Yan, J.S., Cronk, J.K., 1993. Ecological engineering – contrasting experiences in China with the west. Ecol. Eng. 2, 177–191. Mitsch, W.J., 1997. Ecological Engineering: the roots and rational of a new ecological paradigm. In: Etnier, C., Guterstam, B. (Eds.), Ecological Engineering for Wastewater Treatment, 2nd edition. CRC Press, USA, pp. 1–20. Mitsch, W.J., Jørgensen, S.E., 2003a. Ecological engineering: a field whose time has come. Ecol. Eng. 20, 363–377. Mitsch, W.J., Jørgensen, S.E., 2003b. Ecological engineering in China. In: Mitsch, W.J., Jørgensen, S.E. (Eds.), Ecological Engineering and Ecosystem Restoration. John Wiley & Sons, Inc, pp. 309–336. Odum, H.T., 1988. Self-organization, transformation, and information. Science 242, 1132–1139. Odum, H.T., 1996. Environmental Accounting: Emergy and Environmental Decision Making. John Wiley & Sons, Inc, New York. Peng, J.F., Wang, B.Z., Wang, L., 2005. Multi-stage ponds–wetlands ecosystem for effective wastewater treatment. J. Zhejiang Univ. Sci. 6B (5), 346–352. Peterson, S.B., Teal, J.M., 1995. The role of plants in ecologically engineered wastewater treatment systems. Ecol. Eng. 6, 137–148.

1378

D. Zhang et al. / Ecological Engineering 35 (2009) 1367–1378

Shi, L., Wang, B.Z., Cao, X.D., 2004. Performance of a subsurface-flow constructed wetland in southern China. J. Environ. Sci. 16 (3), 476–481. Song, Z.W., Zheng, Z.P., Li, J., Sun, X.F., Han, X.Y., Wang, W., Xu, M., 2006. Seasonal and annual performance of a full-scale constructed wetland system for sewage treatment in China. Ecol. Eng. 26, 272–282. Song, Z., Wu, L., Xu, M., Wen, S., Zhou, Y., Yu, M., 2008. Distribution and survival of six kinds of indicator and pathogenic microorganisms in a full-scale constructed wetlands in China. In: Bioinformatics and Biomedical Engineering, 2008. ICBBE 208. Proceedings of the Second International Conference, May 16–18. Tang, X.Q., Huang, S.L., Scholz, M., 2009. Nutrient removal in pilot-scale constructed wetlands treating eutrophic river water: assessment of plants, intermittent artificial aeration and polyhedron hollow polypropylene balls. Water Air Soil Poll. 197, 61–73. U.S. Department of Commerce, 2005. Water supply and wastewater treatment market in China. U.S. EPA, 1993. Constructed Wetland for Wastewater Treatment and Wildlife Habitat. Office of Research and Development, EPA 832-R-93-005, September 1993. U.S. EPA, 2000. Constructed wetlands treatment of municipal wastewaters manual. Office of Research and Development, EPA-625-R-99-010, September 2000. Varis, O., Vakkilainen, P., 2006. China’s challenges to water resources management. Agrifood Res. Report 68, 115–129. Verhoeven, J.T.A., Meuleman, A.F.M., 1999. Wetlands for wastewater treatment: opportunities and limitations. Ecol. Eng. 12, 5–12. Vymazal, J., 2005. Horizontal sub-surface flow and hybrid constructed wetlands systems for wastewater treatment. Ecol. Eng. 25, 478–490. Vymazal, J., 2006. Removal of nutrients in various types of constructed wetland. Sci. Tot. Environ. 380 (1–3), 48–65. Wang, J., Cai, X., Chen, Y., Yang, Y., Liang, M., Zhang, Y., 1994. Analysis of the configuration and the treatment effect of constructed wetland wastewater treatment system for different wastewaters in South China. In: Proceedings of Fourth International Conference Wetland Systems for Water Pollution Control, Guangzhou, PR China, pp. 114–120. Wang, L., Peng, J., Wang, B., Cao, R., 2005a. Performance of a combined eco-system of ponds and constructed wetlands for wastewater reclamation and reuse. Water Sci. Technol. 51 (12), 315–323. Wang, L., Peng, J., Wang, B.L., Yang, L., 2006a. Design and operation of an ecosystem for municipal wastewater treatment and utilization. Water Sci. Technol. 54 (11–12), 429–436. Wang, Q.H., Duan, L.S., Wu, J.Y., Yang, J., 2008. Growth vitality and pollutantsremoval ability of plants in constructed wetland in Beijing region. Chin. J. Appl. Ecol. 19 (5), 1131–1137. Wang, X., Bai, X., Wang, B., 2005b. Municipal wastewater treatment with pondconstructed wetland system: a case study. Water Sci. Technol. 51 (12), 325–329. Wang, S., Xu, Z.X., Li, H.Z., 2006b. Enhanced strategies in vertical flow constructed wetlands for domestic wastewater treatment. Environ. Sci. 27 (12), 2432–2438.

Wittgren, H.B., Maehlum, T., 1997. Wastewater treatment wetlands in cold climates. Water Sci. Technol. 35 (5), 45–53. Wu, Y., Chung, A., Tan, N.F.Y., Pi, N., Wong, M.H., 2008. Constructed mangrove wetland as secondary treatment system for municipal wastewater. Ecol. Eng. 34, 137–146. Xia, N., Liu, H., Guo, R., Zhang, H., Yang, K., 2006. Research on nitrogen removal and microorganism in a subsurface flow constructed wetland system in Sihong County. J. China Univ. Mining and Tech. 16 (4), 505–508. Yan, J.S., Zhang, Y.S., Wu, X.Y., 1993. Advances of ecological engineering in China. Ecol. Eng. 2, 193–215. Yang, Y., Xu, Z., Hu, K., Wang, J., Wang, G., 1995. Removal efficiency of the constructed wetland: wastewater treatment system at Bainikeng, Shenzhen, China. Water Sci. Technol. 32 (3), 31–40. Yang, Q., Chen, Z.H., Zhao, J.G., Gu, B., 2007. Contaminant removal of domestic wastewater by constructed wetlands: effects of plants species. J. Integr. Plant Biol. 49 (4), 437–446. Yang, Q., Tam, N.F.Y., Wong, Y.S., Luan, T.G., Su, W.S., Lan, C.Y., Shin, P.K.S., Cheung, S.G., 2008. Potential use of mangroves as constructed wetland for municipal sewage treatment in Futian, Shenzhen, China. Mar. Pollut. Bull. 57, 735–743. Yin, W., Ye, M., Lei, A., 2008. Comparison of different types of constructed wetlands. Ren Min Chang Jing, 2. Available http://www.nsbd.com.cn/ NewsDisplay.asp?Id=207283 (accessed April 29, 2009). Yin, H., Shen, W., 1995. Using reed beds for winter operation of wetland treatment system for wastewater. Water Sci. Technol. 32 (3), 111–117. Yin, W., Li., P.J., Guo., W., 2004. Application limitation and operation of subsurface flow constructed wetland. China Water & Wastewater, 20 (11) 1000-4602(2004)11-0036-03 (in Chinese). Zhai, J., He, Q., Kerstens, S., 2006. Experimental study on a new type of hybrid constructed wetland in South China. Report of project: sustainable water management improves tomorrow’s cities’ health (SWICH018530) supported by the sixth framework programme of EU. Zhang, J., Shao, W.S.H.M., Hu, H.Y., Gao, B., 2006. Treatment performance and enhancement of subsurface constructed wetland. Huan Jing Ke Xue 27 (8), 1560–1564 (in Chinese). Zhou, J.B., Jiang, M.M., Chen, B., Chen, G.Q., 2009. Energy evaluations for constructed wetland and conventional wastewater treatments. Commun. Nonlinear Sci. Numeric. Simul. 14 (2009), 1781–1789. Zhu, T., Silora, F.J., 1994. Ammonium and nitrate removal in vegetated and unvegetated gravel bed microcosm wetland. In: Proceedings of Fourth International Conference Wetland Systems for Water Pollution Control, ICWSI94, Secretariat, Guangzhou, PR China, pp. 335–366. Zuo, P., Wan, S.W., Qin, P., Du, J., Wang, H., 2004. A Comparison of the sustainability of original and constructed wetlands in Yancheng biosphere reserve, China: implications from emergy evaluation. Environ. Sci. Policy 7 (204), 329–343.