Effects of seasons and hydraulic loading rates (HLR) on the treatment performance and the response of the microbial community of vertical flow constructed wetland treating tail water were investigated. The seasonal treatment performance was evaluated at four HLR of 125, 250, 375 and 500 mm/d, respectively. The microbial community was detected by MiSeq Illumina platform at HLR 125 and 375 mm/d. The wetland showed significantly higher chemical oxygen demand (COD) and total nitrogen (TN), total phosphorus (TP) at HLR 125 mm/d, compared with other HLR. Overall removal efficiency was 61.47%, 71.40% and 76.31% for COD, TN and TP, respectively, while no significant differences for COD, TN and TP removal were found at HLR of 250, 375 and 500 mm/d. The best removal efficiency for COD and TN was achieved in summer and autumn, while the best TP removal was achieved in winter. Nitrification bacteria (Nitrosomonas and Nitrospira) were significantly higher in HLR 125 mm/d, whereas sequences associated with denitrification had no significant difference at the two HLR. The results can partially explain the significantly higher NH4+-N removal in HLR 125 mm/d and relatively low nitrogen performance in winter.

INTRODUCTION

China is a country with a serious water shortage; meanwhile, freshwater resources in China are subject to increasing pressure in the form of consumptive water use and pollution. Tail water is the treated effluent from conventional wastewater treatment plants, which contains relatively high concentrations of nitrate and low levels of organic matter (Spieles & Mitsch 2000; Andersson et al. 2005; Leverenz et al. 2010). Typically, tail water is discharged into rivers, lakes, estuaries and oceans. Excessive nutrients may cause eutrophication and impact ecosystem health (Greenway 2005). There is an urgent need for an effective and cost-saving technology for the advanced treatment of tail water.

Constructed wetland integrates physical, chemical and biological processes to treat wastewater, which has been considered one of the most promising ecological technologies for wastewater treatment due to its low cost, simple operation and maintenance, minimal secondary pollution, favorable environmental appearance and other ecosystem service benefits (Haberl 1999; Kivaisi 2001; Vymazal 2005). Vertical-subsurface flow constructed wetlands (VFCW) have been widely used in the world for sewage treatment over the past decades (Prochaska et al. 2007). However, the treatment efficiencies vary considerably depending on variables such as system type and design, retention time, hydraulic and nutrient mass loading rates, climate, vegetation, and microbial communities (EPA 1995). Previous studies indicate that hydrologic characteristics such as hydraulic loading rates (HLR), retention time, and water depth are vitally important to determine the treatment performance of a wetland system (Mitsch & Gosselink 1993; Kadlec & Knight 1996). Most studies focused on the treatment of domestic wastewater while there has been less research on the study of tail water treatment in the subtropical area of China.

Microbial communities in the wetland ecosystem play a vital role in the biogeochemical process of water quality improvement in CWs (Mitsch & Gosselink 2007; Saunders et al. 2013). Any shift in the diversity or composition of the microbial community might directly affect the purification performance of CWs. Characterizing the microbial communities in CWs will provide valuable information for understanding the function of the system (Zhong et al. 2015). However, the study of microbial communities from CWs is still limited, especially for tail water treatment. The lack of relevant knowledge may impede the effective design and operation of CWs (Stottmeister et al. 2003; Faulwetter et al. 2009).

Molecular methods have been increasingly used to investigate microbial diversity and activity in the environment such as CWs, which largely overcame the limit of conventional culture-based methodologies (Raghoebarsing et al. 2006; Ansola et al. 2014; Guo et al. 2015; Zhong et al. 2015). The Illumina MiSeq platform is becoming increasingly popular for 16S rRNA gene amplification sequencing because it can generate longer paired end reads (now up to 2 × 300 bp reads) and up to ten times more sequences per run (Caporaso et al. 2012).

Therefore, the objectives of this study were to investigate the treatment performance in different seasons at four HLR, and try to give insight into the variation of the microorganisms with HLR and seasons.

MATERIALS AND METHODS

Configuration of the experimental wetland systems and study site

Four parallel experimental vertical-flow constructed wetland systems were built at the experimental base of Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China (30°30.39′N, 114°28.09′E). Wuhan is located in the subtropical area of China with a humid subtropical monsoon climate. There are abundant rainfall and sunshine as well as four distinct seasons. Summer begins from May. Midsummer starts in July. The maximum temperature during this time mostly stays at 37–39 °C. Autumn starts after October, with temperature gradually declining and the air becoming dry. Winter begins at the end of December and runs through next February, with temperature that can be lower than −5 to 0 °C. Spring is back after March with rapid rise of temperature, even to a maximum above 20 °C.

The wetland chamber (1.25 m × 1 m × 1 m; length × width × depth) was filled with a 40 cm thick layer of gravel (Ø = 10–20 mm) at the bottom, followed by a 35 cm thick layer of Granolithic (Ø = 5–10 mm) at the upper layer. To minimize the fluctuation, synthetic tail wastewater was used in this study. Components and characteristics of the tail water are presented in Table 1. The wetland was planted with Phragmites australis, at a density of 10 plants/m2.

Table 1

Components and characteristics of the influent

ComponentsParametersMeanSTD
Glucose pH 7.47 0.26 
NaCO3 DO/(mg/L) 5.75 2.21 
KH2PO4 EC/(μs/cm) 460.51 22.56 
CaCl2 TN/(mg/L) 12.18 1.30 
MgSO4 NH4+-N/(mg/L) 6.52 1.21 
FeCl3·6H2NO3-N/(mg/L) 5.46 1.39 
NaNO3 TP/(mg/L) 0.71 0.11 
NH4Cl CODCr/(mg/L) 54.50 4.97 
ComponentsParametersMeanSTD
Glucose pH 7.47 0.26 
NaCO3 DO/(mg/L) 5.75 2.21 
KH2PO4 EC/(μs/cm) 460.51 22.56 
CaCl2 TN/(mg/L) 12.18 1.30 
MgSO4 NH4+-N/(mg/L) 6.52 1.21 
FeCl3·6H2NO3-N/(mg/L) 5.46 1.39 
NaNO3 TP/(mg/L) 0.71 0.11 
NH4Cl CODCr/(mg/L) 54.50 4.97 

COD, chemical oxygen demand; DO, dissolved oxygen; EC, electric conductivity; TN, total nitrogen; TP, total phosphorus.

Operation of the wetland systems

Each wetland system was loaded with tail water at the corresponding HLR of 125, 250, 375 and 500 mm/day since October 2013. After three months of the acclimation period, purification performance for pollutants was conducted from February 2014 to January 2015. The loading rates were controlled daily by different levels of inlet valves, with the expected amount of wastewater from the storage tank being drained slowly into the surface of each wetland. The outflow was collected through a row of porous pipes installed at the bottom of the wetland. All of these porous pipes were connected to the main outflow pipe. The outlet level was set just above the bed surface to ensure water saturated conditions in the wetland substrate. While the inflow entered through the top of the wetland, treated wastewater at the bottom will be pushed into the outflow pipe and flow out of the outlet by gravity.

Sampling and analysis

Water samples were collected weekly from the inlet and outlet of each wetland system. Physical parameters such as water temperature (T), dissolved oxygen (DO), pH and electric conductivity (EC) were measured immediately using YSI 556 Multiparameter System (Yellow Springs Instrument Company, USA). Chemical parameters including chemical oxygen demand (COD), total phosphorus (TP), total nitrogen (TN), ammonium nitrogen (NH4+-N) and nitrate nitrogen (NO3-N) were determined according to the methods of AHPA (1998).

Substrate samples were collected from the top layer (5–20 cm) of the wetland for microbial community analysis in the initial period February and March 2014, every two months after March. The samples were taken from five random spots of the wetland surface and well mixed in polyethylene before being stored in a freezer at −20 °C. DNA was extracted from pretreated samples using the PowerSoil®DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA). The V4 region of 16sRNA was amplified using primer 515F 5′-GTGCCAGCMGCCGCGG-3′ and 907R 5′-CCGTCAATTCMTTTRAGTTT-3′. Polymerase chain reaction (PCR) reactions were performed in triplicate 20 μL reactions with 0.4 μM forward and reverse primers, 1 μL template DNA, 250 nM deoxynucleotide triphosphate (dNTP), and 1× FastPfu Buffer. All dilutions were made using certified DNA-free PCR water. Thermal cycling consisted of initial denaturation at 95 °C for 2 min followed by 25 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and an extension at 72 °C for 30 s, with a final extension at 72 °C for 5 min. Replicate amplicons were pooled and visualized on 2.0% agarose gels using SYBR Safe DNA gel stain in 0.5 × Tris/Borate/EDTA (TBE). Amplicons were purified using an AxyPrep™ DNA Gel Extraction Kit (AXYGEN Company) according to the manufacturer's instructions. Sequencing was performed on the MiSeq Illumina sequencing platform. OTU taxonomists' analysis and cluster analysis were used to analyze the diversity and abundance of the screened and filtered sequences.

Statistical analysis

Statistical analysis for all data was performed using SPSS version 18.0 for Windows. One-way analysis of variance (ANOVA) was performed to test the differences of removal efficiencies and difference of microbial community compositions between different seasons at each HLR or between different HLR in each season, with p < 0.05 defined as a significant difference.

RESULTS

Treatment performance of VFCW in different seasons

The seasonal variations of the temperature and removal efficiencies of COD, TP, NH4+-N, NO3-N and TN under different HLR are shown in Figure 1. The removal efficiency of COD in spring was lower than those in other seasons. The highest removal performance of COD was achieved in summer, with removal efficiencies ranging from 60% to 70% during summer, autumn and winter at 125 mm/d, while removal efficiencies at other seasons were around 50% for other three HLR. However, the highest TP removal efficiency was achieved in winter for all HLR (83.81% ± 6.64, 74.55% ± 8.82, 65.95% ± 13.45 and 64.27% ± 13.57 for HLR 125, 250, 375 and 500 mm/d, respectively). No significant difference was detected in TP removal at HLR 250 and 375 mm/d, especially during spring and summer (p > 0.05). TN removal rates in autumn were significantly higher than those in other seasons at HLR of 125 mm/d, while at other HLR the wetland system showed no obvious difference in TN removal in spring, summer, and autumn seasons, while it decreased significantly in winter. A similar trend was found in NH4+-N removal at HLR 125, 250, 375 mm/d; the highest removal rates were achieved in autumn compared with other seasons. The removal of NO3-N remained at a high level in spring, summer and autumn compared with other contaminants. The highest removal rates for each HLR were 97.74% ± 2.52, 96.63% ± 2.31, 94.50% ± 5.20 and 95.81% ± 5.54 for HLR 125, 250, 375 and 500 mm/d, respectively, which was achieved in spring for HLR 125 and 250 mm/d, in autumn for 375 and 500 mm/d. The lowest removal rates were achieved in winter for all HLR.
Figure 1

Seasonal purification performance of VFCW for tail water at different HLR.

Figure 1

Seasonal purification performance of VFCW for tail water at different HLR.

Bacterial diversity and composition

Relative abundance of bacterial taxa was examined at different levels to determine whether there was significant shift in the diversity and composition of the bacterial communities in response to HLR and seasons. At the class level, a total of 46 and 50 classes were detected at the initial stage for HLR of 125 and 375 mm/d. For both wetlands, Betaproteobacteria, Actinobacteria, Gammaproteobacteria, and Alphaproteobacteria were the dominant classes. The dominant classes account for 86.99% and 81.87% of the total sequences, respectively, while a total of 67 and 71 classes were detected in January 2015. Betaproteobacteria, Actinobacteria, Flavobacteria, and Alphaproteobacteria were the dominant classes for both wetlands. The dominant classes account for 80.53% and 78.18% of the classified sequences for HLR 125 and 375 mm/d, respectively. At the order level, a total of 95 and 102 orders were obtained at the initial stage for HLR 125 and 375 mm/d. For both wetlands, Burkholderiales, Micrococcales, and Pseudomonadales were the dominant orders at the initial stage, while Burkholderiales, Flavobacteriales, and Rhodocyclales were the dominant orders at the end of the experiment. At the family level, the dominant species were Comamonadaceae, Flavobacteriaceae, and Rhodocyclaceae for HLR 125 mm/d. Comamonadaceae and Rhodocyclaceae were dominant for HLR 375 mm/d at the initial stage. While Micrococcaceae and Pseudomonadaceae were dominant for HLR 125 mm/d, Comamonadaceae and Micrococcaceae were dominant for 375 mm/d at the end of the experiment.

The results showed that at the initial stage the dominant communities in both wetlands were Arthrobacter, Pseudomonas and Comamonadaceae at the genus level, which comprised 60.76% and 46.48% of the total sequences for HLR 125 and 375 mm/d, respectively, while at the end of the experiment, the dominant bacterial communities were Piscinibacter, Flavobacterium for HLR 125 mm/d wetland samples, and Comamonadaceae, Piscinibacter and Thauera were dominant in HLR 375 mm/d wetland samples. The percentage of dominant sequences are 41.66% and 42.25% for HLR 125 and 375 mm/d, respectively.

A hierarchical bar chart displaying the bacterial community compositions of the samples from the two different wetland systems during the experiment period is shown in Figure 2. The results show a similar pattern of succession of the bacterial composition in the two wetland systems. With the running of the wetlands, the dominant communities such as Arthrobacter began to decrease until they almost disappeared after September. Instead, the abundance of Piscinibacter increased with time and became dominant at the final stage. Pseudomonas was a dominant community at the beginning, and the abundance decreased with time and disappeared after July. Meanwhile, some emerging communities such as Thauera began to appear. The abundance of Thauera increased from May and it became one of the dominant species in September, and higher abundance of Thauera was detected in HLR of 125 mm/d wetland compared with HLR 375 mm/d wetland. At the same time, there were relatively stable communities existing in both systems. The abundance of Flavobacterium first rises and then falls, but it's the main bacterial communities throughout the whole experiment period. By contrast, the abundance of Comamonadaceae first falls, then rises, and was dominant throughout the experiment period.
Figure 2

Hierarchical bar chart for the bacterial community (Genus) in wetlands at HLR 125 mm/d (left) and 375 mm/d (right) (abbreviated month names represent sampling time; numeric values represent HLR, 1 for 125 mm/d and 2 for 375 mm/d).

Figure 2

Hierarchical bar chart for the bacterial community (Genus) in wetlands at HLR 125 mm/d (left) and 375 mm/d (right) (abbreviated month names represent sampling time; numeric values represent HLR, 1 for 125 mm/d and 2 for 375 mm/d).

DISCUSSION

Effect of seasons and HLR on treatment performance

It showed that seasons had no obvious influence on the COD removal in this study. The influent COD loading used in this study was relatively low compared with domestic wastewater (Chang et al. 2012). Old and decaying roots probably have contributed to the relatively high background levels of COD recorded in this study (Trang et al. 2010).

TN and NO3-N removal rates were much more affected by seasons, and the lowest removal efficiencies for both TN and NO3-N were obtained in winter because the activity of microorganisms and plant uptake plays a significant role in nitrogen removal. A value of 15 °C was usually selected as a bound above which efficient nitrogen removal could be achieved due to the properly functioning microorganisms and vegetation (Kuschk et al. 2003; Kadlec & Wallace 2009). The mean temperature in winter was 9.07 °C in this study. This result was inconsistent with the previous studies (Misiti et al. 2011; Chang et al. 2013).

Consequently, the better nitrogen removal during the autumn could be attributed to the suitable temperature and well-developed denitrifying bacteria population with high activities along with the luxuriant plants (Woltemade & Woodward 2008).

The influence of seasons on TP removal was totally different. The highest removal efficiency for TP was observed in winter for all the four HLR. The major processes responsible for P removal in the constructed wetlands are substrate adsorption, precipitation and microbial assimilation and plant uptake. The Granolithic and gravel used as the substrate and the high content of iron, magnesium or calcium might be responsible for the efficient P removal. Efficient adsorption of P to the bed medium and associated high P removals have also been reported by Kern & Idler (1999), but removals are likely to decrease over time due to the saturation of P sorption sites in the medium (Arias & Brix 2005). The TP removal performance suggested that the adsorption capacity of the media was steady during the experiment period, while the increased TP removal over time is probably attributed to the gradual development of the microbial community and the assimilation of P by the microbes.

Optimal HLR is important to achieve good wastewater treatment performance within wetland systems (Yu et al. 2011). At low HLR, the retention time is high, and at high HLR, the water passes rapidly to the outlet, reducing the contact time between the wastewater and the microorganisms, which are considered to be responsible for the degradation processes. The removal performance of contaminants was especially affected by HLR. Lower removal rates are expected at higher HLR, due to the respectively limited contact time between wastewater and microorganisms. However, the removal of COD, TN, and TP in this study was only slightly affected, which is similar to Trang et al. (2010). The results also indicate a shift on the microbial community patterns as they adjust to the different HLRs.

As HLR increased, the COD removal efficiencies didn't decrease as expected. Results obtained show about 61% of COD removal at HLR 125 mm/d, which is significantly higher than those of the other three HLR (p < 0.05), while there is no significant difference in COD removal efficiencies at the other three HLR. And the removal rate of COD over one year at HLR 375 mm/d was even slightly higher than that of HLR 250 mm/d. VFCW is widely used for COD removal from wastewater. Wang et al. (2009) built integrated VFCW for the advanced treatment of tail water, and achieved about 71% removal of COD at HLR 60–80 mm/d. Sklarz et al. (2009) showed an 84% reduction in COD from domestic wastewater using a VFCW. As for TN removal, it appeared from the results that there is no significant difference (p > 0.05) at the HLR 250, 375 and 500 mm/d. Only with HLR of 125 mm/d is there a significantly higher removal performance. Except 125 mm/d, NH4+-N were poorly removed at HLR 250, 375 and 500 mm/d, with a removal percentage of 31.70%, 27.56% and 20.21%, respectively. Meanwhile, HLR seems to have no effect on NO3-N removal; the overall removal of NO3-N at all the four HLR was 80–90%. TP removal decreased with the increasing of HLR. Contaminant removal efficiencies didn't decrease proportionally with the increasing of HLR. Results obtained in this study reveal that VFCW systems are capable of substantial removal reduction of COD, TN, and TP with a good buffering capacity between 250 and 500 mm/d HLR range.

Response of bacterial community to HLR and seasons

No significant differences between the dominant bacterial communities (Genus) at HLR 125 and 375 mm/d were detected (p > 0.05). But some dominant communities, such as Propionicicella (p = 0.053) and Comamonadaceae (p = 0.153), appeared to be more significant than those of others.

Multiple samples similarity analysis of bacterial communities in different months and at different HLR are shown in Figure 3. The similarity trees mainly can be divided into two branches. The first branch consists of the samples in the initial months, February, March and May 2014. The rest of the samples were clustered into the second branch. The similarity trees show that in the initial months, the samples at 125 and 375 mm/d share a high similarity. What's more, it appears from the figure that the samples in the three months are very close to each other, which means the bacterial community of the samples shares a high similarity. The distance of the samples of HLR 125 and 375 mm/d began to increase after July 2014. The distance between the samples of the two wetlands increased with the running time. This is probably because the difference of the diversity and composition in the two wetlands began to increase as a result of different hydraulic conditions.
Figure 3

Multiple samples similarity analysis of bacterial communities over time and at HLR 125 and 375 mm/d.

Figure 3

Multiple samples similarity analysis of bacterial communities over time and at HLR 125 and 375 mm/d.

Response of functional bacteria to HLR and seasons

The genus sequences associated with nitrogen cycling in the samples of the two wetland systems are summarized in Table 2.

Table 2

The number of genera sequences associated with nitrogen cycling in the samples of the two wetlands

ProcessOTU ID (Genus)125 mm/d
FebMarMayJulSepNovJan
Nitrification 
 Ammonium oxidation Nitrosomonas 14 23 11 12 
 Nitrite oxidation Nitrospira 67 92 76 35 
Denitrification 
 Autotrophic Thiobacillus 65 111 38 92 31 22 
 Heterotrophic Dechloromonas 47 64 34 323 244 140 202 
Denitratisoma 32 281 132 86 37 
Flavobacterium 1,081 542 187 14 79 165 1,995 
Hydrogenophaga 388 388 510 300 91 85 249 
Hyphomicrobium 11 24 19 16 11 
Rhodobacter 567 1,087 1,263 597 445 410 367 
Thauera 57 1,204 5,027 2,111 1,397 
Bradyrhizobium 37 32 41 34 13 
375 mm/d
Nitrification 
 Ammonium oxidation Nitrosomonas 
 Nitrite oxidation Nitrospira 34 18 30 12 
Denitrification 
 Autotrophic Thiobacillus 134 65 136 26 15 18 
 Heterotrophic Dechloromonas 366 57 32 169 200 232 294 
Denitratisoma 16 149 102 165 166 
Flavobacterium 2,108 353 107 – 95 172 1,252 
Hydrogenophaga 859 304 175 399 88 112 162 
Hyphomicrobium 12 
Rhodobacter 558 1,631 1,508 422 182 132 81 
Thauera 15 175 4,575 5,317 2,340 2,176 
Bradyrhizobium 14 16 18 10 
ProcessOTU ID (Genus)125 mm/d
FebMarMayJulSepNovJan
Nitrification 
 Ammonium oxidation Nitrosomonas 14 23 11 12 
 Nitrite oxidation Nitrospira 67 92 76 35 
Denitrification 
 Autotrophic Thiobacillus 65 111 38 92 31 22 
 Heterotrophic Dechloromonas 47 64 34 323 244 140 202 
Denitratisoma 32 281 132 86 37 
Flavobacterium 1,081 542 187 14 79 165 1,995 
Hydrogenophaga 388 388 510 300 91 85 249 
Hyphomicrobium 11 24 19 16 11 
Rhodobacter 567 1,087 1,263 597 445 410 367 
Thauera 57 1,204 5,027 2,111 1,397 
Bradyrhizobium 37 32 41 34 13 
375 mm/d
Nitrification 
 Ammonium oxidation Nitrosomonas 
 Nitrite oxidation Nitrospira 34 18 30 12 
Denitrification 
 Autotrophic Thiobacillus 134 65 136 26 15 18 
 Heterotrophic Dechloromonas 366 57 32 169 200 232 294 
Denitratisoma 16 149 102 165 166 
Flavobacterium 2,108 353 107 – 95 172 1,252 
Hydrogenophaga 859 304 175 399 88 112 162 
Hyphomicrobium 12 
Rhodobacter 558 1,631 1,508 422 182 132 81 
Thauera 15 175 4,575 5,317 2,340 2,176 
Bradyrhizobium 14 16 18 10 

It has been reported that Nitrosomonas is the main bacterium involved in ammonium oxidization (Pedersen et al. 2009), and Nitrospira is mainly involved in nitrite oxidization (Hovanec et al. 1998). Table 2 shows that sequences belonging to Nitrosomonas were significantly higher in HLR 125 wetland than in HLR 375 mm/d since March 2014 (p < 0.01), and sequences belonging to Nitrospira were significantly higher in HLR 125 wetland than in HLR 375 mm/d since July 2014. It has partially explained the significantly higher NH4+-N removal in HLR 125 mm/d wetland. Nitrosomonas and Nitrospira are the main AOB and NOB genera detected in the system.

Most denitrifying bacteria are facultative anaerobic chemoheterotrophs that use organic compounds as electron donors with nitrogen oxides (in ionic and gaseous form) as terminal electron acceptors (Ahn 2006). As seen from Table 2, we can observe that the numbers of most of the sequences involved in denitrification are similar between the two wetlands. This probably suggests that the influence of HLR on denitrification is limited. It also explained why HLR seems to have no effect on NO3-N removal.

Nitrogen removal in CWs was achieved mainly by the functioning microorganism degradation and vegetation uptake. Previous studies have suggested that temperature plays a key role in nitrogen removal by affecting microbial activity and vegetation function (Kuschk et al. 2003; Kadlec & Wallace 2009). The result is similar to our experiment result; the lowest removal efficiencies for both NH4+-N and NO3-N in winter were achieved in this study.

CONCLUSION

Constructed wetland showed a good performance in contaminant removal from tail water. The increase of HLR will lead to a decrease of pollutant removal efficiency, while a good buffering capacity was shown between HLR 250 and 375 mm/d. TP removal responded to seasons differently compared with TN and COD removal; the best removal efficiencies for COD and TN were found in summer and autumn, while the best TP removal was found in winter in this study.

No significant differences between the dominant bacterial communities at HLR 125 and 375 mm/d were detected at genus level (p > 0.05). Less similarity was detected after July 2014 than in the initial period between the two wetlands. Nitrification bacteria (Nitrosomonas and Nitrospira) were significantly higher in HLR 125 mm/d wetland compared with HLR 375 mm/d wetland, which partially explained the significantly higher NH4+-N removal in HLR 125 mm/d wetland and relatively lower nitrogen purification performance in winter.

ACKNOWLEDGEMENTS

This work was supported by grants from National Natural Science Foundation of China (51578538, 51408593), Natural Science Foundation of Hubei Province (Y53A15), Zhejiang Province Science and Technology Plan (2015C32011) and Science and technology promotion project of Ministry of Water Resources (TG1520).

REFERENCES

Ahn
Y. H.
2006
Sustainable nitrogen elimination biotechnologies: a review
.
Process Biochemistry
41
(
8
),
1709
1721
.
American Public Health Association (AHPA)
1998
Standard Methods for the Examination of Water and Wastewater
, 20th edn.
American Public Health Association Inc.
,
Washington, DC
.
Andersson
J. L.
Bastviken
S. K.
Tonderski
K. S.
2005
Free water surface wetlands for wastewater treatment in Sweden: nitrogen and phosphorus removal
.
Water Science & Technology
51
,
39
46
.
Arias
C. A.
Brix
H.
2005
Phosphorus removal in constructed wetlands: can suitable alternative media be identified?
Water Science & Technology
51
(
9
),
267
273
.
Caporaso
J. G.
Lauber
C. L.
Walters
W. A.
Berg-Lyons
D.
Huntley
J.
Fierer
N.
Owens
S. M.
Betley
J.
Fraser
L.
Bauer
M.
Gormley
N.
Gilbert
J. A.
Smith
G.
Knight
R.
2012
Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms
.
The ISME Journal
6
,
1621
1624
.
Chang
J. J.
Wu
S. Q.
Dai
Y. R.
Liang
W.
Wu
Z. B.
2012
Treatment performance of integrated vertical-flow constructed wetland plots for domestic wastewater
.
Ecological Engineering
44
,
152
159
.
Chang
J. J.
Wu
S. Q.
Dai
Y. R.
Liang
W.
Wu
Z. B.
2013
Nitrogen removal from nitrate-laden wastewater by integrated vertical-flow constructed wetland systems
.
Ecological Engineering
58
,
192
201
.
EPA
1995
Handbook of Constructed Wetlands: Domestic Wastewater
,
vol. 2
.
Environment Protection Agency
,
Washington, DC
.
Faulwetter
J. L.
Gagnon
V.
Sundberg
C.
Chazarenc
F.
Burr
M. D.
Brisson
J.
Camper
A. K.
Stein
O. R.
2009
Microbial processes influencing performance of treatment wetlands: a review
.
Ecological Engineering
35
,
987
1004
.
Hovanec
T. A.
Taylor
L. T.
Blakis
A.
Delong
E. F.
1998
Nitrospira-like bacteria associated with nitrite oxidation in freshwater aquaria
.
Applied Environmental Microbiology
64
,
258
264
.
Kadlec
R. H.
Knight
R. L.
1996
Treatment Wetlands
.
Lewis Publishers, CRC Press
,
Boca Raton
.
Kadlec
R. H.
Wallace
S. D.
2009
Treatment Wetlands
, 2nd edn.
CRC Press
,
Boca Raton, FL
.
Kern
J.
Idler
C.
1999
Treatment of domestic and agricultural wastewater by reed bed systems
.
Ecological Engineering
,
12
(
1–2
),
13
25
.
Kivaisi
A. K.
2001
The potential for constructed wetlands for wastewater treatment and reuse in developing countries: a review
.
Ecological Engineering
16
(
4
),
545
560
.
Kuschk
P.
Wiebner
A.
Kappelmeyer
U.
Weißbrodt
E.
Kästner
M.
Stottmeister
U.
2003
Annual cycle of nitrogen removal by a pilot-scale subsurface horizontal flow in a constructed wetland under moderate climate
.
Water Research
37
,
4236
4242
.
Leverenz
H. L.
Haunschild
K.
Hopes
G.
Tchobanoglousa
G.
Darby
J. L.
2010
Anoxic treatment wetlands for denitrification
.
Ecological Engineering
36
,
544
551
.
Misiti
T. M.
Hajaya
M. G.
Pavlostathis
S. G.
2011
Nitrate reduction in a simulated free-water surface wetland system
.
Water Research
45
(
17
),
5587
5598
.
Mitsch
W. J.
Gosselink
J. G.
1993
Wetlands
, 2nd edn.
John Wiley
,
New York
.
Mitsch
W. J.
Gosselink
J. G.
2007
Wetlands
, 4th edn.
John Wiley
,
New York
.
Pedersen
L. F.
Pedersen
P. B.
Nielsen
J. L.
Nielsen
P. H.
2009
Peracetic acid degradation and effects on nitrification in recirculating aquaculture systems
.
Aquaculture
296
,
246
254
.
Raghoebarsing
A. A.
Pol
A.
Smolders
A. J.
Ettwig
K. F.
Rijpstra
W. I.
Schouten
S.
Damste
J. S. S.
Camp
H. J. M. O
Jetten
M. S. M.
Strous
M.
2006
A microbial consortium couples anaerobic methane oxidation to denitrification
.
Nature
440
(
7086
),
918
921
.
Sklarz
M. Y.
Gross
A.
Yakirevich
A.
Soares
M. I. M.
2009
A recirculating vertical flow constructed wetland for the treatment of domestic wastewater
.
Desalination
246
(
1–3
),
617
624
.
Stottmeister
U.
Wiebner
K. P.
Kappelmeyer
U.
Kastner
B. O.
Muller
R. A.
Moormann
H.
2003
Effects of plants and microorganisms in constructed wetlands for wastewater treatment
.
Biotechnology Advances
22
,
93
117
.
Wang
X.
Han
B. P.
Shi
Y. Z.
Pang
Z. Q.
2009
Advanced wastewater treatment by integrated vertical flow constructed wetland with Vetiveria zizanioides in north China
. In:
Procedia Earth and Planetary Science: The 6th International Conference on Mining Science & Technology
, pp.
1258
1262
.
Woltemade
C. J.
Woodward
J.
2008
Nitrate removal in a restored spring-fed wetland, Pennsylvania, USA
.
Journal of the American Water Association
44
(
1
),
222
234
.
Zhong
F.
Wu
J.
Dai
Y. R.
Yang
L. H.
Zhang
Z. H.
Cheng
S. P.
Zhang
Q.
2015
Bacterial community analysis by PCR-DGGE and 454-pyrosequencing of horizontal subsurface flow constructed wetlands with front aeration
.
Applied Microbiology and Biotechnology
99
(
3
),
1499
1512
.