- Email: [email protected]
Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
Ecological Engineering xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater Ewa Wojciechowska ∗ , Magdalena Gajewska, Arkadiusz Ostojski Gda´ nsk University of Technology, Faculty of Civil and Environmental Engineering, ul. Narutowicza 11/12, Gda´ nsk, 80-233, Poland
a r t i c l e
i n f o
Article history: Received 21 January 2016 Received in revised form 23 May 2016 Accepted 6 July 2016 Available online xxx Keywords: Landfill leachate Reject water from sewage sludge dewatering Nitrification Denitrification Ammonium adsorption Biodegradability
a b s t r a c t Treatment wetlands have been proved to be more effective than conventional treatment processes in case of high-strength wastewater containing high concentrations of ammonium nitrogen and recalcitrant organic matter. In this study nitrogen removal processes and reliability of nitrogen removal at two identical pilot-scale multistage treatment wetlands (MTWs) receiving real, non-synthetic wastewater were discussed. The wastewater discharged to pilot-scale subsurface flow MTWs contained high ammonium nitrogen concentrations and limited biodegradable organic matter (OM) concentrations. One of the pilot MTWs was fed with landfill leachate (LL) and another one with reject water from sewage sludge centrifugation (RW). In the first season of operation both MTWs reached very high (95–99%) efficiency of total nitrogen (TN) removal, which was explained by adsorption to the substrate. In the later period, TN was removed in a sequence of nitrification and denitrication processes with lower efficiency (40–86%). The denitrification process was the limiting one at the LL treating site because of too low carbon supply. Still, during the whole investigation period TN outflow concentrations were at the level that allows for co-treatment of the effluent in municipal wastewater treatment plants without risk for biological treatment processes. In 90% of samples TN concentration was below 200 mg/l and in 60% of samples it was below 145 mg/l. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The treatment of wastewater containing high concentrations of recalcitrant organic matter (OM) and ammonium nitrogen, like landfill leachate (LL) or reject water from mechanical dewatering of sewage sludge after the digestion process (RW) is hard to accomplish in conventional treatment processes. Poor biodegradability indicates proportions of easily biodegradable OM (expressed in BOD5 ) in total organic matter (expressed in COD) and total Kjeldahl nitrogen (TKN) usually determines the failure of activated-sludge or SBR reactors (Wiesman, 1994). However, the treatment wetlands (TWs) have been reported to perform better and offer reliable treatment results in case of high-strength wastewater like LL (Kadlec, 2003; Kinsley et al., 2006). This significant difference between the conventional biological treatment and treatment wetlands is explained to be the result of more complex processes occurring in the wetland systems and the unique capacity of wetland systems in providing endogenous sources of biodegradable OM (Crites et al., 2006; Kadlec and Wallace, 2009; Gajewska et al., 2015).
∗ Corresponding author. E-mail address: [email protected] (E. Wojciechowska).
Experiences with the application of TWs for LL indicated that significant reduction of BOD, COD and nitrogen concentrations, as well as BTEX is possible. Surface flow systems (SF) have proved to be very effective in LL treatment due to long hydraulic retention time and complex removal processes. (Johnson et al., 1999; Waara et al., 2008; Obarska-Pempkowiak et al., 2015). The removal of extremely high ammonium nitrogen concentrations generally requires a sequence of nitrification and denitrification processes (Makinia et al., 2009; Vymazal, 2009). In case of horizontal subsurface flow systems (HSSF), nitrification is limited due to anaerobic conditions, which could be overcome by additional aeration (Nivala et al., 2007). In vertical subsurface flow systems (VSSF) oxygen transfer is good enough and creates favourable conditions for nitrification of ammonium nitrogen presence in LL (ObarskaPempkowiak et al., 2010; Yalcuk and Ugurlu, 2009; Lavrova and Koumanova, 2010). On the other hand denitrification requires anoxic conditions, as well as the presence of easily biodegradable carbon. Multistage TWs offer a solution with nitrification going on in VSSF beds and denitrification running in HSSF beds (Molle et al., 2008; Masi et al., 2013; Gajewska et al., 2015). Kadlec and Zmarthie (2010) recommend LL treatment in vertical subsurface flow beds (VSSF), willingly followed by a SF system for denitrification. The final SF system can be replaced by a HSSF bed, while a
http://dx.doi.org/10.1016/j.ecoleng.2016.07.006 0925-8574/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model ECOENG-4317; No. of Pages 7
ARTICLE IN PRESS E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
2
Table 1 The average RW and LL composition (in/out). Parameter
Fig. 1. Scheme of twin MTWs for LL and RW treatment with marked sampling points (1–inflow, 2–outflow).
sequence of two VSSF beds working in a series (instead of a single bed) can provide full nitrification of enormous ammonium nitrogen load (Obarska-Obarska-Pempkowiak and Gajewska, 2011). The objective of this paper is to discuss the nitrogen removal processes at two identical MTWs receiving wastewater with low biodegradable OM and high ammonium nitrogen concentrations. The study MTWs were the outdoor pilot scale installations placed at (i) landfill site and fed with real LL and (ii) municipal wastewater treatment plant and fed with real RW from digested sludge centrifugation. In our investigation “twin” installations gave the opportunity to analyze the differences in operation depending on the quality of inflow of RW and LL, as well as they allowed to find similarities in removal processes. The other distinction of our investigation is working with real, and not synthetic wastewater, which gave more complex results due to the presence of organic and inorganic fractions of solids, which could have influence on removal processes in MTWs. The TN removal rates and the reliability of removal processes during three years of operation of LL MTW and two years of operation of RW MTW were evaluated with respect to the adaptation and post-adaptation period, basing on the inflow and outflow pollutants concentrations. In this paper we would like to focus on MTWs consisting of three stages, but taken as a whole unit with regard to the removal process reliability, which is of significance for potential co-treatment of municipal wastewater. 2. Materials and methods
RW
total N [mg/l] NH4 + -N [mg/l] organic N [mg/l] TKN [mg/l] N-NO3 − [mg/l] COD [mg O2 /l] BOD5 [mg O2 /l]
LL
inflow
outflow
inflow
outflow
1054 941 113 1054 0.1 1208 460
123 96 23 119 4.3 264 21
542 434 77 511 2.2 1825 151
190 103 35 138 19.2 953 19
Table 2 The average COD/BOD5 , COD/TKN and BOD5 /TKN ratios in RW and LL. Parameter
COD/BOD5 COD/TKN BOD5 /TKN BOD5 /TN
RW
LL
inflow
outflow
inflow
outflow
2.63 1.25 0.46 0.44
12.50 2.83 0.23 0.17
9.09 2.63 0.28 0.28
33.33 6.45 0.18 0.10
regarded as an adaptation period. It was considered that the conditions during the adaptation of the beds (for instance intensively developing plants and “fresh” gravel media) can influence the treatment performance. The concentrations of COD, BOD5 , total nitrogen, ammonium nitrogen, organic nitrogen and nitrates were measured in samples (inflow and outflow) collected according to the hydraulic retention time. The chemical analyses were performed by an independent laboratory (ISO certificate), according to Polish standard methods (Polish Environmental Ministry Regulations of 24th July 2006 with amendment of 18th January 2009; COD: PN-ISO 15705:2005; total N: PN-73/C-04576.14; ammonium N: PN-ISO 5664:2002, nitrates: PN-EN ISO 10304-1:2009). Polish standards are in agreement with the EU Framework Directive 2000/60/EC and are comparable with APHA (2005).
2.1. Study sites 3. RESULTS and discussion Two identical multistage MTWs for LL and RW treatment consisted of three SSF beds, working in series: two VSSF (7.84 m2 and 5.29 m2 ) beds followed by a HSSF bed (3.19 m2 ), all planted with P. australis. Washed gravel (4–8 mm) with the hydraulic conductivity of 4.2 × 10−2 m/s was used as a substrate. One of the MTWs received leachate from the landfill in Chlewnica near Słupsk and another one treated reject waters from sewage sludge centrifugation at WWTP ´ in Gdansk. The VSSF beds were operated in batch mode (flooded and drained) except for the first operation season (adaptation period) at LL MTW, when both VSSF beds were continuously flooded to promote the development of P. australis that was damaged by birds breeding at the landfill site. The wastewater inflow was equal to 111 l/d. The scheme of twin MTWs is presented in Fig. 1. Further details of the investigation sites were provided in previous papers (Gajewska and Obarska-Pempkowiak, 2011; Wojciechowska and Gajewska, 2013). 2.2. Sampling and analytical procedures The averaged samples of wastewater at the inflow and at the outflow were collected every 3–4 weeks during the vegetation seasons (from April to early November) in 1-l glass bottles and immediately transported to the laboratory in cooling containers. The analyses were conducted in three subsequent vegetation seasons (2009–2011) in the LL MTW and during the two subsequent seasons in RW MTW (2009–10). The first season of operation was
3.1. Composition of raw and treated wastewater The composition of raw wastewater is summarized in Table 1. Both, LL and RW contained high concentrations of total N, mostly in the form of ammonium nitrogen, although the total nitrogen concentration in RW was twice as high as in LL. Organic nitrogen constituted approximately 11% and 14% of the total nitrogen present in RW and LL, respectively. Nitrates concentrations in raw wastewater were low. The outflow concentrations indicate effective removal of ammonium nitrogen from both LL and RW. Organic nitrogen concentrations of 23 mg/l (RW) and 35 mg/l (LL) remained in the effluent. The remaining organic nitrogen was supposed to be built up in the structure of complex organic compounds, like humic substances and thus unavailable for biotransformation processes (Kang et al., 2002; Obarska-Pempkowiak et al., 2015), which is in agreement with very high concentrations of recalcitrant OM in LL, expressed in high inflow/outflow COD concentration and high COD/BOD5 ratio. The LL at the outflow contained nitrates, indicating incomplete denitrification. The OM concentrations expressed in COD were very high, especially in LL. At the same time, the BOD5 mean concentration was four times lower in LL than in RW. In Table 2 the values of characteristic ratios BOD5 /COD, COD/TKN and BOD5 /TKN are summarized. The high COD/BOD5 ratio indicates that recalcitrant forms of organics were present in both types of wastewater. The LL COD/BOD5
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
3
adaptation period 100% 80% 60% 40%
02.12.11
11.10.11
10.09.11
21.09.11
28.08.11
03.09.11
11.08.11
18.08.11
31.07.11
11.07.11
21.07.11
30.06.11
20.06.11
29.05.11
03.06.11
03.11.10
10.10.10
27.09.10
12.09.10
18.08.10
29.08.10
10.08.10
02.08.10
07.07.10
16.07.10
20.06.10
01.07.10
07.06.10
25.05.10
10.11.09
27.10.09
30.09.09
15.10.09
10.09.09
02.09.09
18.08.09
24.08.09
12.08.09
14.07.09
28.07.09
06.07.09
0%
20.06.09
20%
14.06.09
N tot removal efficiency [%]
120%
Fig. 2. Total nitrogen removal efficiency from LL.
N-NH 4 + removal efficiency [%]
120%
adaptation period 100% 80% 60% 40% 20%
2 8.08.11
0 3.09.11
1 0.09.11
2 1.09.11
1 1.10.11
0 2.12.11
28.08.11
03.09.11
10.09.11
21.09.11
11.10.11
02.12.11
1 8.08.11 18.08.11
1 1.08.11
3 1.07.11
2 1.07.11
1 1.07.11
3 0.06.11
2 0.06.11
0 3.06.11
2 9.05.11
0 3.11.10
1 0.10.10
1 2.09.10
2 7.09.10
2 9.08.10
1 8.08.10
1 0.08.10
0 2.08.10
0 7.07.10
1 6.07.10
0 1.07.10
2 0.06.10
0 7.06.10
2 5.05.10
1 0.11.09
2 7.10.09
1 5.10.09
3 0.09.09
1 0.09.09
0 2.09.09
2 4.08.09
1 8.08.09
1 2.08.09
2 8.07.09
1 4.07.09
0 6.07.09
1 4.06.09
2 0.06.09
0%
Fig. 3. Ammonium nitrogen removal efficiency from LL.
adaptation period
100% 80% 60% 40%
31.07.11
11.08.11
21.07.11
11.07.11
30.06.11
20.06.11
03.06.11
29.05.11
10.10.10
27.09.10
12.09.10
18.08.10
29.08.10
10.08.10
02.08.10
16.07.10
07.07.10
01.07.10
20.06.10
25.05.10
07.06.10
10.11.09
27.10.09
15.10.09
30.09.09
10.09.09
02.09.09
24.08.09
18.08.09
12.08.09
28.07.09
14.07.09
06.07.09
0%
20.06.09
20%
14.06.09
Norg removal efficiency [%]
120%
Fig. 4. Organic nitrogen removal efficiency from LL.
ratio was outstandingly low, indicating very poor biodegradability of OM. This explains high outflow COD concentration in LL, while the COD reduction in RW was considerably higher. The outflow BOD5 concentrations were at the residual level, similar for both types of wastewater, showing that the labile OM was almost totally consumed in the biological treatment processes. The values of COD/TKN and BOD5 /TKN ratios indicate that the removal of nitrogen could be problematic. Miksch and Sikora (2010) state that when COD/BOD5 ratio is below 2 and at the same time BOD5 /TN ratio is over 4, the efficiency of organic matter and total nitrogen removal in conventional biological processes is over 90%. In TWs, however, the ratios are considered not to be so rigorous.
According to Crites et al. (2006) and Kadlec and Wallace (2009) effective nitrification can take place in TWs with BOD5 /TKN ratio below 1. After the evaluation of nine multistage TWs working in different climatic conditions Gajewska et al. (2015) concluded that the highest N removal rates were most commonly observed with the BOD5 /TN ratio in the range 1.5–2.5. 3.2. Efficiency of nitrogen removal processes In spite of the fact that the analyzed MTWs received wastewater with outstandingly low BOD5 /TKN ratio below 0.5, the removal of nitrogen species was relatively good.
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
4
120% N tot removal efficiency [%]
adaptation period 100% 80% 60% 40%
Oct 10
Sept 10
Aug 10
Aug 10
July 10
July 10
June 10
June 10
May 10
Apr 10
Oct 09
Sept 09
Aug 09
Aug 09
July 09
July 09
June 09
May 09
Apr 09
0%
June 09
20%
120%
adaptation period
100% 80% 60% 40%
Oct 10
Sept 10
Aug 10
Aug 10
July 10
July 10
June 10
June 10
May 10
Apr 10
Oct 09
Sept 09
Aug 09
Aug 09
July 09
July 09
June 09
May 09
0%
June 09
20%
Apr 09
+
N-NH4 removal efficiency [%]
Fig. 5. Total nitrogen removal efficiency from RW.
Fig. 6. Ammonium nitrogen removal efficiency from RW.
adaptation period
100% 80% 60% 40%
Oct 10
Sept 10
Aug 10
Aug 10
July 10
July 10
June 10
June 10
May 10
Apr 10
Oct 09
Sept 09
Aug 09
Aug 09
July 09
July 09
June 09
June 09
0%
May 09
20% Apr 09
Norg removal efficiency [%]
120%
Fig. 7. Organic nitrogen removal efficiency from RW.
In case of both MTWs the nitrogen removal rates varied in time. During the first operation season (adaptation period) very high TN removal efficiencies were obtained, varying from 95 to 99% for RW MTW, and exceeding 99% in case of LL MTW ((Figs. 2 and 5). In the next operation seasons the performance of both MTWs was deteriorated. Total nitrogen removal efficiencies at RW MTW were quite stable and varied from 78 to 86%. In case of LL MTW the efficiencies of TN removal after the adaptation period varied from 40 to 86%. The treatment efficiencies were influenced by the composition of LL, which contained higher share of recalcitrant OM in 2010 operation season (Wojciechowska, 2015). Ammonium nitrogen removal rates were similar to TN removal rates at both sites, with high and stable removal efficiency (exceeding 95% for RW and above 99% for
LL) in the adaptation period, followed by a decrease in the later seasons (Figs. 3 and 6). Organic nitrogen removal was quite stable during the adaptation period at both MTWs, opposing to the subsequent seasons, when considerable fluctuations were observed (Figs. 4 and 7). The post-adaptation period performance was different for each MTW. Ammonium nitrogen and organic nitrogen were removed with considerably lower efficiency in the LL MTW than in the RW MTW. The outflow concentrations of nitrates (V) were higher in the LL MTW than in the RW MTW. The deterioration of removal efficiency after the first season of operation, which obviously took place at both investigation sites, requires some considerations. Interestingly, very high removal efficiency was obtained in LL treatment site during the adaptation
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
Inflow
5
Outflow
120
adaptation period
-
N-NO3 [mg/l]
100 80 60 40 20
02.12.11
11.10.11
10.09.11
21.09.11
28.08.11
03.09.11
11.08.11
18.08.11
21.07.11
31.07.11
11.07.11
30.06.11
03.06.11
20.06.11
03.11.10
29.05.11
10.10.10
12.09.10
27.09.10
18.08.10
29.08.10
10.08.10
16.07.10
02.08.10
01.07.10
07.07.10
20.06.10
25.05.10
07.06.10
27.10.09
10.11.09
30.09.09
15.10.09
02.09.09
10.09.09
24.08.09
18.08.09
28.07.09
12.08.09
06.07.09
14.07.09
14.06.09
20.06.09
0
Fig. 8. Concentrations of nitrates in LL MTW.
Inflow RW
Outflow RW
10
adaptation period
6
-
N-NO3 [mg/l]
8
4
29.08.10
18.08.10
10.08.10
02.08.10
16.07.10
07.07.10
01.07.10
20.06.10
07.06.10
25.05.10
10.09.09
02.09.09
24.08.09
18.08.09
12.08.09
28.07.09
14.07.09
06.07.09
20.06.09
0
14.06.09
2
Fig. 9. Concentrations of nitrates in RW MTW.
Fig. 10. Extreme value distribution curves of total nitrogen during adaptation period (a), after adaptation period (b) and altogether in the whole investigation period (c).
period, though VSSF beds were flooded during this period. Nitrification was not likely to occur under such circumstances due to limited oxygen transfer. This is supported by low nitrates concentrations measured during the adaptation period (Fig. 8), though according to Nivala et al. (2007) simultaneous nitrification and denitrification can take place in the root matrix due to the development of aerobic and anoxic microzones. However, the adsorption process seems to explain better good ammonium nitrogen removal observed at LL and RW treatments sites in the first operation season and its decrease in the later
operation, due to the saturation of the available attachment sites. According to Reed et al. (1995) the ammonium nitrogen removal is often significantly higher in the first two years of operation than in the later period due to adsorption to the substrate and the intensive uptake by developing plants. Gray et al. (2000) considered substrate adsorption to be an important pathway for ammonium nitrogen removal in TW with the most important processes of ion exchange and physicochemical adsorption to substrate. Zhu et al. (2011a, b) studied kinetic adsorption of ammonium nitrogen by 8 different substrates and compared their efficiency of ammonium removal
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model ECOENG-4317; No. of Pages 7 6
ARTICLE IN PRESS E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
and reported the intermediate sorption capacity of gravel among other analyzed substrates. Another finding of this study was that the adsorption of ammonium nitrogen by 8 analyzed substrates increased with initial NH4 + -N concentration (the initial concentrations of 100 mg/l and 1500 mg/l were compared). According to Kadlec and Wallace (2009) ammonium nitrogen adsorbed by the substrate is loosely bound and could be released when water chemistry changes or is oxidized to nitrate, which is not bound to the substrate and is washed out. After the adaptation period VSSF beds in the LL MTW were operated in batch mode, enabling oxygen transfer. Also, the population of bacteria had already developed at this stage, which has usually required some time. This has created adequate conditions for the nitrification process evidenced by the presence of nitrates in the outflow from the VSSF beds at both investigation sites. In the last stage of treatment, HSSF bed, denitrification of nitrates took place. However, in the LL MTW denitrification was incomplete and nitrates were present in the outflow, probably due to the insufficient supply of labile OM (Faulwetter et al., 2009; Wojciechowska, 2015). Similar situations were previously reported for other TWs treating LL (Kozub and Liehr, 1999; Hamersley and Howes, 2002; Rustige and Nolde, 2006; Nivala et al., 2007). In case of RW MTW the nitrogen removal processes seemed to be more stable in time and adsorption processes were smoothly replaced by other ones. In contrast to LL RW contained higher concentrations of labile OM in the form of BOD5 , which could favor the denitrification process in MTW. Moreover, according to Gajewska et al. (2015) the OM inflowing to TWs with wastewater is captured in the beds and slowly degraded. Then, smaller and more easily biodegradable molecules that can act as electron donors for denitrification are produced. Also the plant litter decay contributes to increasing availability of carbon (Reed et al., 1995; Masi, 2008). Thus, carbon supply was not the limiting factor for denitrification at RW MTW. Interestingly, after the adaptation period nitrate (V) concentration dropped down below 4 mg/l (Fig. 9). At the same time the removal efficiency of ammonium was only a bit smaller in comparison to the adaptation period and, what is interesting, rather stable in time. Most likely simultaneous nitrification and denitrification could occur, which is a very common process in TWs (Kadlec and Wallace, 2009; Saeed and Sun, 2012). Possibly, anaerobic ammonium oxidation (anammox) could be responsible for the partial removal of ammonium, though neither anammox bacteria detection nor anammox activity analyses were performed within the study. Anammox has much lower carbon and oxygen requirements than conventional routes of nitrification and denitrification (Kadlec and Wallace, 2009; Dong and Sun, 2007). In some literature reports (Bishay and Kadlec, 1995) carbon supply was considered too low to support traditional denitrification or losses of ammonium would have required much higher oxygen transfer (Tanner et al., 2002), what could be explained by the alternative pathway of ammonium oxidation. In the recent years several studies have confirmed that the anammox process actually exists in TWs (Dong and Sun, 2007; Zhu et al., 2010; Zhu et al., 2011a, b; Tao et al., 2012), though according to Coban et al. (2015), the importance of anammox process in nitrogen transformations in HSSF wetlands appears to be low. 3.3. Evaluation of reliability of nitrogen removal processes Although the efficiency of TN removal, especially in RW MTW, was relatively good, the outflow concentrations were still quite high. At RW MTW the average TN concentration was equal to 123 mg/l, similar to TN concentration of raw municipal wastewater (120–130 mg/l), allowing for safe recirculation of RW MTW efflu´ In case ent to the mainstream of wastewater in WWTP Gdansk. of treated LL, the average TN concentration was higher (190 mg/l), however, due to low LL flow rates, the co-treatment at municipal
WWTP would have limited impact on the biological treatment processes (Fudala-Ksiazek et al., 2014). The analysis of the empirical cumulative distribution functions (Fig. 10) allows for assessment of the reliability of nitrogen removal processes at multistage TWs. The graphs present the outflow TN concentrations at both MTWs during the adaptation period (a), after the adaptation period (b) and altogether in the whole investigation period (c). During the adaptation period adsorption assured very effective TN removal. In 60% of samples TN concentration was below 15 mg/l and in 90% of samples it was below 80 mg/l. After the adaptation period, the adsorption capacity depleted, however, nitrification and denitrification microflora had already developed. These processes appeared to be less effective in TN removal from wastewater with low labile OM concentration. In 60% of outflow samples TN concentration was 165 mg/l, while in 90% of samples it was 200 mg/l. In the whole investigation period, in 90% of samples TN concentration was below 200 mg/l and in 60% of samples it was 145 mg/l. This indicates that TN outflow concentrations are at the level that allows for recirculation of treated RW to the mainstream of wastewater, as well as the co-treatment of treated LL in municipal WWTP without a risk for biological treatment processes. However, LL on-site treatment would require further processing to remove TN, as well as recalcitrant OM.
4. Conclusions Various nitrogen removal mechanisms can play different roles during the adaptation period and during the operation of mature SSF MTWs. In this study, during the initial operation period of the newly constructed LL and RW MTWs, adsorption of ammonium nitrogen seemed to be a major removal process, probably supported by plant uptake. In the next years of operation of LL pilot, TN removal was less effective, which is considered to be linked with the depletion of adsorption capacity. Thus, removal rates obtained during the initial operation periods could be considerably higher than long-term removal rates of treatment wetlands. In case of RW MTW the total nitrogen removal efficiency also decreased after the adaptation period, but it remained on considerably higher level than in case of LL MTW. Thanks to the higher concentration of labile OM denitrification of nitrates was quite effective in the RW MTW. Most likely simultaneous nitrification and denitrification was responsible for total nitrogen removal in RW MTW. In the whole investigation period TN outflow concentrations of both, LL and RW MTWs were at the level that allows for co-treatment of treated wastewater in the municipal WWTP without any risk for biological treatment processes. Further research on nitrogen transformation processes (conventional and alternative) in constructed wetlands treating high-strength wastewater should bring more information to understand and optimize the removal mechanisms.
Acknowledgements Funding support from the EEA Financial Mechanism (PL0085 and N N523 425237) and a Ministry of Science and Higher Education in Poland (E007/P01/2007/01) is gratefully acknowledged.
References APHA, 2005. American Public Health Association, Standard methods for the examination of water and wastewater, 21st Ed. Washington, DC. Bishay, L., Kadlec, R., 1995. Wetland treatment at musselwhite mine. In: Vymazal, J. (Ed.), Natural and Constructed Wetlands: Nutrients, Metals and Management. Backhuys Publishers, Leiden, The Netherlands, pp. 176–198.
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model ECOENG-4317; No. of Pages 7
ARTICLE IN PRESS E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
Coban, O., Kuschk, P., Kappelmeyer, U., Spott, O., Martienssen, M., Jetten, M.S.M., Knoeller, K., 2015. Nitrogen transforming community in a horizontal subsurface-flow constructed wetland. Water Res. 74, 203–212. Crites, R.W., Middlebrooks, E.J., Reed, S.C., 2006. Natural Wastewater Treatment Systems. Taylor & Francis Group, New York, pp. 552. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Dong, Z., Sun, T., 2007. A potential new process for improving nitrogen removal in constructed wetlands −promoting coexistence of partial-nitrification and ANAMMOX. Ecol. Eng. 31, 69–78. 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. Ecol. Eng. 35, 987–1004. Fudala-Ksiazek, S., Luczkiewicz, A., Fitobor, K., Olanczuk-Neyman, K., 2014. Nitrogen removal via the nitrite pathway during wastewater co-treatment with ammonium-rich landfill leachates in a sequencing batch reactor. Environ. Sci. Pollut. Res. 21 (2014), 7307–7318. Gajewska, M., Obarska-Pempkowiak, H., 2011. The role of SSVF and SSHF beds in concentrated wastewater treatment, design recommendation. Wat. Sci. Tech. 64, 431–439. ´ Gajewska, M., Józwiakowski, K., Ghrabi, A., Masi, F., 2015. Impact of influent wastewater quality on nitrogen removal rates in multistage treatment wetlands. Environ. Sci. Pollut. Res. 22 (17) (12840-1284). Gray, S., Kinross, J., Read, P., Marland, A., 2000. The nutrient assimilative capacity of maerl as a substrate in constructed wetland systems for waste treatement. Water Res. 34, 2183–2190. Hamersley, M.R., Howes, B.L., 2002. Control of denitrification in a septage-treating artificial wetland: the dual role of particulate organic carbon. Wat. Res. 36, 4415–4427. Johnson, K.D., Martin, C.D., Moshiri, G.A., McCrory, W.C., 1999. Performance of constructed wetland leachate treatment system at Chunchula Landfill, Mobile County, Alabama. In: Mulamoottil, G., Mc Bean, E.A., Rovers, F. (Eds.), Constructed Wetlands for the Treatment of Landfill Leachates. Lewis Publishers, Boca Raton, Florida, USA, pp. 57–70. Kadlec, R.H., Wallace, S.D., 2009. Treatment Wetlands, 2009., 2nd edn. CRC Press, Taylor & Francis Group, Boca Raton, London, New York. Kadlec, R.H., Zmarthie, L.A., 2010. Wetland treatment of leachate from a closed landfill. Ecol. Eng 36, 946–957. Kadlec, R.H., 2003. Integrated natural systems for landfill leachate treatment. In: Vymazal, J. (Ed.), Wetlands—Nutrients, Metals and Mass Cycling. Backhuys Publishers, Leiden, The Netherlands, pp. 1–33. Kang, K.-H., Shin, H.S., Park, H., 2002. Characterization of humic substances present in landfill leachates with different landfill ages and its implications. Water Res. 36, 4023–4032. Kinsley, C.B., Crolla, A.M., Kuyucak, N., Zimmer, M., Lafléche, A., 2006. Nitrogen dynamics in a constructed wetland system treating landfill leachate. Lisbon, Portugal In: Proc. of 10th International Conference on Wetland Systems for Water Pollution Control September 23–29 2006, 2006, pp. 295–305. Kozub, D.D., Liehr, S.K., 1999. Assesssing denitrification rate limiting factors in a constructed wetland receiving landfill leachate. Water Sci. Tech 40, 75–82. Lavrova, S., Koumanova, B., 2010. Influence of recirculation in a lab-scale vertical-flow constructed wetland on the treatment efficiency of landfill leachate. Bioresour. Technol. 101 (2010), 1756–1761. Makinia, J., Stensel, H.D., Czerwionka, K., Drewnowski, J., Zapero, D., 2009. Nitrogen transformations and mass balances in anaerobic/anoxic/aerobic batch experiments with full-scale biomasses from BNR activated sludge systems. Water Sci. Technol. 60 (9), 2463–2470. Masi, F., Caffaz, S., Ghrabi, A., 2013. Multi-stage constructed wetlands systems for municipal wastewater treatment. Water Sci. Technol. 67, 1590–1598. Masi, F., 2008. Enhanced denitrification by an hybrid HF-FWS CW in large scale wastewater treatment plant (Jesi). In: Vymazal, J. (Ed.), Wastewater Treatment, Plant Dynamics and Management in Constructed and Natural Wetlands. Springer, New York, pp. 267–275.
7
Miksch, K., Sikora, J., (2010) Biotechnologia s´ cieków. Wydawnictwo Naukowe PWN. Warszawa, (Biotechnology of wastewater), 216–224. Molle, P., Prost-Boucle, S., Lienard, A., 2008. Potential for total nitrogen removal by combining vertical flow and horizontal flow constructed wetlands: a full-scale experiment study. Ecol. Eng. 34, 23–29. Nivala, J., Hoos, M.B., Cross, C., Wallace, S., Parkin, G., 2007. Treatment of landfill leachate using an aerated, horizontal subsurface-flow constructed wetland. Sci. Total Environ. 380, 19–27. Obarska-Pempkowiak, H., Gajewska, M., Wojciechowska, E., 2010. Application of vertical flow constructed wetlands for highly contaminated wastewater treatment: preliminary results. In: Vymazal, J. (Ed.), Water and Nutrient Management in Natural and Constructed Wetlands, 2010. Springer, London, pp. 37–51. Obarska-Pempkowiak, H., Gajewska, M., Wojciechowska, E., Pempkowiak, J., 2015. ´ Treatment wetlands for environmental pollution control. In: Rowinski, P. (Ed.), GeoPlanet: Earth and Planetary Sciences. Springer International Publishing, p. 169. Reed, S.C., Crites, R.W., Middlebrooks, E.J., 1995. Natural Systems for Waste Management and Treatment, 1995., 2nd edn. McGraw-Hill, Inc., New York, pp. 198–199. Rustige, H., Nolde, E., 2006. Nitrogen elimination from landfill leachates using an extra carbon source in subsurface flow constructed wetlands. Lisbon, Portugal In: Proc. of 10th International Conference on Wetland Systems for Water Pollution Control September 23–29 2006, 2006, pp. 229–239. Saeed, T., Sun, G., 2012. A review on nitrogen and organics removal mechanisms in subsurface flow constructed wetlands: dependency on environmental parameters, operating conditions and supporting media. J Environ Manag 112, 429–448. Tanner, C.C., Kadlec, R., Gibbs, M.M., Sukias, J.P., Nguyen, M.L., 2002. Nitrogen processing gradients in subsurface-flow treatment wetlands: influent wastewater characteristics. Ecol. Eng. 18, 499–520. Tao, W., He, Y., Wang, Z., Smith, R., Shayya, W., Pei, Y., 2012. Effects of pH and temperature on coupling nitrification and anammox in biofilters treating dairy wastewater. Ecol. Eng. 47, 76–82. Vymazal, J., 2009. The use of constructed wetlands with horizontal sub-surface flow for various types of wastewater. Ecol. Eng. 35, 1–17. Waara, S., Waara, K.-O., Forsberg, Å., Fridolfsson, M., 2008. An evaluation of the performance of a constructed wetland system for treatment of landfill leachate during 2003–2006. Warwickshire, England, 16–17 September In: Proc. of Waste 2008: Waste and Resource Management—a Shared Responsibility Stratford-upon-Avon, 2008. Wiesman, U., 1994. Biological nitrogen removal from wastewater. Adv. Biochem. Eng. Biotechnol. 51 (Spinger Verlag, Berlin, Heidelberg). Wojciechowska, E., Gajewska, M., 2013. Partitioning of heavy metals in sub-surface flow treatment wetlands receiving high-strength wastewater. Water Sci. Technol. 68, 486–493. Wojciechowska, E., 2015. Removal of nitrogen compounds from landfill leachate in pilot constructed wetlands. Annu. Set Environ. Prot. 17 (2), 1484–1497 (in Polish). Yalcuk, A., Ugurlu, A., 2009. Comparison of horizontal and vertical constructed wetland systems for landfill leachate treatment. Bioresour. Technol. 100 (2009), 2521–2526. Zhu, G.B., Jetten, M.S.M., Kuschk, P., Ettwig, K.F., Yin, C.Q., 2010. Potential roles of anaerobic ammonium and methane oxidation in nitrogen cycle of wetland ecosystem. Appl. Microbiol. Biotechnol. 86, 1043–1055. Zhu, W.L., Cui, L.H., Ouyang, Y., Long, C.F., Tang, X.D., 2011a. Kinetic adsorption of ammonium nitrogen by substrate materials for constructed wetlands. Pedosphere 21, 454–463. Zhu, G.B., Wang, S.Y., Feng, X.J., Fan, G.N., Jetten, M.S.M., Yin, C.O., 2011b. Anammox bacterial abundance, biodiversity and activity in a constructed wetland. Environ. Sci. Technol. 45 (9).
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
Ecological Engineering xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater Ewa Wojciechowska ∗ , Magdalena Gajewska, Arkadiusz Ostojski Gda´ nsk University of Technology, Faculty of Civil and Environmental Engineering, ul. Narutowicza 11/12, Gda´ nsk, 80-233, Poland
a r t i c l e
i n f o
Article history: Received 21 January 2016 Received in revised form 23 May 2016 Accepted 6 July 2016 Available online xxx Keywords: Landfill leachate Reject water from sewage sludge dewatering Nitrification Denitrification Ammonium adsorption Biodegradability
a b s t r a c t Treatment wetlands have been proved to be more effective than conventional treatment processes in case of high-strength wastewater containing high concentrations of ammonium nitrogen and recalcitrant organic matter. In this study nitrogen removal processes and reliability of nitrogen removal at two identical pilot-scale multistage treatment wetlands (MTWs) receiving real, non-synthetic wastewater were discussed. The wastewater discharged to pilot-scale subsurface flow MTWs contained high ammonium nitrogen concentrations and limited biodegradable organic matter (OM) concentrations. One of the pilot MTWs was fed with landfill leachate (LL) and another one with reject water from sewage sludge centrifugation (RW). In the first season of operation both MTWs reached very high (95–99%) efficiency of total nitrogen (TN) removal, which was explained by adsorption to the substrate. In the later period, TN was removed in a sequence of nitrification and denitrication processes with lower efficiency (40–86%). The denitrification process was the limiting one at the LL treating site because of too low carbon supply. Still, during the whole investigation period TN outflow concentrations were at the level that allows for co-treatment of the effluent in municipal wastewater treatment plants without risk for biological treatment processes. In 90% of samples TN concentration was below 200 mg/l and in 60% of samples it was below 145 mg/l. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The treatment of wastewater containing high concentrations of recalcitrant organic matter (OM) and ammonium nitrogen, like landfill leachate (LL) or reject water from mechanical dewatering of sewage sludge after the digestion process (RW) is hard to accomplish in conventional treatment processes. Poor biodegradability indicates proportions of easily biodegradable OM (expressed in BOD5 ) in total organic matter (expressed in COD) and total Kjeldahl nitrogen (TKN) usually determines the failure of activated-sludge or SBR reactors (Wiesman, 1994). However, the treatment wetlands (TWs) have been reported to perform better and offer reliable treatment results in case of high-strength wastewater like LL (Kadlec, 2003; Kinsley et al., 2006). This significant difference between the conventional biological treatment and treatment wetlands is explained to be the result of more complex processes occurring in the wetland systems and the unique capacity of wetland systems in providing endogenous sources of biodegradable OM (Crites et al., 2006; Kadlec and Wallace, 2009; Gajewska et al., 2015).
∗ Corresponding author. E-mail address: [email protected] (E. Wojciechowska).
Experiences with the application of TWs for LL indicated that significant reduction of BOD, COD and nitrogen concentrations, as well as BTEX is possible. Surface flow systems (SF) have proved to be very effective in LL treatment due to long hydraulic retention time and complex removal processes. (Johnson et al., 1999; Waara et al., 2008; Obarska-Pempkowiak et al., 2015). The removal of extremely high ammonium nitrogen concentrations generally requires a sequence of nitrification and denitrification processes (Makinia et al., 2009; Vymazal, 2009). In case of horizontal subsurface flow systems (HSSF), nitrification is limited due to anaerobic conditions, which could be overcome by additional aeration (Nivala et al., 2007). In vertical subsurface flow systems (VSSF) oxygen transfer is good enough and creates favourable conditions for nitrification of ammonium nitrogen presence in LL (ObarskaPempkowiak et al., 2010; Yalcuk and Ugurlu, 2009; Lavrova and Koumanova, 2010). On the other hand denitrification requires anoxic conditions, as well as the presence of easily biodegradable carbon. Multistage TWs offer a solution with nitrification going on in VSSF beds and denitrification running in HSSF beds (Molle et al., 2008; Masi et al., 2013; Gajewska et al., 2015). Kadlec and Zmarthie (2010) recommend LL treatment in vertical subsurface flow beds (VSSF), willingly followed by a SF system for denitrification. The final SF system can be replaced by a HSSF bed, while a
http://dx.doi.org/10.1016/j.ecoleng.2016.07.006 0925-8574/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model ECOENG-4317; No. of Pages 7
ARTICLE IN PRESS E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
2
Table 1 The average RW and LL composition (in/out). Parameter
Fig. 1. Scheme of twin MTWs for LL and RW treatment with marked sampling points (1–inflow, 2–outflow).
sequence of two VSSF beds working in a series (instead of a single bed) can provide full nitrification of enormous ammonium nitrogen load (Obarska-Obarska-Pempkowiak and Gajewska, 2011). The objective of this paper is to discuss the nitrogen removal processes at two identical MTWs receiving wastewater with low biodegradable OM and high ammonium nitrogen concentrations. The study MTWs were the outdoor pilot scale installations placed at (i) landfill site and fed with real LL and (ii) municipal wastewater treatment plant and fed with real RW from digested sludge centrifugation. In our investigation “twin” installations gave the opportunity to analyze the differences in operation depending on the quality of inflow of RW and LL, as well as they allowed to find similarities in removal processes. The other distinction of our investigation is working with real, and not synthetic wastewater, which gave more complex results due to the presence of organic and inorganic fractions of solids, which could have influence on removal processes in MTWs. The TN removal rates and the reliability of removal processes during three years of operation of LL MTW and two years of operation of RW MTW were evaluated with respect to the adaptation and post-adaptation period, basing on the inflow and outflow pollutants concentrations. In this paper we would like to focus on MTWs consisting of three stages, but taken as a whole unit with regard to the removal process reliability, which is of significance for potential co-treatment of municipal wastewater. 2. Materials and methods
RW
total N [mg/l] NH4 + -N [mg/l] organic N [mg/l] TKN [mg/l] N-NO3 − [mg/l] COD [mg O2 /l] BOD5 [mg O2 /l]
LL
inflow
outflow
inflow
outflow
1054 941 113 1054 0.1 1208 460
123 96 23 119 4.3 264 21
542 434 77 511 2.2 1825 151
190 103 35 138 19.2 953 19
Table 2 The average COD/BOD5 , COD/TKN and BOD5 /TKN ratios in RW and LL. Parameter
COD/BOD5 COD/TKN BOD5 /TKN BOD5 /TN
RW
LL
inflow
outflow
inflow
outflow
2.63 1.25 0.46 0.44
12.50 2.83 0.23 0.17
9.09 2.63 0.28 0.28
33.33 6.45 0.18 0.10
regarded as an adaptation period. It was considered that the conditions during the adaptation of the beds (for instance intensively developing plants and “fresh” gravel media) can influence the treatment performance. The concentrations of COD, BOD5 , total nitrogen, ammonium nitrogen, organic nitrogen and nitrates were measured in samples (inflow and outflow) collected according to the hydraulic retention time. The chemical analyses were performed by an independent laboratory (ISO certificate), according to Polish standard methods (Polish Environmental Ministry Regulations of 24th July 2006 with amendment of 18th January 2009; COD: PN-ISO 15705:2005; total N: PN-73/C-04576.14; ammonium N: PN-ISO 5664:2002, nitrates: PN-EN ISO 10304-1:2009). Polish standards are in agreement with the EU Framework Directive 2000/60/EC and are comparable with APHA (2005).
2.1. Study sites 3. RESULTS and discussion Two identical multistage MTWs for LL and RW treatment consisted of three SSF beds, working in series: two VSSF (7.84 m2 and 5.29 m2 ) beds followed by a HSSF bed (3.19 m2 ), all planted with P. australis. Washed gravel (4–8 mm) with the hydraulic conductivity of 4.2 × 10−2 m/s was used as a substrate. One of the MTWs received leachate from the landfill in Chlewnica near Słupsk and another one treated reject waters from sewage sludge centrifugation at WWTP ´ in Gdansk. The VSSF beds were operated in batch mode (flooded and drained) except for the first operation season (adaptation period) at LL MTW, when both VSSF beds were continuously flooded to promote the development of P. australis that was damaged by birds breeding at the landfill site. The wastewater inflow was equal to 111 l/d. The scheme of twin MTWs is presented in Fig. 1. Further details of the investigation sites were provided in previous papers (Gajewska and Obarska-Pempkowiak, 2011; Wojciechowska and Gajewska, 2013). 2.2. Sampling and analytical procedures The averaged samples of wastewater at the inflow and at the outflow were collected every 3–4 weeks during the vegetation seasons (from April to early November) in 1-l glass bottles and immediately transported to the laboratory in cooling containers. The analyses were conducted in three subsequent vegetation seasons (2009–2011) in the LL MTW and during the two subsequent seasons in RW MTW (2009–10). The first season of operation was
3.1. Composition of raw and treated wastewater The composition of raw wastewater is summarized in Table 1. Both, LL and RW contained high concentrations of total N, mostly in the form of ammonium nitrogen, although the total nitrogen concentration in RW was twice as high as in LL. Organic nitrogen constituted approximately 11% and 14% of the total nitrogen present in RW and LL, respectively. Nitrates concentrations in raw wastewater were low. The outflow concentrations indicate effective removal of ammonium nitrogen from both LL and RW. Organic nitrogen concentrations of 23 mg/l (RW) and 35 mg/l (LL) remained in the effluent. The remaining organic nitrogen was supposed to be built up in the structure of complex organic compounds, like humic substances and thus unavailable for biotransformation processes (Kang et al., 2002; Obarska-Pempkowiak et al., 2015), which is in agreement with very high concentrations of recalcitrant OM in LL, expressed in high inflow/outflow COD concentration and high COD/BOD5 ratio. The LL at the outflow contained nitrates, indicating incomplete denitrification. The OM concentrations expressed in COD were very high, especially in LL. At the same time, the BOD5 mean concentration was four times lower in LL than in RW. In Table 2 the values of characteristic ratios BOD5 /COD, COD/TKN and BOD5 /TKN are summarized. The high COD/BOD5 ratio indicates that recalcitrant forms of organics were present in both types of wastewater. The LL COD/BOD5
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
3
adaptation period 100% 80% 60% 40%
02.12.11
11.10.11
10.09.11
21.09.11
28.08.11
03.09.11
11.08.11
18.08.11
31.07.11
11.07.11
21.07.11
30.06.11
20.06.11
29.05.11
03.06.11
03.11.10
10.10.10
27.09.10
12.09.10
18.08.10
29.08.10
10.08.10
02.08.10
07.07.10
16.07.10
20.06.10
01.07.10
07.06.10
25.05.10
10.11.09
27.10.09
30.09.09
15.10.09
10.09.09
02.09.09
18.08.09
24.08.09
12.08.09
14.07.09
28.07.09
06.07.09
0%
20.06.09
20%
14.06.09
N tot removal efficiency [%]
120%
Fig. 2. Total nitrogen removal efficiency from LL.
N-NH 4 + removal efficiency [%]
120%
adaptation period 100% 80% 60% 40% 20%
2 8.08.11
0 3.09.11
1 0.09.11
2 1.09.11
1 1.10.11
0 2.12.11
28.08.11
03.09.11
10.09.11
21.09.11
11.10.11
02.12.11
1 8.08.11 18.08.11
1 1.08.11
3 1.07.11
2 1.07.11
1 1.07.11
3 0.06.11
2 0.06.11
0 3.06.11
2 9.05.11
0 3.11.10
1 0.10.10
1 2.09.10
2 7.09.10
2 9.08.10
1 8.08.10
1 0.08.10
0 2.08.10
0 7.07.10
1 6.07.10
0 1.07.10
2 0.06.10
0 7.06.10
2 5.05.10
1 0.11.09
2 7.10.09
1 5.10.09
3 0.09.09
1 0.09.09
0 2.09.09
2 4.08.09
1 8.08.09
1 2.08.09
2 8.07.09
1 4.07.09
0 6.07.09
1 4.06.09
2 0.06.09
0%
Fig. 3. Ammonium nitrogen removal efficiency from LL.
adaptation period
100% 80% 60% 40%
31.07.11
11.08.11
21.07.11
11.07.11
30.06.11
20.06.11
03.06.11
29.05.11
10.10.10
27.09.10
12.09.10
18.08.10
29.08.10
10.08.10
02.08.10
16.07.10
07.07.10
01.07.10
20.06.10
25.05.10
07.06.10
10.11.09
27.10.09
15.10.09
30.09.09
10.09.09
02.09.09
24.08.09
18.08.09
12.08.09
28.07.09
14.07.09
06.07.09
0%
20.06.09
20%
14.06.09
Norg removal efficiency [%]
120%
Fig. 4. Organic nitrogen removal efficiency from LL.
ratio was outstandingly low, indicating very poor biodegradability of OM. This explains high outflow COD concentration in LL, while the COD reduction in RW was considerably higher. The outflow BOD5 concentrations were at the residual level, similar for both types of wastewater, showing that the labile OM was almost totally consumed in the biological treatment processes. The values of COD/TKN and BOD5 /TKN ratios indicate that the removal of nitrogen could be problematic. Miksch and Sikora (2010) state that when COD/BOD5 ratio is below 2 and at the same time BOD5 /TN ratio is over 4, the efficiency of organic matter and total nitrogen removal in conventional biological processes is over 90%. In TWs, however, the ratios are considered not to be so rigorous.
According to Crites et al. (2006) and Kadlec and Wallace (2009) effective nitrification can take place in TWs with BOD5 /TKN ratio below 1. After the evaluation of nine multistage TWs working in different climatic conditions Gajewska et al. (2015) concluded that the highest N removal rates were most commonly observed with the BOD5 /TN ratio in the range 1.5–2.5. 3.2. Efficiency of nitrogen removal processes In spite of the fact that the analyzed MTWs received wastewater with outstandingly low BOD5 /TKN ratio below 0.5, the removal of nitrogen species was relatively good.
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
4
120% N tot removal efficiency [%]
adaptation period 100% 80% 60% 40%
Oct 10
Sept 10
Aug 10
Aug 10
July 10
July 10
June 10
June 10
May 10
Apr 10
Oct 09
Sept 09
Aug 09
Aug 09
July 09
July 09
June 09
May 09
Apr 09
0%
June 09
20%
120%
adaptation period
100% 80% 60% 40%
Oct 10
Sept 10
Aug 10
Aug 10
July 10
July 10
June 10
June 10
May 10
Apr 10
Oct 09
Sept 09
Aug 09
Aug 09
July 09
July 09
June 09
May 09
0%
June 09
20%
Apr 09
+
N-NH4 removal efficiency [%]
Fig. 5. Total nitrogen removal efficiency from RW.
Fig. 6. Ammonium nitrogen removal efficiency from RW.
adaptation period
100% 80% 60% 40%
Oct 10
Sept 10
Aug 10
Aug 10
July 10
July 10
June 10
June 10
May 10
Apr 10
Oct 09
Sept 09
Aug 09
Aug 09
July 09
July 09
June 09
June 09
0%
May 09
20% Apr 09
Norg removal efficiency [%]
120%
Fig. 7. Organic nitrogen removal efficiency from RW.
In case of both MTWs the nitrogen removal rates varied in time. During the first operation season (adaptation period) very high TN removal efficiencies were obtained, varying from 95 to 99% for RW MTW, and exceeding 99% in case of LL MTW ((Figs. 2 and 5). In the next operation seasons the performance of both MTWs was deteriorated. Total nitrogen removal efficiencies at RW MTW were quite stable and varied from 78 to 86%. In case of LL MTW the efficiencies of TN removal after the adaptation period varied from 40 to 86%. The treatment efficiencies were influenced by the composition of LL, which contained higher share of recalcitrant OM in 2010 operation season (Wojciechowska, 2015). Ammonium nitrogen removal rates were similar to TN removal rates at both sites, with high and stable removal efficiency (exceeding 95% for RW and above 99% for
LL) in the adaptation period, followed by a decrease in the later seasons (Figs. 3 and 6). Organic nitrogen removal was quite stable during the adaptation period at both MTWs, opposing to the subsequent seasons, when considerable fluctuations were observed (Figs. 4 and 7). The post-adaptation period performance was different for each MTW. Ammonium nitrogen and organic nitrogen were removed with considerably lower efficiency in the LL MTW than in the RW MTW. The outflow concentrations of nitrates (V) were higher in the LL MTW than in the RW MTW. The deterioration of removal efficiency after the first season of operation, which obviously took place at both investigation sites, requires some considerations. Interestingly, very high removal efficiency was obtained in LL treatment site during the adaptation
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model
ARTICLE IN PRESS
ECOENG-4317; No. of Pages 7
E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
Inflow
5
Outflow
120
adaptation period
-
N-NO3 [mg/l]
100 80 60 40 20
02.12.11
11.10.11
10.09.11
21.09.11
28.08.11
03.09.11
11.08.11
18.08.11
21.07.11
31.07.11
11.07.11
30.06.11
03.06.11
20.06.11
03.11.10
29.05.11
10.10.10
12.09.10
27.09.10
18.08.10
29.08.10
10.08.10
16.07.10
02.08.10
01.07.10
07.07.10
20.06.10
25.05.10
07.06.10
27.10.09
10.11.09
30.09.09
15.10.09
02.09.09
10.09.09
24.08.09
18.08.09
28.07.09
12.08.09
06.07.09
14.07.09
14.06.09
20.06.09
0
Fig. 8. Concentrations of nitrates in LL MTW.
Inflow RW
Outflow RW
10
adaptation period
6
-
N-NO3 [mg/l]
8
4
29.08.10
18.08.10
10.08.10
02.08.10
16.07.10
07.07.10
01.07.10
20.06.10
07.06.10
25.05.10
10.09.09
02.09.09
24.08.09
18.08.09
12.08.09
28.07.09
14.07.09
06.07.09
20.06.09
0
14.06.09
2
Fig. 9. Concentrations of nitrates in RW MTW.
Fig. 10. Extreme value distribution curves of total nitrogen during adaptation period (a), after adaptation period (b) and altogether in the whole investigation period (c).
period, though VSSF beds were flooded during this period. Nitrification was not likely to occur under such circumstances due to limited oxygen transfer. This is supported by low nitrates concentrations measured during the adaptation period (Fig. 8), though according to Nivala et al. (2007) simultaneous nitrification and denitrification can take place in the root matrix due to the development of aerobic and anoxic microzones. However, the adsorption process seems to explain better good ammonium nitrogen removal observed at LL and RW treatments sites in the first operation season and its decrease in the later
operation, due to the saturation of the available attachment sites. According to Reed et al. (1995) the ammonium nitrogen removal is often significantly higher in the first two years of operation than in the later period due to adsorption to the substrate and the intensive uptake by developing plants. Gray et al. (2000) considered substrate adsorption to be an important pathway for ammonium nitrogen removal in TW with the most important processes of ion exchange and physicochemical adsorption to substrate. Zhu et al. (2011a, b) studied kinetic adsorption of ammonium nitrogen by 8 different substrates and compared their efficiency of ammonium removal
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model ECOENG-4317; No. of Pages 7 6
ARTICLE IN PRESS E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
and reported the intermediate sorption capacity of gravel among other analyzed substrates. Another finding of this study was that the adsorption of ammonium nitrogen by 8 analyzed substrates increased with initial NH4 + -N concentration (the initial concentrations of 100 mg/l and 1500 mg/l were compared). According to Kadlec and Wallace (2009) ammonium nitrogen adsorbed by the substrate is loosely bound and could be released when water chemistry changes or is oxidized to nitrate, which is not bound to the substrate and is washed out. After the adaptation period VSSF beds in the LL MTW were operated in batch mode, enabling oxygen transfer. Also, the population of bacteria had already developed at this stage, which has usually required some time. This has created adequate conditions for the nitrification process evidenced by the presence of nitrates in the outflow from the VSSF beds at both investigation sites. In the last stage of treatment, HSSF bed, denitrification of nitrates took place. However, in the LL MTW denitrification was incomplete and nitrates were present in the outflow, probably due to the insufficient supply of labile OM (Faulwetter et al., 2009; Wojciechowska, 2015). Similar situations were previously reported for other TWs treating LL (Kozub and Liehr, 1999; Hamersley and Howes, 2002; Rustige and Nolde, 2006; Nivala et al., 2007). In case of RW MTW the nitrogen removal processes seemed to be more stable in time and adsorption processes were smoothly replaced by other ones. In contrast to LL RW contained higher concentrations of labile OM in the form of BOD5 , which could favor the denitrification process in MTW. Moreover, according to Gajewska et al. (2015) the OM inflowing to TWs with wastewater is captured in the beds and slowly degraded. Then, smaller and more easily biodegradable molecules that can act as electron donors for denitrification are produced. Also the plant litter decay contributes to increasing availability of carbon (Reed et al., 1995; Masi, 2008). Thus, carbon supply was not the limiting factor for denitrification at RW MTW. Interestingly, after the adaptation period nitrate (V) concentration dropped down below 4 mg/l (Fig. 9). At the same time the removal efficiency of ammonium was only a bit smaller in comparison to the adaptation period and, what is interesting, rather stable in time. Most likely simultaneous nitrification and denitrification could occur, which is a very common process in TWs (Kadlec and Wallace, 2009; Saeed and Sun, 2012). Possibly, anaerobic ammonium oxidation (anammox) could be responsible for the partial removal of ammonium, though neither anammox bacteria detection nor anammox activity analyses were performed within the study. Anammox has much lower carbon and oxygen requirements than conventional routes of nitrification and denitrification (Kadlec and Wallace, 2009; Dong and Sun, 2007). In some literature reports (Bishay and Kadlec, 1995) carbon supply was considered too low to support traditional denitrification or losses of ammonium would have required much higher oxygen transfer (Tanner et al., 2002), what could be explained by the alternative pathway of ammonium oxidation. In the recent years several studies have confirmed that the anammox process actually exists in TWs (Dong and Sun, 2007; Zhu et al., 2010; Zhu et al., 2011a, b; Tao et al., 2012), though according to Coban et al. (2015), the importance of anammox process in nitrogen transformations in HSSF wetlands appears to be low. 3.3. Evaluation of reliability of nitrogen removal processes Although the efficiency of TN removal, especially in RW MTW, was relatively good, the outflow concentrations were still quite high. At RW MTW the average TN concentration was equal to 123 mg/l, similar to TN concentration of raw municipal wastewater (120–130 mg/l), allowing for safe recirculation of RW MTW efflu´ In case ent to the mainstream of wastewater in WWTP Gdansk. of treated LL, the average TN concentration was higher (190 mg/l), however, due to low LL flow rates, the co-treatment at municipal
WWTP would have limited impact on the biological treatment processes (Fudala-Ksiazek et al., 2014). The analysis of the empirical cumulative distribution functions (Fig. 10) allows for assessment of the reliability of nitrogen removal processes at multistage TWs. The graphs present the outflow TN concentrations at both MTWs during the adaptation period (a), after the adaptation period (b) and altogether in the whole investigation period (c). During the adaptation period adsorption assured very effective TN removal. In 60% of samples TN concentration was below 15 mg/l and in 90% of samples it was below 80 mg/l. After the adaptation period, the adsorption capacity depleted, however, nitrification and denitrification microflora had already developed. These processes appeared to be less effective in TN removal from wastewater with low labile OM concentration. In 60% of outflow samples TN concentration was 165 mg/l, while in 90% of samples it was 200 mg/l. In the whole investigation period, in 90% of samples TN concentration was below 200 mg/l and in 60% of samples it was 145 mg/l. This indicates that TN outflow concentrations are at the level that allows for recirculation of treated RW to the mainstream of wastewater, as well as the co-treatment of treated LL in municipal WWTP without a risk for biological treatment processes. However, LL on-site treatment would require further processing to remove TN, as well as recalcitrant OM.
4. Conclusions Various nitrogen removal mechanisms can play different roles during the adaptation period and during the operation of mature SSF MTWs. In this study, during the initial operation period of the newly constructed LL and RW MTWs, adsorption of ammonium nitrogen seemed to be a major removal process, probably supported by plant uptake. In the next years of operation of LL pilot, TN removal was less effective, which is considered to be linked with the depletion of adsorption capacity. Thus, removal rates obtained during the initial operation periods could be considerably higher than long-term removal rates of treatment wetlands. In case of RW MTW the total nitrogen removal efficiency also decreased after the adaptation period, but it remained on considerably higher level than in case of LL MTW. Thanks to the higher concentration of labile OM denitrification of nitrates was quite effective in the RW MTW. Most likely simultaneous nitrification and denitrification was responsible for total nitrogen removal in RW MTW. In the whole investigation period TN outflow concentrations of both, LL and RW MTWs were at the level that allows for co-treatment of treated wastewater in the municipal WWTP without any risk for biological treatment processes. Further research on nitrogen transformation processes (conventional and alternative) in constructed wetlands treating high-strength wastewater should bring more information to understand and optimize the removal mechanisms.
Acknowledgements Funding support from the EEA Financial Mechanism (PL0085 and N N523 425237) and a Ministry of Science and Higher Education in Poland (E007/P01/2007/01) is gratefully acknowledged.
References APHA, 2005. American Public Health Association, Standard methods for the examination of water and wastewater, 21st Ed. Washington, DC. Bishay, L., Kadlec, R., 1995. Wetland treatment at musselwhite mine. In: Vymazal, J. (Ed.), Natural and Constructed Wetlands: Nutrients, Metals and Management. Backhuys Publishers, Leiden, The Netherlands, pp. 176–198.
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006
G Model ECOENG-4317; No. of Pages 7
ARTICLE IN PRESS E. Wojciechowska et al. / Ecological Engineering xxx (2016) xxx–xxx
Coban, O., Kuschk, P., Kappelmeyer, U., Spott, O., Martienssen, M., Jetten, M.S.M., Knoeller, K., 2015. Nitrogen transforming community in a horizontal subsurface-flow constructed wetland. Water Res. 74, 203–212. Crites, R.W., Middlebrooks, E.J., Reed, S.C., 2006. Natural Wastewater Treatment Systems. Taylor & Francis Group, New York, pp. 552. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Dong, Z., Sun, T., 2007. A potential new process for improving nitrogen removal in constructed wetlands −promoting coexistence of partial-nitrification and ANAMMOX. Ecol. Eng. 31, 69–78. 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. Ecol. Eng. 35, 987–1004. Fudala-Ksiazek, S., Luczkiewicz, A., Fitobor, K., Olanczuk-Neyman, K., 2014. Nitrogen removal via the nitrite pathway during wastewater co-treatment with ammonium-rich landfill leachates in a sequencing batch reactor. Environ. Sci. Pollut. Res. 21 (2014), 7307–7318. Gajewska, M., Obarska-Pempkowiak, H., 2011. The role of SSVF and SSHF beds in concentrated wastewater treatment, design recommendation. Wat. Sci. Tech. 64, 431–439. ´ Gajewska, M., Józwiakowski, K., Ghrabi, A., Masi, F., 2015. Impact of influent wastewater quality on nitrogen removal rates in multistage treatment wetlands. Environ. Sci. Pollut. Res. 22 (17) (12840-1284). Gray, S., Kinross, J., Read, P., Marland, A., 2000. The nutrient assimilative capacity of maerl as a substrate in constructed wetland systems for waste treatement. Water Res. 34, 2183–2190. Hamersley, M.R., Howes, B.L., 2002. Control of denitrification in a septage-treating artificial wetland: the dual role of particulate organic carbon. Wat. Res. 36, 4415–4427. Johnson, K.D., Martin, C.D., Moshiri, G.A., McCrory, W.C., 1999. Performance of constructed wetland leachate treatment system at Chunchula Landfill, Mobile County, Alabama. In: Mulamoottil, G., Mc Bean, E.A., Rovers, F. (Eds.), Constructed Wetlands for the Treatment of Landfill Leachates. Lewis Publishers, Boca Raton, Florida, USA, pp. 57–70. Kadlec, R.H., Wallace, S.D., 2009. Treatment Wetlands, 2009., 2nd edn. CRC Press, Taylor & Francis Group, Boca Raton, London, New York. Kadlec, R.H., Zmarthie, L.A., 2010. Wetland treatment of leachate from a closed landfill. Ecol. Eng 36, 946–957. Kadlec, R.H., 2003. Integrated natural systems for landfill leachate treatment. In: Vymazal, J. (Ed.), Wetlands—Nutrients, Metals and Mass Cycling. Backhuys Publishers, Leiden, The Netherlands, pp. 1–33. Kang, K.-H., Shin, H.S., Park, H., 2002. Characterization of humic substances present in landfill leachates with different landfill ages and its implications. Water Res. 36, 4023–4032. Kinsley, C.B., Crolla, A.M., Kuyucak, N., Zimmer, M., Lafléche, A., 2006. Nitrogen dynamics in a constructed wetland system treating landfill leachate. Lisbon, Portugal In: Proc. of 10th International Conference on Wetland Systems for Water Pollution Control September 23–29 2006, 2006, pp. 295–305. Kozub, D.D., Liehr, S.K., 1999. Assesssing denitrification rate limiting factors in a constructed wetland receiving landfill leachate. Water Sci. Tech 40, 75–82. Lavrova, S., Koumanova, B., 2010. Influence of recirculation in a lab-scale vertical-flow constructed wetland on the treatment efficiency of landfill leachate. Bioresour. Technol. 101 (2010), 1756–1761. Makinia, J., Stensel, H.D., Czerwionka, K., Drewnowski, J., Zapero, D., 2009. Nitrogen transformations and mass balances in anaerobic/anoxic/aerobic batch experiments with full-scale biomasses from BNR activated sludge systems. Water Sci. Technol. 60 (9), 2463–2470. Masi, F., Caffaz, S., Ghrabi, A., 2013. Multi-stage constructed wetlands systems for municipal wastewater treatment. Water Sci. Technol. 67, 1590–1598. Masi, F., 2008. Enhanced denitrification by an hybrid HF-FWS CW in large scale wastewater treatment plant (Jesi). In: Vymazal, J. (Ed.), Wastewater Treatment, Plant Dynamics and Management in Constructed and Natural Wetlands. Springer, New York, pp. 267–275.
7
Miksch, K., Sikora, J., (2010) Biotechnologia s´ cieków. Wydawnictwo Naukowe PWN. Warszawa, (Biotechnology of wastewater), 216–224. Molle, P., Prost-Boucle, S., Lienard, A., 2008. Potential for total nitrogen removal by combining vertical flow and horizontal flow constructed wetlands: a full-scale experiment study. Ecol. Eng. 34, 23–29. Nivala, J., Hoos, M.B., Cross, C., Wallace, S., Parkin, G., 2007. Treatment of landfill leachate using an aerated, horizontal subsurface-flow constructed wetland. Sci. Total Environ. 380, 19–27. Obarska-Pempkowiak, H., Gajewska, M., Wojciechowska, E., 2010. Application of vertical flow constructed wetlands for highly contaminated wastewater treatment: preliminary results. In: Vymazal, J. (Ed.), Water and Nutrient Management in Natural and Constructed Wetlands, 2010. Springer, London, pp. 37–51. Obarska-Pempkowiak, H., Gajewska, M., Wojciechowska, E., Pempkowiak, J., 2015. ´ Treatment wetlands for environmental pollution control. In: Rowinski, P. (Ed.), GeoPlanet: Earth and Planetary Sciences. Springer International Publishing, p. 169. Reed, S.C., Crites, R.W., Middlebrooks, E.J., 1995. Natural Systems for Waste Management and Treatment, 1995., 2nd edn. McGraw-Hill, Inc., New York, pp. 198–199. Rustige, H., Nolde, E., 2006. Nitrogen elimination from landfill leachates using an extra carbon source in subsurface flow constructed wetlands. Lisbon, Portugal In: Proc. of 10th International Conference on Wetland Systems for Water Pollution Control September 23–29 2006, 2006, pp. 229–239. Saeed, T., Sun, G., 2012. A review on nitrogen and organics removal mechanisms in subsurface flow constructed wetlands: dependency on environmental parameters, operating conditions and supporting media. J Environ Manag 112, 429–448. Tanner, C.C., Kadlec, R., Gibbs, M.M., Sukias, J.P., Nguyen, M.L., 2002. Nitrogen processing gradients in subsurface-flow treatment wetlands: influent wastewater characteristics. Ecol. Eng. 18, 499–520. Tao, W., He, Y., Wang, Z., Smith, R., Shayya, W., Pei, Y., 2012. Effects of pH and temperature on coupling nitrification and anammox in biofilters treating dairy wastewater. Ecol. Eng. 47, 76–82. Vymazal, J., 2009. The use of constructed wetlands with horizontal sub-surface flow for various types of wastewater. Ecol. Eng. 35, 1–17. Waara, S., Waara, K.-O., Forsberg, Å., Fridolfsson, M., 2008. An evaluation of the performance of a constructed wetland system for treatment of landfill leachate during 2003–2006. Warwickshire, England, 16–17 September In: Proc. of Waste 2008: Waste and Resource Management—a Shared Responsibility Stratford-upon-Avon, 2008. Wiesman, U., 1994. Biological nitrogen removal from wastewater. Adv. Biochem. Eng. Biotechnol. 51 (Spinger Verlag, Berlin, Heidelberg). Wojciechowska, E., Gajewska, M., 2013. Partitioning of heavy metals in sub-surface flow treatment wetlands receiving high-strength wastewater. Water Sci. Technol. 68, 486–493. Wojciechowska, E., 2015. Removal of nitrogen compounds from landfill leachate in pilot constructed wetlands. Annu. Set Environ. Prot. 17 (2), 1484–1497 (in Polish). Yalcuk, A., Ugurlu, A., 2009. Comparison of horizontal and vertical constructed wetland systems for landfill leachate treatment. Bioresour. Technol. 100 (2009), 2521–2526. Zhu, G.B., Jetten, M.S.M., Kuschk, P., Ettwig, K.F., Yin, C.Q., 2010. Potential roles of anaerobic ammonium and methane oxidation in nitrogen cycle of wetland ecosystem. Appl. Microbiol. Biotechnol. 86, 1043–1055. Zhu, W.L., Cui, L.H., Ouyang, Y., Long, C.F., Tang, X.D., 2011a. Kinetic adsorption of ammonium nitrogen by substrate materials for constructed wetlands. Pedosphere 21, 454–463. Zhu, G.B., Wang, S.Y., Feng, X.J., Fan, G.N., Jetten, M.S.M., Yin, C.O., 2011b. Anammox bacterial abundance, biodiversity and activity in a constructed wetland. Environ. Sci. Technol. 45 (9).
Please cite this article in press as: Wojciechowska, E., et al., Reliability of nitrogen removal processes in multistage treatment wetlands receiving high-strength wastewater. Ecol. Eng. (2016), http://dx.doi.org/10.1016/j.ecoleng.2016.07.006