- Email: [email protected]
Nitrogen transformation in horizontal and vertical flow constructed wetlands applied for dairy cattle wastewater treatment in southern Brazil
Ecological Engineering 73 (2014) 307–310
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Short communication
Nitrogen transformation in horizontal and vertical flow constructed wetlands applied for dairy cattle wastewater treatment in southern Brazil Catiane Pelissari, Pablo Heleno Sezerino * , Samara Terezinha Decezaro, Delmira Beatriz Wolff, Alessandra Pellizzaro Bento, Orlando de Carvalho Junior, Luiz Sérgio Philippi Department of Sanitary and Environmental Engineering, Federal University of Santa Catarina, Trindade, Florianópolis, Santa Catarina CEP 88040-900, Brazil
A R T I C L E I N F O
A B S T R A C T
Article history: Received 10 March 2014 Received in revised form 3 September 2014 Accepted 26 September 2014 Available online xxx
The aim of this study was to evaluate the nitrogen transformations in horizontal and vertical flow constructed wetlands (HFCW and VFCW), working in parallel and applied for dairy cattle wastewater treatment. Both HFCW (26.50 m2 of surface area) and VFCW (14.30 m2 of surface area) were filled up with sand (d10 of 0.3 mm and uniformity coefficient of 2.50) as bed media and planted with Typha domingensis Pers. HFCW and VFCW worked with an influent flow rate of 3.98 m3 week1 and 4.50 m3 week1, respectively. Applying an average loading rate of 151.4 gCOD m2 week1, 10.3 gTKN m2 week1 and 8.2 gNH4+-N m2 week1 in HFCW, it was possible to achieve 59% of TN and 58% of NH4+-N removals. In VFCW an average loading rate of 317.2 gCOD m2 week1, 21.6 gTKN m2 week1 and 13.7 gNH4+-N m2 week1 were applied and was obtained 23% of TN and 80% NH4+-N removals, where 73% of ammonia removal was due to nitrification process. The macrophytes removed 5.1% and 0.88% of influent N loading rate in HFCW and VFCW, respectively. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Dairy cattle Wastewater Constructed wetlands Nitrogen Macrophytes
1. Introduction In constructed wetlands, transformations of nitrogen occur in diverse metabolic pathways. According to Saeed and Sun (2012), traditional forms of nitrogen transformation and removal are associated with ammonification, nitrification, denitrification, volatilization, uptake by macrophytes, adsorption by the filtration media, and uptake by microbial biomass. The various performance levels of constructed wetlands in terms of nitrogen removal is associated with the following conditions: (i) low availability of dissolved oxygen in the medium, thus preventing nitrification due to high concentrations of organic compounds (Platzer, 1999); and (ii) lack of organic carbon available for the denitrification process (Zhai et al., 2013). Since, the magnitude of nitrogen transformations in constructed wetlands is complex and largely associated with environmental and operational factors, such as influent hydraulic and organic loadings, characteristics of hydraulic flows, time of
* Corresponding author. Tel: +55 48 37212606. E-mail addresses: [email protected], [email protected] (P.H. Sezerino). http://dx.doi.org/10.1016/j.ecoleng.2014.09.085 0925-8574/ ã 2014 Elsevier B.V. All rights reserved.
hydraulic retention and use of macrophytes, there is a need for more in-depth studies to understand the importance of each of the possible nitrogen metabolic pathways in these systems. Given the above, this study aimed to assess the transformations of nitrogen forms in horizontal flow constructed wetlands (HFCW) and vertical flow constructed wetlands (VFCW) being operated simultaneously with the system of dairy wastewater treatment under southern Brazil climate conditions. 2. Material and methods The wetland systems were implemented in dairy facilities in the city of Frederico Westphalen, southern Brazil (latitude 27 21033” south and longitude 53 230 40” west, altitude 566 m), producing on average 140 L of milk per day. All wastewater produced in the milking parlor flowed down by gravity into a storage pond (SP) which served as primary treatment. After going through this unit the wastewater flowed into an equalizer tank, from which a fraction was then transported by gravity into the HFCW, and the other portion pumped into the VFCW. The wetlands operated in parallel for the purposes of comparison of performance of both systems. Fig. 1 presents the layout of the studied treatment plant.
308
C. Pelissari et al. / Ecological Engineering 73 (2014) 307–310
VFCW), the analysis of variance–ANOVA was applied using software MicrosoftTM Excel. 3. Results and discussion Table 1 describes the means and standard deviation (SD) measured in the wetlands influent (after the equalizer tank) and effluent, as well the average organic and inorganic loadings applied to the HFCW and VFCW. Fig. 1. Schematic layout of the treatment plant.
3.1. Nitrogen removal and transformation in the HFCW The HFCW (6.70 3.95 0.80 m; length x width x usable depth, and a total surface area of 26.50 m2) operated with an average flow of 3980 L week1, being fed by gravity four times a week, with a daily flow of 995 L during four hours per day at an average hydraulic rate of 37.55 mm d1. The VFCW (4.40 3.25 0.80 m; length x width x usable depth, and a total surface area of 14.30 m2) was fed intermittently assisted by a pumping system, with an average flow of 4500 L week1, which was distributed in 4 pulses of 375 L d1, 3 days a week, at an average hydraulic rate of 105 mm d1. Both HFCW and VFCW were filled up with sand (d10 of 0.3 mm and uniformity coefficient of 2.50) as bed media and planted with Typha domingensis Pers. To quantify nitrogen transformations and removals in the wetlands, the contents of nitrogen in the macrophytes’ foliar tissue were analyzed as well as the concentrations of the forms of nitrogen in the influent and effluent of each unit. The quantification of nitrogen concentrations incorporated to the plant tissue of macrophyte Typha domingensis Pers. was achieved by pruning the macrophytes to a height of 30 cm above the filter bed and by monitoring the plants growth. The analysis of N concentrations in the foliar tissue was performed as recommended by Tedesco et al. (1995) in different stages of the macrophytes growth. The wetlands influents and effluents were monitored over one year (November/2011–October/2012) totaling 35 samples (weekly frequency) in three sampling sites, namely, after the equalizer tank, after the HFCW and after the VFCW. Evaluated parameters were pH, Alkalinity, Total Kjeldhal Nitrogen (TKN), Chemical Oxygen Demand (COD), Ammonium Nitrogen (NH4+-N) and Nitrate Nitrogen (NO3–N), as recommended by Standard Methods (APHA, 2005), except for ammonium nitrogen, which was evaluated according to Vogel (1981). To determine possible differences in the performances of both treatments (HFCW and
For an average loading of 8.2 gNH4+-N m2 week1(Table 1), an average removal of 58% of ammonium nitrogen was achieved. Since nitrification was not apparent in this unit and that ammonia volatilization can be disregarded once the influent pH was close to neutral, it may be inferred that the removal of ammonium nitrogen was basically associated with the following mechanisms: (i) uptake by macrophytes; (ii) uptake by the bacterial biomass; (iii) adsorption to the filter medium. There was a variation in the ammonium nitrogen removal in the HFCW over the monitoring period (Fig. 2). The highest removal rate occurred when the HFCW operated with the largest loading of ammonium nitrogen (21.6 gNH4+-N m2 week1), which coincided with the period following pruning, attaining mean removal efficiencies as high as 74%, compared to 32% of removal before pruning. It should be noted that from the 20th sampling, corresponding to the winter period, there was a decrease of the influent concentration of ammonium nitrogen. During this period, the mean loading dropped to 7.8 gNH4+-N m2, and a mean removal of 85% was achieved. Vymazal (2007) reports that macrophytes can perform a considerable removal of nitrogen when pruning is made regularly. The mean nitrogen levels in the foliar tissue of macrophyte Typha domingensis Pers. were 25.6 g kg1, considering a growth period of 150 days. Taking into account a total of 850 plants in the HFCW (density of 50 plants m2 of planted filter) and that Typha domingensis Pers. incorporates 7.35 g of dry matter per growth meters, removals of nitrogen corresponded to 0.3 kg N of 6 kg N applied, that represents 1.07 g m2 week1. Therefore, the macrophytes are responsible by 5.1% removal of the total nitrogen loading applied to the HFCW. Low concentrations of nitrate formed may be related to low concentrations of oxygen available in the HFCW, since it was operated in such a way that the filtration medium and the rhizosphere remained saturated with the wastewater undergoing
Table 1 Average values of HFCW and VFCW characteristics and ANOVA statistics. Parameters
Applied loading rate HFCW
*
HFCW effluent
VFCW effluent
18 4 7.2 0.4 670 3 1,009 298
20 4 6.4 0.2 455 300 262 84
20 4 6.9 0.2 290 139 323 101
69 30
28 15
20 10
0.03201
55 27
23 21
11 10
0.00516
54
31
37 14
2.40 1022
VFCW
T ( C) pH Alkalinity (mgCaCO3 L1) COD (mg L1)* 2 COD (g m week1) TKN (mg L1)* 2 TKN (g m week1) NH4+-N (mg L1) 2 NH4+-N (g m week1) NO3 -N (mg L1)
p-value (HFCW–VFCW)**
Influent
151.4
317.2
10.3
21.6
8.2
17.3
23 Samples. Statistical result between concentrations in mg L1 for the constructed wetlands: 5% significance level (a = 0.05); p = significance level (p > a accepts H0; p < a rejects H0). H0 = Null hypothesis: there is no difference between the arithmetical means of the groups; H1 = Alternative hypothesis: there is difference between the arithmetical means of the groups samples. **
C. Pelissari et al. / Ecological Engineering 73 (2014) 307–310
309
Fig. 2. Concentrations of ammonium nitrogen in the HFCW influent and effluent, and concentrations of nitrate nitrogen in the HFCW effluent over the monitoring period.
treatment. Zhang et al. (2011) evaluated the performance of three HFCW operating with loads of 15.4 gNH4+-N m2 week1, leading to forced aeration of 4 mg L1 of dissolved oxygen over the studied period. The authors found a large accumulation of nitrite, which reached concentrations of up to 22.7 mg L1, concluding that partial nitrification occurred in these modules as a function of oxygen added to the medium. Denitrification is largely dependent on the source of organic carbon. However, in this study the C:N ratio was higher than 2.3. Therefore, organic carbon was not a limiting factor for the occurrence of denitrification. Since, the concentrations of influent nitrate nitrogen in the HFCW were low, denitrification in this module was not the main way of nitrogen removal, even with the HFCW operating with available source of carbon. Regarding the HFCW, a mean removal rate of 59% of the influent TN was observed. Based on the results, since there was no oxidation of ammonium nitrogen, it was noticed that 5% of nitrogen was absorbed by the macrophytes, 55% partially by the sum of microorganisms assimilation, filter medium adsorption and also denitrification, and 40% remained as residual TKN. 3.2. Nitrogen removal and transformation in the VFCW For average loadings of 17.3 gNH4+-N m2 week1 and 21.6 gTKN m2 week1, an average removal of 80% of ammonium nitrogen
was achieved. Nitrification was the main pathway of ammonium nitrogen transformation and utilization in the VFCW and 73% of the ammonium nitrogen was converted into nitrate nitrogen. Nitrification in this wetland unit was facilitated by the reduced average load used (21.6 gTKN m2 week1, see Table 1), corroborating Platzer’s recommendations (1999), which indicate a maximum value of 6.5 gTKN m2 d1 (45.5 gTKN m2 week1) for nitrification to be effective in constructed wetlands using sand as filter medium. One of the key factors that contributed to the occurrence of nitrification, in addition to the good adaptation of the nitrifying micro biota, was the effective transference of oxygen to the filter bed due to intermittent feeding. Average nitrogen levels in the foliar tissue of macrophyte Typha domingensis Pers. over a 100 day growth period were 27.7 g kg1. Considering 220 plants in the VFCW (density of 15 plants m2), average nitrogen removal was 0.18 g m2 week1 in the VFCW. The macrophytes accounted for a removal of only 0.88% of the nitrogen load to the VFCW. VFCW has presented an average TN removal of 23%. Based on this performance, the nitrogen oxidation represented an average of 47%. Adsorption by the filtration medium, microorganisms and also by denitrification reached 24% in average, and 28% remained as residual TKN.
Fig. 3. Average proportions of different nitrogen forms in the treatments studied (a) Influent; (b) HFCW effluent; (c) VFCW effluent.
310
C. Pelissari et al. / Ecological Engineering 73 (2014) 307–310
3.3. Average proportions of nitrogen forms in the HFCW and VFCW The ANOVA test showed that significant differences occurred between the units under study (Table 1), when compared to the effluent concentrations of the nitrogen forms. In general, nitrogen transformations occurred in different pathways and magnitudes in the systems. In an analysis of Fig. 3, it can be seen that the distribution of the nitrogen forms is quite different in the two constructed wetlands under study. NH4+-N is the predominant form of nitrogen in the HFCW influent and effluent, followed by organic nitrogen, indicating low oxygen concentration in the system and the predominance of a reducing environment. In contrast, in the VFCW it can be seen that the most oxidized form of nitrogen in the liquid phase is the most abundant form, indicating occurrence of expressive nitrification achieved by aeration provided by intermittent feeding of the unit. The better nitrification efficiency of VFCW compared to HFCW is related to controlled intermittent feeding (4 pulses per day in the VFCW–free drainage) which allow better oxygen input in the media. Was obtained in average a positive balance of 145 g O2 pulse1 by Platzer’s oxygen dimensioning model (Platzer, 1999). It was observed that in the HFCW the average percentage of organic nitrogen increased by 25%, when compared to the influent. This might have occurred by detachment of the biofilm that formed in the rhizosphere and in the filter medium as well as by decomposition of the macrophytes roots. 4. Conclusion 1. By operating the HFCW with an average loading of 154 gCOD
m2 week1, 10.3 gTKN m2 week1 and 8.2 gNH4+-N m2 week1, an average removal of 59% of total nitrogen and 58% of ammonium nitrogen was attained. Nitrification was not observed; 2. By operating the VFCW with an average loading of 317.2 gCOD m2 week1, 21.6 gTNK m2 week1 and 17.3 gNH4+-N m2
week1, an average removal of 23% of total nitrogen and 80% of ammonium nitrogen was observed; 3. Nitrification was the main removal mechanism of ammonium nitrogen in the VFCW, responsible for removing 73% of the influent ammonium nitrogen; 4. To the HFCW the average nitrogen levels in the foliar tissue of macrophyte Typha domingensis Pers. were 25.6 g kg1, considering a growth period of 150 days. In the VFCW the average nitrogen levels in the foliar tissue over a 100 days growth period were 27.7 g kg1;
Acknowledgement The authors would like to thank the National Council of Scientific and Technological Development (CNPq) for funding this research. References APHA, AWWA, WEF, 2005. Standard methods for the examination of water and wastewater. 21st ed. Washington. Platzer, C., 1999. Design recommendation for subsurface flow constructed wetlands for nitrification and denitrification. Water Sci. Technol. 40 (3), 257–263. 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. Manage. 112 (15), 429–448. M.J. Tedesco C. Gianello C.A. Bissani H. Bohnen S.J. Volkweiss Análise de solo, plantas e outros materiais Porto Alegre, Universidade Federal do Rio Grande do Sul 174 p. 1995 (in portuguese). A.I. Vogel Química analítica qualitativa. 5. ed, editora Mestre Jou São Paulo, p.665 1981 (in portuguese). Vymazal, J., 2007. Removal of nutrients in various types of constructed wetlands. Sci. Total Environ. 380 (1–3), 48–65. Zhai, X., Piwpuan, N., Arias, A.C., Headley, T., Brix, H., 2013. Can root exudates from emergent wetland plants fuel denitrification in subsurface flow constructed wetland systems? Ecol. Eng. 61 (Part B), 555–563. Zhang, L., Xia, X., Zhao, Y., Xi, B., Yan, Y., Guo, X., Xiong, Y., Zhan, J., 2011. The ammonium nitrogen oxidation process in horizontal subsurface flow constructed wetlands. Ecol. Eng. 37 (11), 1614–1619.
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Short communication
Nitrogen transformation in horizontal and vertical flow constructed wetlands applied for dairy cattle wastewater treatment in southern Brazil Catiane Pelissari, Pablo Heleno Sezerino * , Samara Terezinha Decezaro, Delmira Beatriz Wolff, Alessandra Pellizzaro Bento, Orlando de Carvalho Junior, Luiz Sérgio Philippi Department of Sanitary and Environmental Engineering, Federal University of Santa Catarina, Trindade, Florianópolis, Santa Catarina CEP 88040-900, Brazil
A R T I C L E I N F O
A B S T R A C T
Article history: Received 10 March 2014 Received in revised form 3 September 2014 Accepted 26 September 2014 Available online xxx
The aim of this study was to evaluate the nitrogen transformations in horizontal and vertical flow constructed wetlands (HFCW and VFCW), working in parallel and applied for dairy cattle wastewater treatment. Both HFCW (26.50 m2 of surface area) and VFCW (14.30 m2 of surface area) were filled up with sand (d10 of 0.3 mm and uniformity coefficient of 2.50) as bed media and planted with Typha domingensis Pers. HFCW and VFCW worked with an influent flow rate of 3.98 m3 week1 and 4.50 m3 week1, respectively. Applying an average loading rate of 151.4 gCOD m2 week1, 10.3 gTKN m2 week1 and 8.2 gNH4+-N m2 week1 in HFCW, it was possible to achieve 59% of TN and 58% of NH4+-N removals. In VFCW an average loading rate of 317.2 gCOD m2 week1, 21.6 gTKN m2 week1 and 13.7 gNH4+-N m2 week1 were applied and was obtained 23% of TN and 80% NH4+-N removals, where 73% of ammonia removal was due to nitrification process. The macrophytes removed 5.1% and 0.88% of influent N loading rate in HFCW and VFCW, respectively. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Dairy cattle Wastewater Constructed wetlands Nitrogen Macrophytes
1. Introduction In constructed wetlands, transformations of nitrogen occur in diverse metabolic pathways. According to Saeed and Sun (2012), traditional forms of nitrogen transformation and removal are associated with ammonification, nitrification, denitrification, volatilization, uptake by macrophytes, adsorption by the filtration media, and uptake by microbial biomass. The various performance levels of constructed wetlands in terms of nitrogen removal is associated with the following conditions: (i) low availability of dissolved oxygen in the medium, thus preventing nitrification due to high concentrations of organic compounds (Platzer, 1999); and (ii) lack of organic carbon available for the denitrification process (Zhai et al., 2013). Since, the magnitude of nitrogen transformations in constructed wetlands is complex and largely associated with environmental and operational factors, such as influent hydraulic and organic loadings, characteristics of hydraulic flows, time of
* Corresponding author. Tel: +55 48 37212606. E-mail addresses: [email protected], [email protected] (P.H. Sezerino). http://dx.doi.org/10.1016/j.ecoleng.2014.09.085 0925-8574/ ã 2014 Elsevier B.V. All rights reserved.
hydraulic retention and use of macrophytes, there is a need for more in-depth studies to understand the importance of each of the possible nitrogen metabolic pathways in these systems. Given the above, this study aimed to assess the transformations of nitrogen forms in horizontal flow constructed wetlands (HFCW) and vertical flow constructed wetlands (VFCW) being operated simultaneously with the system of dairy wastewater treatment under southern Brazil climate conditions. 2. Material and methods The wetland systems were implemented in dairy facilities in the city of Frederico Westphalen, southern Brazil (latitude 27 21033” south and longitude 53 230 40” west, altitude 566 m), producing on average 140 L of milk per day. All wastewater produced in the milking parlor flowed down by gravity into a storage pond (SP) which served as primary treatment. After going through this unit the wastewater flowed into an equalizer tank, from which a fraction was then transported by gravity into the HFCW, and the other portion pumped into the VFCW. The wetlands operated in parallel for the purposes of comparison of performance of both systems. Fig. 1 presents the layout of the studied treatment plant.
308
C. Pelissari et al. / Ecological Engineering 73 (2014) 307–310
VFCW), the analysis of variance–ANOVA was applied using software MicrosoftTM Excel. 3. Results and discussion Table 1 describes the means and standard deviation (SD) measured in the wetlands influent (after the equalizer tank) and effluent, as well the average organic and inorganic loadings applied to the HFCW and VFCW. Fig. 1. Schematic layout of the treatment plant.
3.1. Nitrogen removal and transformation in the HFCW The HFCW (6.70 3.95 0.80 m; length x width x usable depth, and a total surface area of 26.50 m2) operated with an average flow of 3980 L week1, being fed by gravity four times a week, with a daily flow of 995 L during four hours per day at an average hydraulic rate of 37.55 mm d1. The VFCW (4.40 3.25 0.80 m; length x width x usable depth, and a total surface area of 14.30 m2) was fed intermittently assisted by a pumping system, with an average flow of 4500 L week1, which was distributed in 4 pulses of 375 L d1, 3 days a week, at an average hydraulic rate of 105 mm d1. Both HFCW and VFCW were filled up with sand (d10 of 0.3 mm and uniformity coefficient of 2.50) as bed media and planted with Typha domingensis Pers. To quantify nitrogen transformations and removals in the wetlands, the contents of nitrogen in the macrophytes’ foliar tissue were analyzed as well as the concentrations of the forms of nitrogen in the influent and effluent of each unit. The quantification of nitrogen concentrations incorporated to the plant tissue of macrophyte Typha domingensis Pers. was achieved by pruning the macrophytes to a height of 30 cm above the filter bed and by monitoring the plants growth. The analysis of N concentrations in the foliar tissue was performed as recommended by Tedesco et al. (1995) in different stages of the macrophytes growth. The wetlands influents and effluents were monitored over one year (November/2011–October/2012) totaling 35 samples (weekly frequency) in three sampling sites, namely, after the equalizer tank, after the HFCW and after the VFCW. Evaluated parameters were pH, Alkalinity, Total Kjeldhal Nitrogen (TKN), Chemical Oxygen Demand (COD), Ammonium Nitrogen (NH4+-N) and Nitrate Nitrogen (NO3–N), as recommended by Standard Methods (APHA, 2005), except for ammonium nitrogen, which was evaluated according to Vogel (1981). To determine possible differences in the performances of both treatments (HFCW and
For an average loading of 8.2 gNH4+-N m2 week1(Table 1), an average removal of 58% of ammonium nitrogen was achieved. Since nitrification was not apparent in this unit and that ammonia volatilization can be disregarded once the influent pH was close to neutral, it may be inferred that the removal of ammonium nitrogen was basically associated with the following mechanisms: (i) uptake by macrophytes; (ii) uptake by the bacterial biomass; (iii) adsorption to the filter medium. There was a variation in the ammonium nitrogen removal in the HFCW over the monitoring period (Fig. 2). The highest removal rate occurred when the HFCW operated with the largest loading of ammonium nitrogen (21.6 gNH4+-N m2 week1), which coincided with the period following pruning, attaining mean removal efficiencies as high as 74%, compared to 32% of removal before pruning. It should be noted that from the 20th sampling, corresponding to the winter period, there was a decrease of the influent concentration of ammonium nitrogen. During this period, the mean loading dropped to 7.8 gNH4+-N m2, and a mean removal of 85% was achieved. Vymazal (2007) reports that macrophytes can perform a considerable removal of nitrogen when pruning is made regularly. The mean nitrogen levels in the foliar tissue of macrophyte Typha domingensis Pers. were 25.6 g kg1, considering a growth period of 150 days. Taking into account a total of 850 plants in the HFCW (density of 50 plants m2 of planted filter) and that Typha domingensis Pers. incorporates 7.35 g of dry matter per growth meters, removals of nitrogen corresponded to 0.3 kg N of 6 kg N applied, that represents 1.07 g m2 week1. Therefore, the macrophytes are responsible by 5.1% removal of the total nitrogen loading applied to the HFCW. Low concentrations of nitrate formed may be related to low concentrations of oxygen available in the HFCW, since it was operated in such a way that the filtration medium and the rhizosphere remained saturated with the wastewater undergoing
Table 1 Average values of HFCW and VFCW characteristics and ANOVA statistics. Parameters
Applied loading rate HFCW
*
HFCW effluent
VFCW effluent
18 4 7.2 0.4 670 3 1,009 298
20 4 6.4 0.2 455 300 262 84
20 4 6.9 0.2 290 139 323 101
69 30
28 15
20 10
0.03201
55 27
23 21
11 10
0.00516
54
31
37 14
2.40 1022
VFCW
T ( C) pH Alkalinity (mgCaCO3 L1) COD (mg L1)* 2 COD (g m week1) TKN (mg L1)* 2 TKN (g m week1) NH4+-N (mg L1) 2 NH4+-N (g m week1) NO3 -N (mg L1)
p-value (HFCW–VFCW)**
Influent
151.4
317.2
10.3
21.6
8.2
17.3
23 Samples. Statistical result between concentrations in mg L1 for the constructed wetlands: 5% significance level (a = 0.05); p = significance level (p > a accepts H0; p < a rejects H0). H0 = Null hypothesis: there is no difference between the arithmetical means of the groups; H1 = Alternative hypothesis: there is difference between the arithmetical means of the groups samples. **
C. Pelissari et al. / Ecological Engineering 73 (2014) 307–310
309
Fig. 2. Concentrations of ammonium nitrogen in the HFCW influent and effluent, and concentrations of nitrate nitrogen in the HFCW effluent over the monitoring period.
treatment. Zhang et al. (2011) evaluated the performance of three HFCW operating with loads of 15.4 gNH4+-N m2 week1, leading to forced aeration of 4 mg L1 of dissolved oxygen over the studied period. The authors found a large accumulation of nitrite, which reached concentrations of up to 22.7 mg L1, concluding that partial nitrification occurred in these modules as a function of oxygen added to the medium. Denitrification is largely dependent on the source of organic carbon. However, in this study the C:N ratio was higher than 2.3. Therefore, organic carbon was not a limiting factor for the occurrence of denitrification. Since, the concentrations of influent nitrate nitrogen in the HFCW were low, denitrification in this module was not the main way of nitrogen removal, even with the HFCW operating with available source of carbon. Regarding the HFCW, a mean removal rate of 59% of the influent TN was observed. Based on the results, since there was no oxidation of ammonium nitrogen, it was noticed that 5% of nitrogen was absorbed by the macrophytes, 55% partially by the sum of microorganisms assimilation, filter medium adsorption and also denitrification, and 40% remained as residual TKN. 3.2. Nitrogen removal and transformation in the VFCW For average loadings of 17.3 gNH4+-N m2 week1 and 21.6 gTKN m2 week1, an average removal of 80% of ammonium nitrogen
was achieved. Nitrification was the main pathway of ammonium nitrogen transformation and utilization in the VFCW and 73% of the ammonium nitrogen was converted into nitrate nitrogen. Nitrification in this wetland unit was facilitated by the reduced average load used (21.6 gTKN m2 week1, see Table 1), corroborating Platzer’s recommendations (1999), which indicate a maximum value of 6.5 gTKN m2 d1 (45.5 gTKN m2 week1) for nitrification to be effective in constructed wetlands using sand as filter medium. One of the key factors that contributed to the occurrence of nitrification, in addition to the good adaptation of the nitrifying micro biota, was the effective transference of oxygen to the filter bed due to intermittent feeding. Average nitrogen levels in the foliar tissue of macrophyte Typha domingensis Pers. over a 100 day growth period were 27.7 g kg1. Considering 220 plants in the VFCW (density of 15 plants m2), average nitrogen removal was 0.18 g m2 week1 in the VFCW. The macrophytes accounted for a removal of only 0.88% of the nitrogen load to the VFCW. VFCW has presented an average TN removal of 23%. Based on this performance, the nitrogen oxidation represented an average of 47%. Adsorption by the filtration medium, microorganisms and also by denitrification reached 24% in average, and 28% remained as residual TKN.
Fig. 3. Average proportions of different nitrogen forms in the treatments studied (a) Influent; (b) HFCW effluent; (c) VFCW effluent.
310
C. Pelissari et al. / Ecological Engineering 73 (2014) 307–310
3.3. Average proportions of nitrogen forms in the HFCW and VFCW The ANOVA test showed that significant differences occurred between the units under study (Table 1), when compared to the effluent concentrations of the nitrogen forms. In general, nitrogen transformations occurred in different pathways and magnitudes in the systems. In an analysis of Fig. 3, it can be seen that the distribution of the nitrogen forms is quite different in the two constructed wetlands under study. NH4+-N is the predominant form of nitrogen in the HFCW influent and effluent, followed by organic nitrogen, indicating low oxygen concentration in the system and the predominance of a reducing environment. In contrast, in the VFCW it can be seen that the most oxidized form of nitrogen in the liquid phase is the most abundant form, indicating occurrence of expressive nitrification achieved by aeration provided by intermittent feeding of the unit. The better nitrification efficiency of VFCW compared to HFCW is related to controlled intermittent feeding (4 pulses per day in the VFCW–free drainage) which allow better oxygen input in the media. Was obtained in average a positive balance of 145 g O2 pulse1 by Platzer’s oxygen dimensioning model (Platzer, 1999). It was observed that in the HFCW the average percentage of organic nitrogen increased by 25%, when compared to the influent. This might have occurred by detachment of the biofilm that formed in the rhizosphere and in the filter medium as well as by decomposition of the macrophytes roots. 4. Conclusion 1. By operating the HFCW with an average loading of 154 gCOD
m2 week1, 10.3 gTKN m2 week1 and 8.2 gNH4+-N m2 week1, an average removal of 59% of total nitrogen and 58% of ammonium nitrogen was attained. Nitrification was not observed; 2. By operating the VFCW with an average loading of 317.2 gCOD m2 week1, 21.6 gTNK m2 week1 and 17.3 gNH4+-N m2
week1, an average removal of 23% of total nitrogen and 80% of ammonium nitrogen was observed; 3. Nitrification was the main removal mechanism of ammonium nitrogen in the VFCW, responsible for removing 73% of the influent ammonium nitrogen; 4. To the HFCW the average nitrogen levels in the foliar tissue of macrophyte Typha domingensis Pers. were 25.6 g kg1, considering a growth period of 150 days. In the VFCW the average nitrogen levels in the foliar tissue over a 100 days growth period were 27.7 g kg1;
Acknowledgement The authors would like to thank the National Council of Scientific and Technological Development (CNPq) for funding this research. References APHA, AWWA, WEF, 2005. Standard methods for the examination of water and wastewater. 21st ed. Washington. Platzer, C., 1999. Design recommendation for subsurface flow constructed wetlands for nitrification and denitrification. Water Sci. Technol. 40 (3), 257–263. 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. Manage. 112 (15), 429–448. M.J. Tedesco C. Gianello C.A. Bissani H. Bohnen S.J. Volkweiss Análise de solo, plantas e outros materiais Porto Alegre, Universidade Federal do Rio Grande do Sul 174 p. 1995 (in portuguese). A.I. Vogel Química analítica qualitativa. 5. ed, editora Mestre Jou São Paulo, p.665 1981 (in portuguese). Vymazal, J., 2007. Removal of nutrients in various types of constructed wetlands. Sci. Total Environ. 380 (1–3), 48–65. Zhai, X., Piwpuan, N., Arias, A.C., Headley, T., Brix, H., 2013. Can root exudates from emergent wetland plants fuel denitrification in subsurface flow constructed wetland systems? Ecol. Eng. 61 (Part B), 555–563. Zhang, L., Xia, X., Zhao, Y., Xi, B., Yan, Y., Guo, X., Xiong, Y., Zhan, J., 2011. The ammonium nitrogen oxidation process in horizontal subsurface flow constructed wetlands. Ecol. Eng. 37 (11), 1614–1619.