removal efficiency
RESULTS AND DISCUSSIONS
The presence and removal of BOD5 and FC are monitoring for 8.5 weeks after the stabilization of the system. No significant differences in the number of FC with respect to different sampling points, registering a gradual reduction in the point of sampling next to the output were. The system showed an efficient removal of FC 99.99% (4.81 logarithmic units) and BOD5 a 94.46% (Table 3). These results coincide with those reported by Arias et al.
(2003) in has two stages, which recorded a removal of indicator bacteria between 99.5 -
99.9%, and by Li et al. (2012) in its work related to sand filters, which recorded values of removal of FC of 99.9%.
Contrary to the proceeds by Song et al. (2006) who reported percentages less removal than those obtained in our study, since they recorded a removal of FC of 99.6%.
Table 3. Removal of FC and BOD5 for stage.
Parameter
Stage 1 Stage 2 System
complete Inlet flow Flow rate Removal
(%) Inlet flow Flow rate Removal (%)
Removal (%) FC (NMP/100 mL) 1.50 × 106 2.12 × 102 99.94 2.12 × 102 9.38 × 100 94.75 99.99
BOD5(mg/l) 151.3 19.65 86.58 19.65 8.1 58.72 94.46
On the other hand, regarding the BOD5 results are similar to those obtained in this work (94.46%) (Figure 3), for example Langergraber et al. (2014) obtained a percentage of 95% in vertical wetlands in two stages and Jong et al. (2016) obtained values of 92-96%.
Figure 3.Removal of BOD5.
During the operation of the SSVFCW were presented different concentrations of FC, which ranged from 3.38 × 105 - 3.89 × 107MPN/100 mL and BOD5 of 150 - 250 mg/L; however the system was kept in an average percentage of removal from the order of 99.99% for FC and a 94.46% for the BOD5. These results corroborate cited by several authors (Kadlec et al. 2008, Garcia et al. 2008, Molleda 2011) that assume that the CW are able to withstand fluctuations in loading of contaminants.
To corroborate this information a linear regression was made in which you can see that the data between the concentration of FC and reduction of logarithmic units follow a pattern similar to the linear with a value (R2) close to 1 (0.90), which indicates that there is a high correlation coefficient between the concentration of input and the reduction of FC log units throughout the system (Figure 4); This data is similar to that obtained by Molleda (2011), whose correlation coefficient R2= 0.99, was between the burden of FC and its rate of removal in a SSVFCW of urban and livestock waste water treatment with high concentrations of pathogens.
Figure 4. Linearity of concentration and reduction of FC.
On the other hand, the results obtained regarding the BOD5, shows a linearity with a coefficient of R2 = 0.87 (Figure 5), is not ideal but it is an acceptable result; being this coefficient less than introducing Molleda (2011), between the load to the wetland and its elimination rate (R2 = 0.98).
Figure 5. Linearity of concentration and removal of BOD5. CONCLUSIONS
Indicator bacteria (FC) are effectively reduced by a SSVFCW reaching removals of 99.99%, i.e. in the order of 4.8 logarithmic units (MPN/100 mL); and regarding the BOD5 removal average of 94.46% values are reached. These results are lower than those recorded in similar wetlands of horizontal type with HRT.
The system is not destabilized in terms of FC and BOD5 removal by fluctuations in concentration of input, which corroborates that this type of wetlands can withstand pollutants concentration peaks without affecting its efficiency.
The removal of indicator bacteria and BOD5 in this paper was conducted mainly in stage 1, so it is recommended to do a deeper analysis in terms of the size of such systems.
Organic loading rates analyzed in the present work are low compared with rates that use systems on a large scale, so it is still studying whether the concentration of output reaches
REFERENCES
i. Arias C.A., Cabello A., Brix H., Johansen N.H. 2003. Removal of indicator bacteria from municipal wastewater in an experimental two-stage vertical flow constructed wetland system. Wat. Sci. Tech.,48(5), 35-41.
ii. Ávila S., Estupiñan S., Mejia A. y Mora L. 2014. La calidad bacteriológica del agua del humedal Jaboque (Bogotá, Colombia) en dos épocas contrastantes. Caldasia, 36(2), 323-329. DOI: 10.15446/caldasia.v36n2.47490.
iii. Brix H. and Arias C. 2005. Danish Guidelines for small-scale constructed wetlands systems for onsite treatment of domestuc sewage. Water Sci. Tech., 51(9), 1-9.
iv. García J. and Corzo A., 2008. Depuración con Humedales Construidos. Departamento de Ingeniería Hidráulica, Marítima y Ambiental de la Universidad Politécnica de Catalunya, Barcelona, España.
v. Gopal B. 1999. Natural and constructed wetlands for wastewater treatment: potentials and problems. Water Sci. Tech,40(3), 27-35. DOI: 10.1016/S0273-1223(99)00468-0.
vi. Jin S., Lin Y., Lee D. and Wang T. 2001. Using constructed wetland systems to remove solids from highly polluted river water. Wat. Sci. Tech: Water Supply, 1(1), 89-96.
vii. Karathanasis A., Potter C. and Coyne M. 2003. Vegetation effects bacteria, BOD, and suspended solid removal in constructed wetlands treating domestic wastewater.
Ecological Engineering, 20(2) 157-16. DOI: 10.1016/S0925-8574(03)00011-9.
viii.Kadlec R. and Wallace S. 2008.Treatment wetlands (2da Ed.). United States of America:
Taylor y Francis Group.
ix. Khatiwada N., Polprasert C. 1999. Kinetics of fecal coliform removal in constructed wetlands, Water Sci. Tech,40(3), 109–116. DOI: 10.1016/S0273-1223(99)00446-1.
x. Langergraber G., Pressi A. and Haberl R. 2014. Experiences from the full-scale implementation of a new two-stage vertical flow constructed wetland desing. Wat. Sci.
Tech.,69(2), 335-342. DOI: 10.2166/wst.2013.708.
xi. Li Y., Yu J., Lui Z. and M. T. 2012. Estimation and modeling of direct rapid sand filtration for total fecal coliform removal from secondary clarifier effluents. Wat. Sci.
Tech,65 (9), 1615-1623. DOI: 10.2166/wst.2012.054.
xii. Molleda M. P. 2011. Aplicación de humedales construidos en la reducción de patógenos y otros contaminantes en agua residual urbana y ganadera. Tesis, Universidad de Leon, pp.150.
xiii. Song Z. W., Zheng Z.P. and Li J. 2006. Seasonal and annual performance of a full-scale constructed wetlands system for sewage treatment in China. Ecol. Eng. 26(3), 272–282.
DOI: 10.1016/j.ecoleng.2005.10.008.
xiv. Vacca G., Wand H., Nikolausz M.Kuschk P. and Kastner M. 2005. Effect of plants and filter materials on bacteria removal in pilot-scale constructed wetlands. Water Research, 39(7), 1361-1373. DOI: 10.1016/j.watres.2005.01.005.
xv. Vymazal J. 2010. Constructed wetlands for wastewater treatment. Water, 2(3), 530-549.
DOI: 10.3390/w2030530.
xvi. Wang H., Jiang D., Yang Y. and Cao G. 2013. Analysis of chemical reaction kinetics of depredating organic pollutants from secondary effluent of wastewater treatment plant in constructed wetlands. Wat. Sci. Tech., 67(2), 353-358. DOI: 10.2166/wst.2012.521.
Parallel session 4 - Treatment wetlands 2 14:00-15:20 Session chair: Thomas Aubron
588 - Onsite one stage constructed wetland fed by raw wastewater: performances and resilience of the system DUBOIS Vivien - Irstea
296 - The new draft German constructed wetland guideline for treatment of domestic and municipal waste-water
RUSTIGE Heribert - AKUT Umweltschutz Ingenieure Burkard und Partner
496 - On-site wastewater treatment in France: statistical analysis of the outlet effluent quality of 240 facilities OLIVIER Laurie - Irstea
439 - Green Bio-sorption Reactor: Making wastewater treatment plant an attractive scenario ZHAO Yaqian - Xi’an University of Technology
Parallel session 5 - Treatment wetlands 3 15:40-17:00
Session chair: Mathieu Gautier
690 - Nkolfoulou landfill leachate treatment performances in VFCWs planted with Echinochloa Pyramidalis during dry season
FENDOUNG Guy - IMT Atlantique
689 - Design parameters influence on pollution removal performance of full-scale constructed-wetlands treatments plants for domestic wastewater
RODRIGUEZ VASQUEZ Sebastian - IMT Atlantique
641 - Electroactive Biofilm-based Constructed Wetland (EABB-CW): Testing of an innovative setup for waste-water treatment
ARIAS Carlos - Aarhus University
498 - Nutrients removals in vertical flow constructed wetlands combined or not with biological and/or chemical treatment processes
KIM Boram - INSA Lyon, DEEP laboratory
Parallel session 6 - Treatment Wetland 4 17:20-18:20
Session chair: Tjaša Griessler Bulc
443 - The key factors of building a microbial fuel cell in a pilot-scale constructed treatment wetland TANG Cheng - Chang’an University
631 - Peat as substrate for small-scale constructed wetlands polishing secondary effluents CHAMPAGNE Pascale - Queen’s University, Department of Civil Engineering
688 - The effect of nitrification-denitrification process in one single stage Vertical Flow Constructed Wet-lands (VFCW) treating domestic wastewater
RUIZ Hernan - Institute Mines Telecom Atlantique
Tuesday 24 October
Salle G
Onsite one stage constructed wetland fed by raw wastewater: performances and resilience of the system.
Ref # S2SMALL-83775 V. Dubois1, and P. Molle1.
1Irstea, Freshwater Systems, Ecology and Pollutions Research Unit, 5 rue de la Doua, BP 32108, 69616 Villeurbanne, France
* Presenting Author: vivien.dubois@irstea.fr
Keywords: onsite sanitation, constructed wetlands, performances robustness.
INTRODUCTION
Onsite sanitation systems in Europe are evaluated through a EU labeled procedure done on a platform test under a specific schedule of loads (EN 12566). Nevertheless this test procedure conditions does not represent the real conditions of treatment systems in term of wastewater characteristics and loads. On another angle, in France, systems implemented for capacities above 20 p.e. do not need EU labeled procedure but have to comply with performances requirements. For constructed wetlands, it leads to a situation where design for 21 p.e. can be more compact than for 15 p.e.. Some companies have fulfil EU labeled procedure for one stage vertical flow constructed design between 1.2 and 2 m² per habitant to implement more compact system. This study aims at evaluating, under real conditions the behaviour of a one stage unsaturated/saturated vertical flow constructed wetlands (as designed for small communities, Pringent et al., 2013; Silveira et al., 2015) and fed by raw wastewater (no septic tank). The objectives were to evaluate performances, robustness and resilience of the system when facing load variations.
MATERIAL AND METHODS
The monitoring have been conducted on a unsaturated/saturated vertical flow constructed wetland treating the wastewater from 6 habitants family in the south of France (Crest, Drôme department). The system consists in:
xA pumping station receiving all wastewaters from the house (no screening) and delivering batches to the treatment system. The batch feeding flow is of 3 m3/h for an average batch volume of 80 liters.
xA unsaturated/saturated vertical flow constructed wetlands divided in two filters in parallel to implement feeding and resting periods (one week/one week). The filters comprises a 70 cm of 2/6 mm crushed and washed gravel (unsaturated layer), an aeration pipe and a saturation layer composed of pea gravel (20/40 mm). The saturation layer has been tested between 22 and 40 cm of depth. Each filter has a surface of 7 m².
Treated wastewaters are infiltrated into the ground in winter and reused for brush irrigation the rest of the year.
The monitoring, over one year and half, started after 3 years of operation and consisted in:
xCounting the number of people present in the house for the night, the lunch and the dinner, every days.
xMeasuring the daily flow by a pressure probe (STS) in the pumping station measuring on a one minute time step
xPerforming 24 h flow proportional samples at the inlet/outlet of the system to analyse global parameters (COD, dissolved COD, TOC, DOC, TSS, BOD5, KN, N-NH4, N-NO2, N-NO3, P-PO4, TP). 31 campaigns have been done in different seasons.
xFor some campaigns outlet online measurement of N-NO3/N-NH4 have been done (WTW Varion).
RESULTS AND DISCUSSION
Characterization of habitant pollutant production
Wastewater characteristics observed on the site reveals that concentrations are much higher than the one usually used in EU labeled tests (see Table 1).
Table 1: Wastewater characteristics
average 521 1853 467 1032 767 209 152 29
median 347 1780 470 1065 784 207 148 29
min 67
(XX)*: Reference values for EU labeled tests
The hydraulic load received varies significantly with some huge loads (maximum of 4.4 m3/d representing a hydraulic load on the filter in operation of 63 cm/d) for very special events.
Excepting the specials events, the daily volume production and classical variation within a day can be seen on Figure 1.
Figure 1.daily hydraulic load (left) and average hydraulic load distribution during the day (right)
The average daily flow volume is of 290 liters per day. When comparing to human presence in the house each day, the generated daily volume per capita is of 54 liters in average (Figure 2).
Figure 2.daily generated volume per capita
The measurements allow quantifying volume variations, which can double quite often even, for special events, be multiplied by 3 or more. The unique special event, with more than ten times the average hydraulic load, corresponds to a forgotten open tap associated with more people in the house. If this event can be seen as an accident and unusual situation it is important to evaluate if such an event can be accepted by the system or damage its functioning.
Treatment performances
Based on a 60g of BOD5/p.e./d and 150 g of COD/p.e./d the system received an corresponding organic load of about 4 p.e. (one habitant is not equal to a people equivalent) with a maximum monitored load of 10.5 p.e. Even in term of organic load, variations are important and sudden.
Inlet outlet concentrations for all measured parameters are presented in Table 2. Whatever inlet concentration variations, outlet concentrations remain always low for carbon pollutant parameters, solids and nitrification. It allows realize the stability of the system regarding inlet hydraulic and organic load variations. Indeed, Figure 2 presents the organic and hydraulic load variations over consecutive days for two seasons. If hydraulic load remains globally low compared to what such system can accept (Molle et al., 2006), except for the exceptional event, the organic load vary a lot which could affect biological stability and efficiency.
Figure 3 :organic and hydraulic loads variations over consecutive monitoring days for two seasons
Table 2 : Inlet – outlet concentrations during 24 h proportional flow composite samples campaigns
TSS (mg/l)
COD (mg/l)
CODd (mg/l)
BOD5
(mg/l)
TOC (mg/l)
DOC (mg/l)
KN (mg/l)
N-NH4
(mg/l)
N-NO2
(mg/l)
N-NO3
(mg/l)
TP (mg/l)
P-PO4
(mg/l) in out in out in out in out in out in out in out in out in out in out in out in out average 1032 10 1853 62 467 49 767 7 578 25 163 19 209 12 152 7.8 0.08 0.19 1.4 69 29 17 10 14 Median 1065 9 1780 57 470 49 784 7 588 24 157 19 207 9 148 5.4 0.04 0.16 0.5 62 29 20 9 15 Min 213 2 358 23 53 20 530 3 435 19 116 14 29 3 17 0.3 0.02 0.02 0.5 15 6 8 2 7 Max 1590 22 2750 100 708 82 1100 16 762 30 240 24 330 29 252 22.5 0.51 0.53 5.1 148 44 23 20 21 Nb of
value 30 30 31 31 31 31 13 13 17 17 13 13 31 22 31 31 31 31 31 31 28 28 28 28
Despite these high load variations, performances removal stay stable as it can be seen on Figure 3. It appears that removal performances are extremely stables with, on average, more than 98 %, 99 %, 94 % and 97 % for TSS, BOD5, COD and KN respectively. Even if the hydraulic load is usually low, organic load can reach values similar to nominal design of French vertical flow constructed wetlands (Molle et al., 2005).
Figure 4: applied and treated loads on the filter in operation.
TN removal still not complete and no clear correlation can be done according to season. It appears that increasing the depth of the bottom saturation from 22 cm to 40 cm allowed to improve total nitrogen removal (Figure 3). With 22 cm of saturation TN removal is about 45
% and not stable, while with 40 cm total nitrogen removal reached around 70 %. It is likely not possible, with such a unsaturated layer (70 cm), to reach better TN performances as the outlet available carbon is very low. Average and median outlet COD concentrations were of 68 and 69 mgCOD/l when implementing a 22 cm saturation and of 46 and 45 mgCOD/l when implementing a 40 cm saturation depth. It is likely that those last values are close to the non-biodegradable part of COD.
Phosphorus is not correctly treated in the filter as no specific materials are implemented. Only 40 % on average is removed and correspond mainly to the particulate form of P that is filtrated by the system.
Feeding the system with raw wastewater should lead to an organic deposit accumulation on the filter’s surface (Molle et al., 2005). Due to the low load applied, and despite that reeds are cut but not removed from the filter, it is no possible to measure a deposit layer depth after 5 years of operation. Main of the deposit is mineralized, which easier the sludge management and confirm the interest in feeding the system with raw wastewater.
CONCLUSIONS
The real scale monitoring over one year and half of an onsite constructed wetlands system allows to draw some global information.
In term of wastewater production and variation on a single house:
x Wastewater concentration characteristics are much higher than the one used in EU labeled tests. Onsite system needs to face to higher concentrations.
x There is not a proportional relation between the number of people in the house and the amount of emitted water. The emitted volume per person is on average three times lower than the reference value used in the EU labeled tests but can vary between 4 and 304 liters.
This could question the relevance of the EU test and the difference of treatment systems behavior between labelled test and real life.
x Organic load variation can vary up to a ten factor from one day to another one and requires treatment system to be robust and resilient to un-expected events.
In term of performances of the unsaturated/saturated vertical flow constructed wetlands tested, we noted that the system stay very stable whatever the variations observed during the monitoring. Removal performances are very high and outlet concentrations low and stable despite inlet variations. If performances are high for solids, carbon removal and nitrification, TN removal is impacted by the depth of the saturation level. The designed implemented seems not able to go higher than 70 % of TN removal as it doesn’t still carbon source for denitrification at the outlet. To reach higher level, design should be modified, balancing nitrification performances and TN removal.
REFERENCES
Prigent, S., Paing, J., Andres, Y., Chazarenc, F., 2013. Effects of a saturated layer and recirculation on nitrogen treatment performances of a single stage verticalflow constructed wetland (VFCW). Water Sci. Technol. 68 (7), 1461–1467.
Silveira D.D., Belli Filho P., Philippi L.S., Kim B., Molle P. (2015) Influence of partial saturation on total nitrogen removal in a single-stage French constructed wetland treating raw domestic wastewater. In Ecological Engineering, 77, pp. 257-264.
Molle P., Liénard A., Grasmick A., Iwema A., (2006). Effect of reeds and feeding operations on hydraulic behaviour of vertical flow constructed wetlands under hydraulic overloads.
Water research, Vol 4 0 (3), pp. 606-612
Molle P., Liénard A., Boutin C., Merlin G., Iwema A., (2005), How to treat raw sewage with constructed wetlands: An overview of the French systems. Wat. Sci. & Tech, vol 51, n°9, pp
The IWA S2Small2017 Conference on Small Water & Wastewater Systems and Resources Oriented Sanitation