Performance evaluation of different horizontal subsurface flow wetland types by characterization of flow behavior, mass removal and depth-dependent contaminant load
Graphical abstract
Highlights
► Plants improve the contaminant removal efficiency in the gravel CW. ► The hydroponic plant root mat achieves better contaminant removal than the planted gravel CW. ► Plants induce flow and thus contaminant flux into the root zone of the gravel CW. ► Local residence time distributions allow to observe dead zones and preferential flow paths.
Introduction
Constructed wetlands (CWs) have proven to be an efficient ecological technology for the treatment of various kinds of contaminated waters (Williams, 2002; Haberl et al., 2003; Kadlec and Wallace, 2008), including domestic and agricultural wastewater (Konnerup et al., 2009; Vymazal and Kröpfelová, 2009), landfill leachate (Bulc, 2006; Yalcuk and Ugurlu, 2009), industrial effluents (Vymazal, 2009) and groundwater contaminated with organic chemicals (Braeckevelt et al., 2008; Seeger et al., 2011a). The use of CWs has been successfully tested in pilot- (Braeckevelt et al., 2011) and field-scale (Ferro et al., 2002; Moore et al., 2002) applications, providing data on overall contaminant removal efficiency on the basis of either concentrations or loads.
CW treatment efficiency primarily depends on the contact time between the contaminated water and the filter material, including biota, as longer residence times enhance contaminant turnover (Werner and Kadlec, 1996). Hence, the characterization of hydraulic flow in CWs is crucial for the evaluation and optimization of system designs and performances (Persson et al., 1999). Moreover, detailed knowledge on flow behavior is essential for the validation and improvement of existing models. The hydraulic characterization of CWs is mostly based on tracer experiments, where the analysis of outlet breakthrough curves (BTCs) using the temporal moment method allows the calculation of the CWs' flow characteristic parameters, such as the mean residence time (τ), porosity (θ), variance (σ2), and dispersion parameters. These system-specific parameters may reveal whether the CW is actually operating with ideal flow, short-circuiting, or dead zones, and allow for a comparison between differently designed and operated CWs (Werner and Kadlec, 1996; García et al., 2004; Holland et al., 2004). Since the connection between hydraulic residence time and treatment efficiency has been recognized, many studies have already evaluated the effects of wetland design parameters (e.g. aspect ratio, inlet/outlet configuration, filter medium size, or plant development) and operational modes on the flow behavior or the removal efficiency (García et al., 2005; Kjellin et al., 2007; Ascuntar Ríos et al., 2009; Su et al., 2009; Hijosa-Valsero et al., 2010). Nevertheless, investigations that simultaneously encompass the hydraulic characterization of different CW systems, the change of hydraulic parameters due to operational mode adaptation, and the respective contaminant removal efficiency are rare (Ascuntar Ríos et al., 2009). Most studies evaluate treatment efficiency based on the black box concept, only focusing on overall removal rates. Some studies additionally provide contaminant concentrations of the pore water. However, loads within the wetland have not been described previously, given that actual flow at the sampling points was unknown.
Therefore, the goal of our work was to evaluate the flow characteristics of various pilot-scale horizontal subsurface flow CWs (two conventional wetlands with a gravel filter (planted and unplanted) and a plant root mat system without gravel matrix) and to relate the hydraulic parameters obtained from tracer tests to the achieved treatment efficiencies. For the plant root mat, the effect of low and high water level on both the hydraulic parameters and the pollutant removal was also investigated. Furthermore, local flow rates and local contaminant loads were assessed for the gravel-based CWs in order to identify preferential flow paths and zones with enhanced contaminant flux. The analysis is based upon local tracer BTCs monitored at various sampling points in the porous media.
Section snippets
Constructed wetland design
The study was carried out at the so-called CoTra (Compartment Transfer) research site in Leuna, Germany, where six pilot-scale horizontal subsurface flow (HSSF) CWs were set up during 2007 in order to evaluate the optimized CW design for the remediation of the local groundwater, which is contaminated with benzene (∼20 mg L−1), methyl tert-butyl ether (MTBE) (∼3.7 mg L−1) and ammonium (57 mg L−1) (Seeger et al., 2011a,b). All CWs consisted of steel basins (5 m length × 1.1 m width × 0.6 m
Normalized residence time distribution
Transferring time and concentration to the dimensionless flow-weighted time variable Φ = Vout(t)/Vsys and normalized concentration c′(Φ) = c(ti) Vsys/Mout, respectively, displays dimensionless BTCs, which allow the direct visual comparison between differently designed and operated CWs (Holland et al., 2004; Wahl et al., 2010). This procedure also accounts for different tracer mass that might have been applied to the CWs (Werner and Kadlec, 1996; Holland et al., 2004; Headley and Kadlec, 2007). Φ
Results and discussion
The results and discussion part is divided into three sections. The first Section 4.1 addresses the evaluation of the hydraulic flow behavior of the gravel CWs and the hydroponic plant root mat by analyzing outlet tracer BTCs using the temporal moment method. The second part (4.2) is about the relation between the goodness of the hydraulic flow behavior measured by the hydraulic indices and the achieved purification efficiencies. The third paragraph (4.3) presents contaminant loads within the
Conclusion
In the present work we present a detailed evaluation of the hydraulic performance and the load removal efficiency of several HSSF CWs (a planted and unplanted gravel-based CW) and a hydroponic plant root mat system (operated at two water levels) treating gasoline and ammonium contaminated groundwater. We investigated the flow behavior following up the state-of-the-art procedure that comprises the determination of system-specific overall hydraulic indices by analyzing outlet tracer response
Acknowledgements
This work was supported by the Helmholtz Centre for Environmental Research – UFZ within the scope of the SAFIRA II Research Programme (Revitalization of Contaminated Land and Groundwater at Megasites, subproject ‘‘Compartment Transfer – CoTra”), and the European Union (European Commission, FP7 Contract No. 265946, Minotaurus). The authors thank F. Löper, P. Weiβbach and M. Schröter for their assistance in the field and laboratory.
Glossary
- Ain
- area perpendicular to the flow direction [L²]
- Asurf
- surface area [L²]
- c′(Φ)
- normalized concentration [–]
- ctracer
- fluorescein concentration of the tracer solution [M L³]
- cx,y
- contaminant concentration at the sampling position (x,y) [M L³]
- D
- dispersion coefficient [L² t−1]
- Dn
- dispersion number [–]
- fOC
- organic carbon content [%]
- h
- water table [L]
- Kd
- solid water distribution coefficient [L3 M–1]
- KOC
- organic carbon water distribution coefficient [L3 M–1]
- KOW
- octanol water distribution coefficient [–]
- L
- length of the
References (75)
- et al.
Assessment of the plug flow and dead volume ratios in a sub-surface horizontal-flow packed-bed reactor as a representative model of a sub-surface horizontal constructed wetland
Ecol. Eng.
(2012) - et al.
Changes of flow patterns in a horizontal subsurface flow constructed wetland treating domestic wastewater in tropical regions
Ecol. Eng.
(2009) - et al.
Treatment of chlorobenzene-contaminated groundwater in a pilot-scale constructed wetland
Ecol. Eng.
(2008) - et al.
Chlorobenzene removal efficiencies and removal processes in a pilot-scale constructed wetland treating contaminated groundwater
Ecol. Eng.
(2011) Long term performance of a constructed wetland for landfill leachate treatment
Ecol. Eng.
(2006)- et al.
Hydrodynamics of horizontal subsurface flow constructed wetlands
Ecol. Eng.
(2003) - et al.
Spatial variations of temperature, redox potential and contaminants in horizontal flow reed beds
Ecol. Eng.
(2003) - et al.
Hydraulic behaviour of horizontal subsurface flow constructed wetlands with different aspect ratio and granular medium size
Ecol. Eng.
(2004) - et al.
Effect of key design parameters on the efficiency of horizontal subsurface flow constructed wetlands
Ecol. Eng.
(2005) - et al.
Comparison of sulfur hexafluoride, fluorescein and rhodamine dyes and the bacteriophage PRD-1 in tracing subsurface flow
J. Hydrol.
(2003)