Page | 1  

 

 

Determination of the Kinetics of Contaminant Degradation 

Processes as a Precursor to Improved Constructed Wetland Design 

 

A dissertation for the fulfilment of the requirements of Master of Engineering 

 

Prepared by 

 

Lara Aylward 

300324 

 

Submitted to 

 

School of Chemical and Metallurgical Engineering, Faculty of Engineering and the Built Environment, 

University of the Witwatersrand, Johannesburg, South Africa 

 

Supervisor(s): 

Dr Craig Sheridan Pr Eng. (Wits University)  

Dr Uwe Kappelmeyer (UFZ, Leipzig, Germany) 

May, 2019 



Page | 2  

 

Declaration  

 

I declare that this dissertation is my own unaided work. It is being submitted for the degree of Master of 

Engineering to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, 

Johannesburg.  

It has not been submitted previously for any degree or examination to any other University.  

 

 

 

 

Lara Anne Aylward 

 

23rd day of May 2019 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



Page | 3  

 

Acknowledgements 

 

I would like to thank the following people and organizations for their generous funding and invaluable advice 

and support: 

 Prof. Craig Sheridan and Dr Uwe Kappelmeyer for their exceptional supervision and endless patience; 

 The staff of the School of Chemical and Metallurgical Engineering at the University of the 

Witwatersrand; 

 The staff of the Department of Environmental Biotechnology, as well as Dr Marie Kurz, Dr Christian 

Schmidt and Dr Tanja Brandt, at the Helmholtz Center of Environmental Research (UFZ), Leipzig, 

Germany; 

 Ricky Bonner for his inspirational work ethic, his dedication to our shared experimental work and his 

contributions to each of my published papers; 

 Philipp Hecht (TU Dresden), Michael van Zyl (Wits University) and Alyona Lepilova (University 

Thübingen) for their assistance in gathering and analysing data for this research as part of their diploma 

(Philipp) and masters (Michael, Alyona) research stays at the UFZ and their great company over many 

hours outside at the wetlands, in all weather conditions.  

 My mom and dad for their unwavering support and encouragement and always providing a listening ear. 

For financial assistance: 

 Golder Associates South Africa  

 The South African Water Research Commission 

 The South African Department of Science and Technology 

 The University of the Witwatersrand Friedel Sellschop Award 2015 

 The Helmholtz Center of Environmental Research 

 The National Research Foundation (NRF) 

 The German Academic Exchange Service (DAAD) 

 Water4Crops (an EU initiative) 

 

  



Page | 4  

 

Table of Contents 

Table of Figures.............................................................................................................................. 8 

List of Tables ................................................................................................................................ 13 

List of Nomenclature .................................................................................................................... 15 

List of Acronyms .......................................................................................................................... 18 

 Introduction ........................................................................................................................................20 

1.1. Background.........................................................................................................................................20 

1.1.1. The water conundrum in South Africa ......................................................................... 20 

1.1.2. Contamination of South Africa’s water resources ........................................................ 20 

1.1.3. South African water legislation .................................................................................... 21 

1.1.4. Sustainable water infrastructure and the green economy ............................................. 21 

1.1.5. The role of constructed wetlands and biomimicry ....................................................... 22 

1.2. The relevance of this research ............................................................................................................23 

1.3. Research objectives ............................................................................................................................24 

1.4. Structure of the thesis .........................................................................................................................24 

1.5. List of Publications from this study ....................................................................................................25 

 Literature Review ..............................................................................................................................26 

2.1. Introduction to constructed wetlands ..................................................................................................26 

2.1.1. Natural wetland ecosystems ......................................................................................... 26 

2.1.2. Historical development of constructed wetlands .......................................................... 27 

2.1.3. Types of constructed wetlands ..................................................................................... 28 

2.1.4. Wetland vegetation ....................................................................................................... 29 

2.1.5. Advantages of constructed wetlands ............................................................................ 34 

2.2. Wastewater quality .............................................................................................................................34 

2.3. Remediation of wastewater using constructed wetlands ....................................................................34 

2.3.1. Contaminant removal mechanisms in constructed wetlands ........................................ 34 

2.3.2. Removal of organic carbon ........................................................................................... 35 



Page | 5  

 

2.3.3. Removal of nitrogen ..................................................................................................... 36 

2.3.4. Removal of phosphorus ................................................................................................ 38 

2.3.5. Removal of additional contaminants ............................................................................ 39 

2.4. Constructed wetland hydraulics..........................................................................................................39 

2.4.1 Chemical engineering reactor theory ............................................................................ 39 

2.4.2. Types of fluid flow ....................................................................................................... 41 

2.4.3. Hydraulic tracer tests .................................................................................................... 43 

2.4.4. The residence time distribution for steady-flow systems ............................................. 44 

2.4.5. Normalization of flow data ........................................................................................... 45 

2.4.6. The residence time distribution for variable flow systems ........................................... 45 

2.4.7. Moments of the RTD .................................................................................................... 46 

2.4.8. Flow behaviour in constructed wetland systems .......................................................... 47 

2.4.9. Hydraulic efficiency ..................................................................................................... 49 

2.4.10. Hydraulic indexes ......................................................................................................... 51 

2.5. Chemical kinetics ...............................................................................................................................54 

2.6. Biomimicry and constructed wetland design ......................................................................................54 

2.6.1. Ecological and Life’s Principles ................................................................................... 54 

2.6.2. Biomimetic approach to wastewater treatment ............................................................. 56 

 Materials and Methods .......................................................................................................................59 

3.1. Experiments at the University of the Witwatersrand ..........................................................................59 

3.1.1. Constructed wetland description .................................................................................. 59 

3.1.2. Wits Experimental timeline .......................................................................................... 60 

3.1.3. Wetland preparation ..................................................................................................... 61 

3.1.4. Hydraulic studies .......................................................................................................... 61 

3.2. Experiments at the Helmholtz UFZ ....................................................................................................62 

3.2.1. Constructed wetland description .................................................................................. 62 

3.2.2. UFZ Experimental timeline .......................................................................................... 64 



Page | 6  

 

3.2.3. Hydraulic studies .......................................................................................................... 64 

3.2.4. Chemical and kinetic studies ........................................................................................ 66 

 A comparison of three different residence time distribution modelling methodologies for horizontal 

subsurface flow constructed wetlands ................................................................................................................69 

4.1. Summary.............................................................................................................................................69 

4.2. Statement of individual contribution ..................................................................................................69 

4.3. Ecol. Eng. (2017). Vol. 99 pp. 99-113 ...............................................................................................70 

4.3.1. Abstract......................................................................................................................... 70 

4.3.2. Introduction .................................................................................................................. 71 

4.3.3. Background ................................................................................................................... 71 

4.3.4. Research objectives ...................................................................................................... 75 

4.3.5. Materials and methods .................................................................................................. 76 

4.3.6. Results and discussion .................................................................................................. 86 

4.3.7. Conclusion .................................................................................................................... 96 

 Investigation into the kinetics of constructed wetland degradation processes as a precursor to 

biomimetic design ..............................................................................................................................................99 

5.1. Summary.............................................................................................................................................99 

5.2. Statement of individual contribution ..................................................................................................99 

5.3. Water SA (2017) Vol. 43 No. 4 pp. 655-665......................................................................................99 

5.3.1. Abstract......................................................................................................................... 99 

5.3.2. Introduction ................................................................................................................ 100 

5.3.3. Materials and methods ................................................................................................ 103 

5.3.4. Results and discussion ................................................................................................ 106 

5.3.5. Conclusion .................................................................................................................. 115 

 Hydraulic study of a non-steady horizontal sub-surface flow constructed wetland during start-up 117 

6.1. Summary...........................................................................................................................................117 

6.2. Statement of individual contribution ................................................................................................117 



Page | 7  

 

6.3. Sci. Tot. Environ. (2019) Vol. 646 pp. 880-892 ...............................................................................117 

6.3.1. Highlights ................................................................................................................... 117 

6.3.2. Graphical abstract ....................................................................................................... 117 

6.3.3. Abstract....................................................................................................................... 118 

6.3.4. Introduction ................................................................................................................ 118 

6.3.5. Theoretical background .............................................................................................. 119 

6.3.6. Research objectives .................................................................................................... 125 

6.3.7. Materials and methods ................................................................................................ 125 

6.3.8. Results and discussion ................................................................................................ 129 

6.3.9. Conclusion .................................................................................................................. 139 

 Contaminant degradation in a pilot-scale constructed wetland in the early start-up phase .............140 

7.1. Summary...........................................................................................................................................140 

7.2. Statement of Individual Contribution ...............................................................................................140 

7.3. Experimental Results ........................................................................................................................140 

7.3.1. Baseline physico-chemical characteristics ................................................................. 140 

7.3.2  Contaminant degradation and transformation ............................................................ 145 

 Discussion and concluding remarks .................................................................................................157 

8.1. Discussion.........................................................................................................................................157 

8.2. Conclusion ........................................................................................................................................167 

 References ........................................................................................................................................169 

Appendix A: Experimental Data, UFZ Leipzig ............................................................................................180 

Appendix B: Matlab Code for Flow Test Data Analysis ..............................................................................209 

Appendix C: Tracer Flow Test Data, UFZ Leipzig ......................................................................................241 

Appendix D: Climatic Data (2016) ..............................................................................................................266 

Appendix E: Supplementary information for Sci. Tot. Environ. (2019) Vol. 646 pp. 880-892 ....................280 

Appendix F: Phragmites australis fluoride tolerance experiments ..............................................................286 

 



Page | 8  

 

Table of Figures 

Figure 2-1: The definition of natural wetlands, their distinguishing characteristics and their classification 

according to the dominant plant species (Kivaisi, 2001). ..............................................................26 

Figure 2-2: The functional roles of a natural wetland (the dashed arrows represent contaminant removal by the 

wetland system) (Kivaisi, 2001). ...................................................................................................27 

Figure 2-3: Classification of constructed wetlands (Brix, 1994b). .....................................................................28 

Figure 2-4: The impulse and step change output concentration curves for ideal (dotted lines) and non-ideal 

(dashed lines) reactors in response to an impulse or step increase in input concentration (bold lines 

with arrow heads). (a) and (b) depict plug flow and (c) and (d) represent complete mixing (Headley 

and Kadlec, 2007). .........................................................................................................................43 

Figure 2-5: (a) Hypothetical concentration breakthrough curve for an idealized impulse-response tracer test and 

(b) the corresponding residence time distribution function. ..........................................................44 

Figure 2-6: (a) Three hypothetical RTD’s showing the effect of the degree of dispersion on peak shape, breadth 

and position and (b) Four hypothetical RTDs showing the influence of the amount of dead volume 

on the position of the RTD (Thackston et al., 1987). ....................................................................48 

Figure 2-7: The typical exponentially decaying concentration time profile of a perfectly mixed reactor and 

hypothetical concentration time breakthrough curves for two systems displaying various 

distributions of advective and stagnant zones (Thackston et al., 1987). ........................................49 

Figure 3-1: The experimental constructed wetlands at the IMWaRU site, Wits University showing the five zones 

of vegetation (left), the planted and unplanted wetlands side-by-side and the Jojo tank set-up 

(right). The outlet configurations (inlet configurations are not shown but are of the same design), 

the syphon break and the sample ports are also visible (right). .....................................................59 

Figure 3-2: IMWaRU (Wits University) constructed wetland configuration showing a removable cage and 

sample tubing to three depths. The outlet distribution network is also visible and the inlet 

distribution network is of the same arrangement. The red dashed lines illustrate the division of the 

wetland system into a grid of sample points. .................................................................................60 

Figure 3-3: Experimental timeline for the experiments conducted in the constructed wetlands at the University 

of the Witwatersrand, 2015. ..........................................................................................................60 

Figure 3-4: (a) Aerial view of the constructed wetland showing seven internal sample ports, three inlet valves 

and one outlet valve and (b) cross-sectional view showing the gravel bed and valve heights. .....63 

Figure 3-5: (a) Photograph of the pilot, gravel horizontal subsurface flow constructed wetland at the Helmholtz 

UFZ, Leipzig (July 2015) and (b) the middle basket when removed from the wetland showing a 

well-developed root system (Sept 2015). ......................................................................................63 



Page | 9  

 

Figure 3-6: Timeline of experiments conducted in the constructed wetlands at the Helmholtz UFZ during 2015.

 .......................................................................................................................................................64 

Figure 3-7: Timeline of experiments conducted in the constructed wetlands at the Helmholtz UFZ during 2016.

 .......................................................................................................................................................64 

Figure 3-8: Stainless steel sampling devices (samplers), peristaltic pump, flow cell and fluorometer 

configuration for the impulse-response flow tests. ........................................................................65 

Figure 4-1: Overview of the three different RTD modelling methodologies used in this study. The initial phase 

consists of data generation in the form of the hydraulic tracer study being performed on the wetland 

system and subsequent development of the concentration-time curve. The second phase consists 

of data processing and various numerical methods to obtain the hydraulic characteristics of the 

system. ...........................................................................................................................................76 

Figure 4-2: Generic concentration-time data generated from the impulse and step change tracer experiments 

(Fogler, 1999). The response curves of both the impulse and step change tracer experiments 

typically differ considerably from their corresponding injection curves which indicates non-ideal 

flow patterns such as dispersion and short-circuiting effects. .....................................................77 

Figure 4-3: The relationship between the concentration-time curve and the RTD function used in the impulse 

response modelling methodology (Levenspiel, 1999a). The mathematical transformation includes 

computing the area underneath the concentration-time curve and hence the use of numerical 

integration techniques such as Simpson’s integration formulae. ...................................................78 

Figure 4-4: Determining the mean residence time of the fluid by computing the area above the F(t) curve 

(Levenspiel, 1999a). ......................................................................................................................79 

Figure 4-5: The relationship between the RTD function, E(t), and the cumulative distribution function, F(t) 

(Levenspiel, 1999a). The conversion of F(t) to E(t) requires the use of numerical integration techniques 

such as the central difference approximation. ..................................................................................80 

Figure 4-6: TIS model demonstrated for the real reactor (left) using a combination of idealised tanks in series 

(right). Adapted from (Fogler, 1999). ............................................................................................81 

Figure 4-7: Normalized E(t) (left) and F(t) (right) curves for different values of n (Fogler, 1999). As the number 

of stirred tanks in series increases the system approaches plug flow behaviour whereas the number 

of tanks in series approach unity the degree of back mixing increases and the system resembles a 

completely stirred tank reactor. .....................................................................................................81 

Figure 4-8: Spatial arrangement of inlet, outlet and internal sampling ports as well as the system grid resolution. 

All dimensions are expressed in mm. Samples were taken from three different depths within each 

of the thirteen sampling wells as well as at the system outlet. ......................................................83 



Page | 10  

 

Figure 4-9: Zonal distribution of wetland vegetation in planted system. Five different species were used and 

were strategically planted according to their tolerance of wastewater with the more resistant species 

planted towards the inlet of the wetland where higher contaminant concentrations would be 

expected. ........................................................................................................................................84 

Figure 4-10: E(t) and F(t) curves from the impulse and step change experiments, respectively. In Fig. 10(a) and 

(c) hydraulic data from the impulse response experiment on unplanted and planted systems are 

presented respectively and likewise in Fig. 10(b) and (d) using the step change response 

experiment. ....................................................................................................................................87 

Figure 4-11: RTD curves from the impulse experiments on the planted and unplanted systems. In Figures (a) 

and (b) hydraulic data from the surface layer are presented. In Figures (c) and (d) data from the 

intermediate layer are presented and in Figures (e) and (f) data from along the bottom of the 

wetland bed are presented. Each curve within the figures represents hydraulic data collected from 

a sampling point a specified distance from the system inlet thus allowing for the evolution of the 

RTD curve to be depicted as a function of system length. ............................................................92 

Figure 4-12: Cumulative distribution curves from the impulse experiments on the planted and unplanted 

systems. In Figures (a) and (b) hydraulic data from the surface layer are presented. In Figures (c) 

and (d) data from the intermediate layer are presented and in Figures (e) and (f) data from along 

the bottom of the wetland bed are presented. Each curve within the figures represents hydraulic 

data collected from a sampling point at a specified distance from the system inlet. .....................93 

Figure 4-13: Comparison of 𝑡𝑚𝜏 for when 3-point subintervals and when one subinterval is used for numerical 

integration using Simpson’s 1/3 rule. The comparison performed on the hydraulic data obtained 

from the unplanted system is presented in Fig. 13(a) and for the planted system in Fig. 13(b) and 

is performed in each case for the impulse response as well as the step change derivative and step 

change integral modelling approaches...........................................................................................95 

Figure 4-14: Normalized E(t) for planted and unplanted systems with regions of noise highlighted on the E(t) 

curves from the step change derivative approach. .........................................................................96 

Figure 5-1: (a) Aerial view of the CW showing the sample port, three inlet valve and single outlet valve locations 

and (b) cross-sectional view of the CW showing one stainless steel sampling tube. ..................103 

Figure 5-2: (a) A photograph of the pilot-scale, horizontal subsurface flow CWs at the Helmholtz UFZ (July 

2015) and (b) the stainless steel basket removed from the centre of the wetland showing an already 

well-developed root system (September 2015). ..........................................................................104 

Figure 5-3: A schematic diagram of the sampling device, pump, flow cell and fluorometer configuration for the 

tracer flow tests. ...........................................................................................................................105 



Page | 11  

 

Figure 5-4: Baseline characterization showing (a) the dissolved oxygen concentration and (b) the oxidation-

reduction potential prior to the introduction of artificial wastewater. Data was obtained from 24 

cm below the gravel bed surface and at increasing distance along the flow path during August, 

September and October 2015. .....................................................................................................108 

Figure 5-5: The normalized RTD function for the pilot CW determined from an impulse-response tracer test, 

with τ = 2.16 d (θ = 0.78), tm = 2.89 d (θ = 1.0) and a peak time at maximum concentration of 2.31 

d (θ = 0.80). .................................................................................................................................110 

Figure 5-6: (a) Total organic carbon concentration (avg. inlet [TOC] = 573 ± 62 mg/ℓ; baseline TOC = ± 3.0 

mg/ℓ) and (b) total nitrogen concentration (avg. inlet [TN] = 230 ± 21 mg/ℓ; baseline TN = ± 1.7 

mg/ℓ) as a function of time at the wetland outlet up to 7 weeks after the introduction of artificial 

wastewater into the system. .........................................................................................................112 

Figure 5-7: An illustration of the general trend in (a) TOC concentration (avg. inlet [TOC] = 573 ± 62 mg/ℓ) 

and (b) the dissolved oxygen concentration at increasing distance from the wetland inlet over the 

7 week experimental period (avg. inlet [DO] = 11.158 ± 1.208 mg/ℓ; avg. inlet temp = 10.9 ± 

2.6°C)...........................................................................................................................................113 

Figure 5-8: The change in oxidation-reduction potential as a function of time at the wetland outlet up to 7 weeks 

after the introduction of artificial wastewater into the system (avg. inlet potential = 336.1 ± 30.1 

mV). .............................................................................................................................................114 

Figure 5-9: The variability in the rate of decomposition of total organic carbon, kTOC, and the rate of 

transformation of total nitrogen, kTN ............................................................................................115 

Figure 6-1: Summary and comparison of the mathematics underlying hydraulic theory for steady versus variable 

flow systems ................................................................................................................................122 

Figure 6-2: (a) An aerial view of the constructed wetland showing seven internal sample ports, three inlet valves 

and one outlet valve and (b) cross-sectional view showing the gravel bed and valve heights. ...126 

Figure 6-3: (a) 2015 and (b) 2016 idealized concentration breakthrough curves at various distances from the 

wetland inlet and mid-depth (24cm) below the gravel surface, assuming a tracer mass of 30 mg 

and hydraulic residence time of 7 days........................................................................................131 

Figure 6-4: Normalized RTDs, assuming steady flow, at various distances along the flow path at (a) mid-depth 

(2015) and (b) 12 cm, (c) 24 cm and (d) 36 cm depth (2016). ....................................................133 

Figure 6-5: Comparison of the 2015 and 2016 normalized RTDs at various distances along the flow path and at 

mid-depth (24cm) below the gravel surface. ...............................................................................134 

Figure 6-6: 2016 normalized RTDs at the wetland outlet for experiments in duplicate, calculated using standard 

(dashed line) and variable flow (solid line) methodologies. ........................................................135 



Page | 12  

 

Figure 6-7: Evapotranspiration, rainfall, relative humidity and global radiation measured in Leipzig during 

April, August and October 2016. .................................................................................................138 

Figure 7-1: (a) Redox potential and (b) dissolved oxygen concentration along the constructed wetland at a depth 

of 12cm over the 5 week baseline sampling period. ....................................................................142 

Figure 7-2: (a) Redox potential and (b) dissolved oxygen concentration along the constructed wetland at a depth 

of 24cm over the 5 week baseline sampling period. ....................................................................143 

Figure 7-3: (a) Redox potential and (b) dissolved oxygen concentration along the constructed wetland at a depth 

of 36cm over the 5 week baseline sampling period. ....................................................................144 

Figure 7-4: Total organic carbon hydraulic loading (avg. inlet TOC loading = 10.1 ± 1.7 g/m3.d; baseline TOC 

loading = 0.81 ± 0.22 g/m3.d) as a function of time at the wetland outlet up to 10 weeks after the 

introduction of artificial wastewater into the system against the variation in volumetric flowrate 

(avg. inlet flowrate = 6.2 ± 0.3 L/d; avg. outlet flowrate = 5.3 ± 0.4 L/d for week 1-5). ............146 

Figure 7-5: Total nitrogen hydraulic loading (avg. inlet TN loading = 4.5 ± 0.8 g/m3.d; baseline TN loading = 

0.34 ± 0.16 g/m3.d) as a function of time at the wetland outlet up to 10 weeks after the introduction 

of artificial wastewater into the system against the variation in volumetric flowrate (avg. inlet 

flowrate = 6.2 ± 0.3 L/d; avg. outlet flowrate = 5.3 ± 0.4 L/d for week 1-5). .............................147 

Figure 7-6: The change in dissolved oxygen concentration as a function of time at the wetland outlet up to 10 

weeks after the introduction of artificial wastewater into the system (inlet dissolved oxygen 

concentration and saturated oxygen concentration also shown). .................................................148 

Figure 7-7: The change in oxidation-reduction potential as a function of time at the wetland outlet up to 10 

weeks after the introduction of artificial wastewater into the system (avg. inlet potential = 109.2 ± 

42.4). ............................................................................................................................................148 

Figure 7-8: (a) Total organic carbon concentration profile (avg. inlet [TOC] = 78.6 ± 11.0 mg/L) and (b) total 

nitrogen concentration profile (avg. inlet [TN] = 35.1 ± 5.0 mg/L) at 24cm below the bed surface 

over 5 weeks in July 2016. ..........................................................................................................149 

Figure 7-9: (a) Dissolved oxygen concentration profile (avg. inlet [DO] = 8.80 ± 0.37 mg/L) and (b) redox 

potential profile 24cm below the bed surface over 5 weeks in July 2016. ..................................150 

Figure 7-10: The TOC hydraulic loading density profile for the week 15th – 21st July 2016 (key in top right-hand 

side indicates range of hydraulic loading rates in g/m3.d). ..........................................................154 

Figure 7-11: The TN hydraulic loading density profile for the week 15th – 21st July 2016 (key in top right-hand 

side indicates range of hydraulic loading rates in g/m3.d). ..........................................................154 

 

  



Page | 13  

 

List of Tables 

Table 2-1: Common species of macrophyte (Brisson and Chazarenc, 2008) .....................................................29 

Table 2-2: Investigation of the removal efficiencies of various pollutants between Phragmites australis and a 

sub-species of Typha in a horizontal subsurface flow mesocosms (Brisson and Chazarenc, 2008)

 .......................................................................................................................................................32 

Table 2-3: The primary contaminants and their mechanisms of removal from CW systems (Vymazal, 2005, 

Brisson and Chazarenc, 2008) .......................................................................................................35 

Table 2-4: Classes of chemical reactors (Froment et al., 2011) .........................................................................40 

Table 2-5: The Biomimetic Ecological Principles .............................................................................................55 

Table 2-6: Summary of successful biomimetic wetland designs (Dama-Fakir et al., 2015) ..............................57 

Table 3-1: Legal limits on environmental discharges for a treatment plant of Class 2 in Germany (based on an 

inflow of 300–600 kg·d−1 BOD5) ..................................................................................................67 

Table 3-2: Feed water and artificial wastewater inflow loading rates during 2015 ...........................................68 

Table 3-3: Feed water and artificial wastewater inflow loading rates during 2016 ...........................................68 

Table 4-1: Average flow rates used for hydraulic tracer studies as well as the nominal retention time for each 

experiment. ....................................................................................................................................84 

Table 4-2: Dynamic sampling regime employed for the impulse tracer studies where X indicates a sample being 

taken ..............................................................................................................................................85 

Table 4-3: Dynamic sampling regime employed for the step change tracer studies where X indicates a sample 

being taken .....................................................................................................................................86 

Table 4-4: Calculated hydraulic parameters for the planted and unplanted systems using the impulse and step 

change modelling methodologies ..................................................................................................88 

Table 4-5: Comparison of hydraulic parameters for the unplanted system, scaled to a flow rate of 4.5 l/min ..89 

Table 4-6: Comparison of hydraulic parameters for the planted system, scaled to a flow rate of 4.5 l/min ......89 

Table 5-1: Primary contaminants and mechanisms of removal from CW systems ..........................................101 

Table 5-2: Average bed temperature, pH, TOC and TN concentration with the saturated oxygen concentration 

(fresh water, 760 mm Hg) at the time of the baseline measurements ..........................................106 

Table 5-3: Various hydraulic parameters describing the behaviour of the pilot CW as determined from the 

impulse-response tracer test and system RTD .............................................................................111 

Table 6-1: A comparison of steady and non-steady systems with regards to flow conditions, retention time and 

standard residence time distribution theory .................................................................................120 

Table 6-2: Time to maximum peak concentration at various distances along the flow path (2015 versus 2016)

 .....................................................................................................................................................131 



Page | 14  

 

Table 6-3: Peak time and hydraulic moments at the wetland outlet using standard and variable flow 

methodologies (sample distance = 6.0 m; system volume = 1009 L) ..........................................136 

Table 7-1: Average baseline pH, TOC concentration and TN concentration with the average temperature and 

saturated oxygen concentration (fresh water, 760 mm Hg) in the pilot CW ...............................141 

Table 7-2: Inlet temperature and dissolved oxygen concentration with the saturated oxygen concentration (fresh 

water, 760 mm Hg) at the baseline ..............................................................................................141 

Table 7-3: Average weekly total organic carbon and total nitrogen hydraulic loading rates and percentage 

removal (day of sampling is the second date mentioned) ............................................................146 

Table 7-4: Estimated internal TOC and TN hydraulic loading 12cm below the wetland surface ....................153 

Table 7-5: Estimated internal TOC and TN hydraulic loading 24cm below the wetland surface ....................153 

Table 7-6: Estimated internal TOC and TN hydraulic loading 36cm below the wetland surface ....................153 

 

  



Page | 15  

 

List of Nomenclature 

ɑ   exposed wetland surface area 

A   wetland cross sectional area  

B   width of the wetland 

CA0   concentration of species A in the feed  

CA   concentration of species A at the system outlet 

Cmax  concentration of dye as t → ∞ 

C(t)   time variant tracer concentration (for a steady flow system) 

C(θ)  dimensionless concentration (for a steady-flow system) 

C(Φ)  dimensionless concentration (non-steady flow system) 

d   dispersion index 

D   dispersion number 

Dp    equivalent spherical diameter of the packing material 

e   effective volume utilisation ratio 

E(t)   residence time distribution function (for a steady flow system) 

E(θ)  normalized residence time distribution (for a steady-flow system) 

E(Φ)   dimensionless residence time distribution (non-steady flow system) 

ET   water lost by evapotranspiration  

F(t)   cumulative distribution function 

F(θ)  Normalized cumulative distribution function 

h   wetland water level 

kTOC  reaction rate constant for the decomposition of TOC 

kTN   reaction rate constant for the transformation of TN 

L   length of the wetland 

M, mi  total mass of tracer injected (at the start of the flow test) 

M0   zeroth moment of the residence time distribution 

M1   first moment of the residence time distribution (equivalent to t̅m) 

M2   second moment of the residence time distribution 

n   number of tanks in series (T-I-S model) 

N   fractional recovery of tracer material 

Pe   Peclet number 

Q   steady-state volumetric flow rate 

Q̅in   average inlet volumetric flow rate (assuming a steady-flow system) 

Qinternal, Qi estimated local flow rate at a specified distance from the wetland inlet 



Page | 16  

 

Q̅out   average outlet volumetric flow rate (assuming a steady-flow system) 

Qin(t)  time-variant inlet volumetric flow rate (non-steady flow system) 

Qout(t)   time-variant outlet volumetric flow rate (non-steady flow state) 

Q̅sys   average volumetric flow rate (assuming a steady-flow system) 

qin   inlet hydraulic loading rate  

qin   outlet hydraulic loading rate  

R   percentage recovery of tracer 

Rep   Reynold’s number for packed beds 

ri   rate of formation (ri > 0) or consumption (ri < 0) of species i 

t   time after injection, or age 

t0   time signalling the start of a tracer experiment, or time of injection 

tf   time marking the end of the flow test 

t10   time taken 10% of the injected tracer mass to reach the system outlet 

t90    time taken for 90% of the injected tracer mass to reach the system outlet 

t99   time taken for 99% of the injected tracer mass to reach the system outlet 

tm, t̅m  mean residence time (1st moment of the RTD)  

tp, tpeak  time corresponding to the maximum peak concentration recorded at the wetland outlet 

Us, us   superficial fluid velocity  

V, Vsys  total system fluid volume (taking into account material voidage) 

Veff   effective volume utilisation (volume of fluid inside reactor, excluding the dead volume) 

V(t)  time-variant system volume 

Vout(t)  cumulative discharge volume 

v0, �̇�  system volumetric flowrate 

xi   position along the length of the wetland, where i = 1 - 6 and designates a sampling point 

X   conversion or extent of reaction 

Greek and Special Characters 

ε  voidage of the gravel 

δ  increase in depth of water due to rainfall 

λ  hydraulic efficiency 

Φ  dimensionless flow weighted time 

ρ  fluid density  

ρveg  vegetation density 

σ2  variance (2nd moment of the RTD) 



Page | 17  

 

τ  nominal residence time, or hydraulic retention time, for a steady flow system 

θ  normalized time for a steady flow system 

θi  dimensionless short-circuiting index  

θpeak  normalized peak time 

μ   fluid viscosity  

Chemical Species Formulae 

C   carbon 

CH2O  formaldehyde 

CO2  carbon dioxide 

CO3
2-  carbonate 

H+   hydronium ion 

HCO3
-  bicarbonate 

H2O  water 

N2   nitrogen gas 

N2O  nitrous oxide 

NH3  ammonia   

NH4
+  ammonium 

NO2
-  nitrite 

NO3
-  nitrate 

O2    oxygen gas 

OH-  hydroxide 

PO4
3-  phosphate 

 

 

  



Page | 18  

 

List of Acronyms 

AMD   acid mine drainage 

ANAMMOX anaerobic ammonia oxidation 

AWW   artificial wastewater 

BOD   biological oxygen demand 

BTC   (concentration) breakthrough curve 

COD   chemical oxygen demand 

CSTR   continuous stirred tank reactor 

CW    constructed wetland 

DEA   Department of Environmental Affairs 

DNRA   dissimilatory nitrate reduction, or nitrate ammonification 

DO    dissolved oxygen 

DST   Department of Science and Technology 

DWS   Department of Water and Sanitation 

FC    faecal coliforms 

HRT   hydraulic residence time 

HSSF   horizontal sub-surface flow 

IMWaRU  Industrial and Mining Wastewater Research Unit 

IC    inorganic carbon 

MDI   Morril Dispersion Index 

MI    Moment Index 

Mpre   pre-nominal moment about the nominal divide 

NDP   National Development Plan 

NPOC   non-purgeable organic carbon 

NRM   National Resource Management programme 

NWRS   National Water Resource Strategy 

PFR   plug flow reactor 

RTD   residence time distribution 

SANBI   South African National Biodiversity Institute 

SIP    Strategic Integrated Project 

SPACE   Systems for People to Access a Clean Environment 

TC    total carbon 

T-I-S   Tanks-In-Series model 

TKN   total Khejdahl nitrogen 



Page | 19  

 

TN    total nitrogen 

TOC   total organic carbon 

TP     total phosphorus 

TSS   total suspended solids 

VF    vertical flow 

WRC   Water Research Commission 

WSA   water source area 

 

  



Page | 20  

 

      Introduction 

1.1. Background 

1.1.1. The water conundrum in South Africa 

South Africa is a water scarce country – ranked 30th driest globally (GreenCape, 2016) – with an annual rainfall 

of just 50% of the world average (WWF-SA, 2016). Moreover, precipitation is seasonal and fluctuates (WWF-

SA, 2016) and rainfall patterns could be further impacted by climate change (GreenCape, 2016). As a 

consequence, South Africa’s fresh water supply in its catchments, rivers, wetlands and aquifers is not guaranteed 

to be replenished at the time and to the extent required.   

South Africa’s water source areas (WSAs) – the land areas that produce run-off into the river systems – are 

located in a belt along the coastal region of the country, inland toward the mountainous Drakensberg area and 

then in the north along the Mozambican border (WWF-SA, 2016). The WSAs are of key importance because 

the country relies predominantly on surface water from the river systems into which they drain. At present, only 

16% of the WSAs are protected in national parks, nature reserves and nature conservancies. This percentage is 

too low, given that these areas are some of South Africa’s most valuable natural assets. It is concerning that 

some of the WSAs in the Drakensberg, Mfolozi and Mpumalanga regions are overlapped by coalfields (WWF-

SA, 2016). Coal mining and coal fired power contribute significantly to the South African economy (DOE, 

2018, Koko, 2018).  This is just one example of the challenge that faces multiple industries and government 

departments alike: ensuring that productivity is maximized within the constraints of sufficient protection of the 

natural environment. 

1.1.2. Contamination of South Africa’s water resources 

Poor water quality has a major negative impact on the livelihood of all South African citizens, as well as the 

surrounding ecosystem. Water quality in South Africa’s natural resources has declined over the previous two 

decades; in some cases to such an extent that the water has become a serious health threat. The primary 

contributors to the pollution of the country’s water resources are:  

 Release of raw sewage 

 Acid mine drainage 

 Release of untreated industrial effluent 

 Excess nutrients from agricultural runoff (WWF-SA, 2016) 

Exacerbating the problem is the number of municipal water service authorities plagued by poor service delivery 

and any of a range of factors, including increased urbanization, aging of the engineered infrastructure, lack of 

maintenance and shortage of skills could be blamed. To compound this, many inhabited areas of the country are 



Page | 21  

 

either in remote locations or are unplanned, rapidly expanding informal settlements; neither of which are 

serviced by the existing, although deteriorating, water treatment facilities (WWF-SA, 2016, Swartz, 2009). 

Without urgent attention, South Africa could be faced with a serious water quality and availability crisis. 

1.1.3. South African water legislation 

South Africa’s constitution declares that access to clean, safe water is a basic right for every citizen. The manner 

in which this constitutional right is upheld is also important because the unique constitution also grants the 

environment the basic right to water. Hence, provision of water in South Africa becomes a careful balance 

between adequate supply and protection of the country’s natural water resources (WWF-SA, 2016). 

South Africa’s water resources are governed by the Water Services Act of 1997 and the National Water Act of 

1998 (WWF-SA, 2016, Govt.SA., 2017). The South African Department of Water and Sanitation (DWS) is 

mandated to protect, manage, develop and control the use of the country’s water resources. Closely related to 

this is the provision of sanitation services; also falling under the DWS umbrella. In line with the requirements 

of the two water acts and government’s National Development Plan (NDP), the DWS issued the revised National 

Water Resource Strategy (NWRS2). The NWRS2 outlines how the department and other key role players will 

achieve effective management of the national water resources. The strategy includes security of water supply, 

managing environmental degradation and curbing pollution of water resources.  

The National Resource Management Programme (NRM) was initiated in 1995 by the South African Department 

of Environmental Affairs (DEA) with the ‘working for water’ initiative. This programme aimed to conserve 

water by eliminating invasive alien plant species. Subsequently, the NRM has expanded to include the ‘working 

for fire’, ‘working for forests’, ‘working for wetlands’ and ‘working for ecosystems’ programmes. The country 

has, since 2002, achieved a 5% improvement in the provision of fresh water to households. However, there are 

still approximately 2.5 million people without access to this essential resource (Govt.SA., 2017) and new and 

innovative programmes will be required to reduce this figure further.   

1.1.4. Sustainable water infrastructure and the green economy 

The increasing demand for water, coupled with the low average annual rainfall input, means that planning and 

development are essential to increase capacity and maintain an adequate supply of fresh water, but all too often 

this is done without sufficient consideration of the ecological infrastructure. The DEA and the South African 

National Biodiversity Institute (SANBI) define ecological infrastructure as “the nature-based equivalent of built 

infrastructure”. In other words, the ecosystems, as a result of their natural functionality, provide essential 

services to society and support socio-economic development in terms of, for example, the provision of water 

and soil and climate control. The DEA and SANBI have, thus, announced the 19th Strategic Integrated Project 



Page | 22  

 

(SIP 19) in order to rehabilitate, develop and protect “Ecological Infrastructure for Water Security” (WWF-SA, 

2016). This project is still in its early phases, but pilot projects are underway (SANBI, 2014).  

As South Africa works towards developing its ecological infrastructure, opportunity arises to move from 

centralized to decentralized wastewater treatment facilities; examples of which include point-of-use household 

technologies (SEAL, 2018, CWC, 2015, Nimbus, 2017), rainwater harvesting installations, grey-water green 

roofs and water recycling and reclamation systems (GreenCape, 2016). The popularity of decentralized systems 

is growing as a result of the need to: 

 reduce or eliminate transport and pumping costs 

 improve access to water and sanitation services; particularly in remote areas and establishments that are 

not connected to the sewage infrastructure 

 prevent pollution of water bodies and   

 find solutions that are easy to install, quick to implement, span a broader range of treatment capacities, 

have minimal power requirements and have low maintenance costs (GreenCape, 2016). 

1.1.5. The role of constructed wetlands and biomimicry 

Constructed wetlands (CWs) are engineered systems, designed to utilize the soil, vegetative and microbial 

processes of natural wetlands to assist in the treatment of wastewater of different origins (Vymazal, 2005). CWs 

have been identified as a viable, green technology solution to water management in the following areas:  

 regulation of water supply 

 drought alleviation  

 regulation of water quality (and particularly for biological and temperature control) 

 purification of water (GreenCape, 2016) 

Biomimicry is defined as “the practice of learning from and then emulating nature’s genius to solve human 

problems and create more sustainable solutions” (Biomimicry.SA., 2015). As the types of wastewater become 

more diverse, so the need for more economical, efficient and robust systems to cope with the treatment demands 

emerges. The biomimetic principles (incorporating both the ecological and life’s principles) (Todd and 

Josephson, 1996, Benyus, 2017) provide valuable tools to inform the design of improved CW systems. 

Biomimetic wetlands are self-contained, ecologically engineered systems where waste streams are recycled 

wherever practicable. In addition, they can be a source of nutrient-rich fertilizers and renewable energy and can 

act as carbon sinks, air quality regulators, temperature regulators and habitats for biodiversity (Dama-Fakir et 

al., 2015). 

The Department of Science and Technology (DST) is at the helm of creating awareness around and moving 

forward with the ecological infrastructure work initiated by the DWS and SANBI. In addition, stakeholders in 



Page | 23  

 

environmental consultancy, research institutions, the Water Research Commission (WRC) and Biomimicry SA 

are coordinating research projects to develop biomimetic design solutions for the country’s water-related 

challenges. There is real scope for CWs to provide a valuable addition to South Africa’s ecological 

infrastructure. They offer all of the advantages of decentralized systems and, when designed using biomimetic 

principles, can offer benefits beyond just water remediation. A case in point is the successful prototype in the 

Langrug informal settlement close to Franschoek (Biomimicry.SA., 2018). This is one of the first Systems for 

People to Access a Clean Environment (SPACE) projects, initiated five years ago by the Western Cape 

government and carried out in full co-operation with the local community. Langrug has been converted into an 

Eco-machine (JTED, 2014, U.S.EPA, 2002), which uses biomimicry to address water purification, stormwater 

and solid waste management, as well as provide potential for revenue generation (WWF-SA, 2016). Overall, 

CWs have an important role to play in expanding the ecological infrastructure and biomimicry offers a more 

holistic and sustainable approach to the design and implementation of these systems. 

 

1.2. The relevance of this research 

Much of the literature views CWs as “black boxes” and an in-depth knowledge of the complexities and 

interdependencies of the chemical, hydraulic, kinetic and microbiological processes is still lacking; even though 

more recent studies have begun to report data from various internal of locations along the length of the wetland 

bed (Sheridan et al., 2014a, Bonner et al., 2017b, Bonner et al., 2017a). This research focuses on an in-depth 

hydraulic characterization and baseline physico-chemical description of a newly established pilot-scale CW; 

followed by a five month period of feeding artificial domestic wastewater into the wetland and continually 

monitoring the chemical and physical changes. In addition, external climatic conditions and vegetation 

development will be monitored. The analytical procedure involves dividing the CW into a grid of sample ports: 

seven ports down the length of the bed, each of which is divided into three depths. The multitude of sampling 

locations and samples will provide insight into the internal spatial development and operation of the wetland 

system. More specifically, the hydraulic characterization will describe fluid flow behaviour throughout the 

wetland and provide insight into the validity of making steady-state mathematical assumptions in a real, 

dynamic system. The chemical characterisation will provide insight into the kinetics of carbon and nitrogen 

transformation. It is envisaged that this research may enhance the existing body of knowledge on design 

methodology and visual representation of CW systems by including more information on the internal fluxes 

and, thus, act as a framework for future research and development.  

 



Page | 24  

 

1.3. Research objectives 

1. To use impulse-response tracer tests to determine the hydraulic characteristics of the Industrial and 

Mining Water Research Unit’s (IMWaRU) pilot-scale CWs (namely the residence time distribution, 

velocity profiles, reactor behaviour, location of dead zones and areas of short-circuiting/bypass) and 

then to compare, and possibly validate, the findings using step-change tracer studies in the same systems 

(in conjunction with Ricky Bonner).  

2. To investigate the hydraulic and chemical kinetic performance of a pilot-scale CW in the start-up phase 

at the Helmholtz Centre for Environmental Research (Helmholtz UFZ) in Leipzig, Germany. The 

investigative method was to be chosen based on the conclusions drawn from objective 1. 

3. To collate and analyse the data collected and develop a framework for improved CW design from the 

perspective of:  

a. optimizing fluid flow pathways  

b. maximizing fluid residence time in the areas of the bed where the rates of contaminant removal 

are greatest   

4. To compare the impulse-response performance of a fluoride probe with that of a fluorescent chemical 

tracer to investigate the suitability of a salt-based tracer.  

5. To determine whether micro-organisms may be used as reliable tracers. 

 

1.4. Structure of the thesis 

This thesis comprises 8 chapters plus the appendices of experimental data. Chapters 4, 5 and 6 are made up of 

published papers in Ecological Engineering, Water SA and Science of the Total Environment, respectively. The 

authors plan to submit the experimental findings in Chapter 7 for publication. Certain sections of the theory 

presented in the literature review (Chapter 2) and the materials and methods (Chapter 3) may be repeated in 

similar sections of the papers. The author requests the reviewers’ patience in this regard.  

The appendices contain the experimental data, Matlab code and other supplementary information to support this 

thesis. More data and information is given on the CDs and anything additional can be made available on request. 

The experimental data supporting the paper in Chapter 4 is presented in the thesis written by Ricky Bonner and 

submitted to the Faculty of Engineering and the Built Environment of the University of the Witwatersrand. 

 

  



Page | 25  

 

1.5. List of Publications from this study 

i. Bonner, R., Aylward, L.A., Kappelmeyer, U. and Sheridan, C.M. 2017. A comparison of three different 

residence time distribution modelling methodologies for horizontal subsurface flow constructed 

wetlands. Ecological Engineering, 99, 99 – 113.  

ii. Bonner, R., Aylward, L.A., Kappelmeyer, U. and Sheridan, C.M. 2017. Heat as a hydraulic tracer for 

horizontal subsurface flow constructed wetlands. Journal of Water Process Engineering, 16, 183 – 192.  

iii. Aylward, L.A., Kappelmeyer, U., Bonner, R., Hecht, P. and Sheridan, C.M. 2017. Investigation into 

the kinetics of constructed wetland degradation processes as a precursor to biomimetic design. Water 

SA, 43(4), 655 – 665. 

iv. Luciana Schultze‐Nobre, L., Wiessner, A., Bartsch, C., Paschke, H., Stefanakis, A.I., Aylward, L.A. 

and Kuschk, P. 2017. Removal of dimethylphenols and ammonium in laboratory‐scale horizontal 

subsurface flow constructed wetlands. Engineering in Life Science, 17 (12), 1224 – 1233. 

v. Kappelmeyer, U. and Aylward, L.A. 2018. Chapter 2 Rhizospheric processes for water treatment: 

background principles, existing technology and future use. In: Artificial or constructed wetlands – a 

suitable technology for sustainable water management. Durán-Domínguez-de-Bazúa, M.C., Navarro-

Frómeta, A.E. and Bayona, J.M. (Eds.). CRC Press Taylor & Francis Group, Mexico.  

vi. Aylward, L.A., Bonner, R., Kappelmeyer, U. and Sheridan, C.M. 2019. Hydraulic study of a non-

steady horizontal sub-surface flow constructed wetland during start-up.  Science of the Total 

Environment, 646, 880 – 892. 

vii. Bonner, R., Aylward, L.A., Kappelmeyer, U. and Sheridan, C.M. 2018. Combining tracer studies and 

biomimetic design principles to investigate clogging in constructed wetlands. Water SA, 44(4), 764 – 

770. 

  

https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Schultze-Nobre%2C+Luciana
https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Wiessner%2C+Arndt
https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Bartsch%2C+Cindy
https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Paschke%2C+Heidrun
https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Stefanakis%2C+Alexandros+I
https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Aylward%2C+Lara+A
https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Kuschk%2C+Peter


Page | 26  

 

      Literature Review 

2.1. Introduction to constructed wetlands 

2.1.1. Natural wetland ecosystems 

Natural wetlands are areas intermediate between water and land; examples of which include swamps, marshes 

and bogs. Figure 2-1 gives the definition and classification of natural wetland systems (Kivaisi, 2001). 

 

Figure 2-1: The definition of natural wetlands, their distinguishing characteristics and their classification according to the 

dominant plant species (Kivaisi, 2001).  

 

Natural wetlands have a number of important ecological functional roles; namely water purification, water 

capture and storage, flow regulation, flood attenuation and shoreline stabilization and protection. These roles 

are represented diagrammatically in Figure 2-2. Without their diversity of vegetation, microbial and aquatic 

species and wildlife, wetland ecosystems would not support these functions (Kivaisi, 2001). It is, thus, not 

surprising that natural wetlands should be the inspiration for a solution to water treatment challenges; namely 

the constructed wetland.  



Page | 27  

 

 

Figure 2-2: The functional roles of a natural wetland (the dashed arrows represent contaminant removal by the wetland 

system) (Kivaisi, 2001). 

 

2.1.2. Historical development of constructed wetlands 

Historically, wetlands were considered wastelands of little use and poor aesthetic appeal and often converted 

into landforms of supposed better value. However, as knowledge grew of how vital a part of the natural 

environment these ecosystems were, focus turned towards their preservation and protection and the development 

of CWs for wastewater remediation (Reed, 1993). Quoting Jan Vymazal of the Duke University Wetland Centre, 

“CWs are engineered systems that have been designed and constructed to utilize the natural processes involving 

wetland vegetation, soils and the associated microbial assemblages to assist in treating wastewaters” (Vymazal, 

2005); often acting as polishing systems following the primary and secondary water treatment stages (Brix, 

1994b).  CWs have three characterising features:  

1. A soil, gravel or sand filled bed (Brix, 1994b, Haberl et al., 2003).  

2. One of more types of specialised macrophyte, which are adapted to the hydric and oxygen depleted 

conditions, known to aid in the removal of various types of wastewater contaminants and withstand 

harsh chemical environments (Brix, 1994b, Haberl et al., 2003). 



Page | 28  

 

3. Specific hydrologic conditions in that the soil matrix be saturated with water up to a depth of 2m, either 

continually or intermittently, during the vegetation’s growth season (Haberl et al., 2003). 

Organic matter and nutrients are removed from the water flowing through the wetland owing to its ability to 

transform or store these species (Brix, 1994b). Originally, CWs treated domestic and municipal sewage, but 

their use was later extended to industrial and agricultural wastewater, landfill leachate, acid mine drainage 

(AMD) and stormwater and mining run-off remediation (Vymazal, 2005, Reed, 1993). High levels of removal 

of sulphate, manganese, iron and certain heavy metals from acid mine water have also been observed (Smith, 

1997). 

2.1.3. Types of constructed wetlands 

CWs are broadly categorized according to the class of dominant macrophyte and then the major type of fluid 

flow through the system. This classification is illustrated by means of the diagram in Figure 2-3. Free-floating 

macrophyte systems were developed largely as a means of enhancing the performance of stabilization ponds; 

whereas submerged macrophyte systems were intended to act as polishing systems post the primary and 

secondary water treatment stages. These two classes of CWs are not widely used and it is the emergent 

macrophyte systems that dominate; hence their further classification based on the primary fluid flow path (Brix, 

1994b).  

 

Figure 2-3: Classification of constructed wetlands (Brix, 1994b). 

 

 



Page | 29  

 

The horizontal subsurface flow (HSSF) CW emerged as the preferred system for wastewater remediation and 

wetlands of this type find were most commonly installed. However, as the number of classes of wastewaters 

requiring treatment continued to increase, it became evident that single-stage HSSF CW systems were not 

always adequate, depending on the level of treatment required. This lead to the development of hybrid CWs. 

Hybrid CWs consist of two or more classes of wetlands connected together, in various configurations of series 

or parallel units, to better utilize the treatment capabilities of the individual systems. For example, nitrogen 

removal efficiency is poor in single-stage HSSF systems, but HSSF and vertical flow (VF) hybrid CWs improve 

the overall efficiency of nitrogen removal (Vymazal, 2005). 

2.1.4. Wetland vegetation 

There have been a number of debates over the importance of CW’s being planted versus unplanted, but the 

majority of studies have concluded that wetland vegetation play a vital part in wastewater remediation in these 

systems (Brix, 1994a, Brisson and Chazarenc, 2008, Brix, 1997, Stottmeister et al., 2003, Tanner, 2001). 

Types of wetland vegetation 

CWs may be planted with one or more types of macrophyte. The term macrophyte broadly describes 

macroscopic, aquatic species which may be emergent, submerged or free-floating. The distinguishing feature 

of the macrophytes is their ability to survive in water-saturated soils or in aquatic environments. In addition, 

they also show fast growth, quick establishment, high rates of biomass generation and have well-developed root 

and support systems (Brix, 1994a, Brisson and Chazarenc, 2008). Table 2-1 summarizes the macrophytes most 

commonly used or studied in treatment wetland systems. Brisson and Chazarenc (2008) reviewed 35 different 

experiments involving a variety of wetland species. Phragmites australis and Typha latifolia appear in the most 

studies; followed by Typha angustifolia and Schoenoplectus validus.  

Table 2-1: Common species of macrophyte (Brisson and Chazarenc, 2008) 

Species Sub-species 

Baumea  articulata 

Bolboschoenus  fluviatilis 

Canna  indica 

Carex  lacustris, rostrata 

Commelina  communis 

Cyperus  corymbosus, dubius, grandis, immensus, involucratus, papyrus 

Digitaria  bicornis 

Echinochloa  pyramidalis, cordifolius 

Eriocaulon  sexangulare 

Glyceria  maxima 

Iris  pseudacorus 

Juncus  effusus 

Kyllinga  erectus 



Page | 30  

 

Leptochloa  fusca 

Ludwigia  octovalvis 

Pennisetum  purpureum 

Phalaris  arundinacea 

Phragmites  sp., australis, mauritianus, vallatoria 

Sagittaria  latifolia 

Schoenoplectus  
sp., acutus (Syn. Scirpus acutus), mucronatus (Syn. Scirpus mucronatus), 

pungens (Syn. Scirpus pungens), validus (Syn. Scirpus validus) 

Scirpus  sp., atrovirens, cyperinus, globulosus, grossus 

Spartina  patens 

Stenotaphrum  secundatum 

Typha  sp., angustifolia, capensis, domingensis, latifolia, orientalis, subulata 

Urochloa  mutica 

Vetiveria  zizanoides 

Zizania  latifolia 

Zizaniopsis  bonariensis 

 

Role of vegetation in contaminant removal 

Wetland plants have important physical and metabolic functions in CWs (Brix, 1994a), which will be discussed 

in this section. The physical functions are the following: 

 Wetland surface stabilization, particle filtration and light attenuation 

Primarily, wetland vegetation prevents erosion of the surface layers of soil (Brix, 1994a); although this 

function is somewhat less important in gravel-based HSSF CWs which are bounded by impermeable 

walls. The dense root region created by macrophyte species aids in trapping suspended solids travelling 

through the wetland. The aerial vegetation reduces the intensity of ultra-violet light reaching the surface 

of the wetland; thereby preventing excessive algal growth (Brix, 1994a).  

 Preventing clogging in vertical flow systems 

 Insulating the wetland 

Decaying plant matter forms a litter layer on the wetland surface which shelters the soil from frost 

during winter and keeps it cool and moist during summer (Brix, 1994a). 

 Provision of oxygen in the root zone (rhizosphere) 

The water saturated state of a CW lends itself towards being anaerobic and supporting anaerobic 

mechanisms of pollutant removal. However, macrophytes have efficient mechanisms of internal oxygen 

transport from their aerial organs to their submerged roots; thereby providing an oxygen-rich zone 

which is essential for aerobic degradation processes and nitrification in the rhizosphere (Brix, 1994a). 

Most of the oxygen is released from the apical region (region of new growth) behind the root apex, or 

tip (Armstrong, 1979). Some oxygen may leak into the rhizosphere from young fine laterals protruding 

from the root base, but no oxygen is released from established roots or rhizomes (Armstrong and 



Page | 31  

 

Armstrong, 1988). The vegetation provides in excess of 90% of the oxygen available in the root zone 

via their various internal transport mechanisms (Reddy et al., 1989).  

 A habitat for wildlife 

 Aesthetic appeal 

Flowering wetland plants, such as the Iris pseudacorus (Yellow Flag), Canna indica (Canna lily or 

Indian shot) (Brix, 1994a) or Zantedeschia aethiopica (Arum lily) (Bonner et al., 2017b) can be chosen 

to increase the visual appeal of a CW.  

The metabolic functions include: 

 Providing a large surface area (specifically in the root zone) for microbial (biofilm) attachment and 

growth 

 Plant uptake of nutrients from the wastewater 

Due to their characteristic high productivity, wetland plants absorb and store larger amounts of nutrients in their 

biomass. However, this amount is typically small in comparison to the influent contaminant loading from the 

wastewater (Brix, 1994a, Haberl et al., 2003).   

Macrophyte species selection 

A number of factors must be considered and off-set against one another when choosing the most suitable 

vegetation for a planted CW, such as root depth, wastewater loading tolerance and plant health and productivity. 

Wetland design, the type of wastewater and contaminants, climate and the time available for wetland 

establishment also play a role (Brisson and Chazarenc, 2008). For example, Phragmites australis take up to 

three years to reach full maturity (Vymazal and Krőpfelová, 2005). Moreover, if aerobic degradation processes 

and, thus, oxygenation of the rhizosphere are important for the type of wastewater being treated, then plants 

supporting high convective through-flow of gases should be chosen (Brix, 1994a). Phragmites australis are a 

good example of macrophyte which transport gases via convective through-flow (Armstrong and Armstrong, 

1991).  

Brisson and Chazarenc (2008) reviewed 35 studies regarding the influence of macrophyte species on pollutant 

removal efficiency in various types of CW. The focus was largely on microcosm studies comparing at least two 

species of macrophyte cultivated under identical experimental conditions and measuring the influent and 

effluent concentrations as an indication of pollutant removal efficiency. Microcosm experiments are not truly 

representative of real systems and replication was minimal (and often carried out in the same unit rather than 

across identical units), so the results should be treated with caution (Fraser and Keddy, 1997). Nevertheless, 

they do give an indication of the relative performance between common wetland species within the constraints 



Page | 32  

 

of reasonable cost and time. Validation of the microcosm findings in a pilot-scale or full-scale wetland system 

is still required (Brisson and Chazarenc, 2008).  

The pollutants under consideration in the studies reviewed by Brisson and Chazarenc (2008) were total 

suspended solids (TSS), organic matter in terms of chemical or biological oxygen demand (COD or BOD), total 

Khejdahl nitrogen (TKN), total nitrogen (TN), nitrate (NO3
-), total phosphorus (TP), phosphate (PO4

3-) and 

ammonium (NH4
+). Overall, they found that removal efficiency for a number of contaminants was different 

from species to species, but no generalizations (based on definite correlations across the board) could be 

formulated. 

Differences in nitrogen (and particularly NO3
-) removal efficiencies between species were most commonly 

observed; most likely because nitrogen is known to be assimilated in the biomass or transformed by aerobic 

microbial activity in the root zone. This is evidence of the advantage of combining more than one species of 

macrophyte in CWs to utilize the differences in removal efficiencies under different conditions. No trends were 

apparent for suspended solids, organic matter and phosphorus removal (Brisson and Chazarenc, 2008). The 

studies involving HSSF mesocosms planted with Phragmites australis are summarized in Table 2-2. In general, 

no difference in removal efficiency was seen between Phragmites australis and Typha latifolia or Typha 

angustifolia, with the exception of experiment 2 where Typha latifolia showed a higher removal efficiency for 

nitrogen and phosphorus. 

Table 2-2: Investigation of the removal efficiencies of various pollutants between Phragmites australis and a sub-species 

of Typha in a horizontal subsurface flow mesocosms (Brisson and Chazarenc, 2008) 

Experiment 

number 
Macrophyte species Removal efficiency observations 

2 
Phragmites australis 

Typha latifolia 

 BOD, COD: no difference 

 Typha lat. Showed higher removal of TKN, TP, NH4
+, PO4

3- 

26 
Phragmites australis 

Typha angustifolia 
 COD, TSS, TKN, TP: no difference 

27 
Phragmites australis 

Typha angustifolia 
 BOD, COD, TSS, TKN, TP, PO4

3-: no difference 

28 
Phragmites australis 

Typha angustifolia 
 COD, TSS, TSS: no difference 

 

The majority of studies reviewed by Brisson and Chazarenc (2008) showed that there were differences in 

removal efficiencies between many species; even though these were experimental condition specific and the 

results should be treated with caution. Nonetheless, the review is evidence that macrophyte selection is an 

important consideration in CW design. It is not yet possible to make a set of general guidelines for macrophyte 

species selection as more research is required into the different mechanisms of contaminant removal between 



Page | 33  

 

the common types of wetland plants and better factors correlating measureable plant features to treatment 

efficiency must be found.  

Importance of root exudates 

As part of their metabolic activities (Marschner, 1995), plant roots secrete a mixture of compounds (Flores et 

al., 1999), collectively referred to as root exudates. These exudates are thought to play a vital role in the 

maintenance of rhizospheric microbial communities, which are, in turn, essential for some of the remediative 

processes occurring in CWs.  

Specialized transporter protein molecules are responsible for the active transport of exudate material across the 

root plasma membrane (Baetz and Martinoia, 2014). Plants can alter the chemical composition of the exudate 

material in response to various internal and external stimuli and it is the make-up of the exudates that largely 

defines the surrounding soil environment and microbial community structure (Baetz and Martinoia, 2014). 

There are few studies devoted to a detailed characterization of the exudate composition according to plant 

species, but they can be grouped broadly into two categories. The first is the low molecular weight class, 

examples of which include amino acids, organic acids, simple sugars and phenolic compounds. The second 

category is the high molecular weight species, such as long chain polysaccharides and proteins (Walker et al., 

2003). The expressed proteins often act as transporter molecules for other root exudate constituents (Baetz and 

Martinoia, 2014).  

Plant root exudates have a number of important functions; namely regulation of the microbial community in the 

surrounding soil, alteration of the soil’s physical and chemical characteristics, plant defence, support of 

symbiotic relationships in the rhizosphere and prevention of invasion by competing plant species (Nardi et al., 

2000). Root exudates are also the medium of transport of photosynthetically fixed carbon from the plant to the 

rhizosphere (Marschner, 1995) and it is believed that they serve as chemical messengers between plant roots, 

the rhizosphere and soil micro-organisms (Walker et al., 2003). As an example, parasitic plants secrete root 

exudates to stimulate the growth of invasive organs which can suffocate or drain nutrients from other plant 

species (Keyes et al., 2000).  

The role of root exudates in plant defence is a vital but complex one (Baetz and Martinoia, 2014). Root 

secretions may contain antimicrobial compounds for the protection of the plant against herbivorous species, 

fungi and pathogens (Nardi et al., 2000). Many of the compounds initially associated with defence were low 

molecular weight species and enzymes, but it has been subsequently discovered that high molecular weight 

proteins and extracellular DNA also play a role by binding, trapping and aggregating pathogenic bacteria (Wen 

et al., 2007a, Wen et al., 2007b, Wen et al., 2009). Exudates can also contain a mix of regulatory compounds. 



Page | 34  

 

Regulators are required, in one instance, for tightly controlling the release of photosynthetically synthesized 

carbon from the plant which, if excessive, is a large carbon drain for the plant (Baetz and Martinoia, 2014). 

2.1.5. Advantages of constructed wetlands 

Provided that an adequate area of land is available at an affordable price, CWs are reliable, cost-effective, 

energy-efficient and low-maintenance systems that pose minimal risk of public exposure to contaminants within 

the system (Reed, 1993). They can also provide various additional benefits; namely carbon sequestration, air 

quality regulation, evapotranspiration, temperature regulation, a habitat for biodiversity and nutrient cycling 

(Dama-Fakir et al., 2015). 

 

2.2. Wastewater quality 

There are various physicochemical parameters which provide information about water quality. For example: 

 A positive redox potential indicates an oxidizing solution. Many pollutants require strong oxidants to 

facilitate their decomposition (Bellingham, 2012) yet some of the oxidative processes (processes 

requiring oxygen) generate sulphur dioxide, nitrate and nitrite. These reduced species produce a drop 

in redox potential which could impede further oxidative decomposition.  

 NH4
+ is a by-product of the decomposition of urea. High levels of NH4

+ can produce an environment 

which is toxic to aquatic organisms.  

 Organic carbon is also an important water quality indicator, for which maximum allowable limits are 

commonly published in environmental legislation; particularly for the chemical oxygen demand (COD) 

and biological oxygen demand (BOD).  

 

2.3. Remediation of wastewater using constructed wetlands 

2.3.1. Contaminant removal mechanisms in constructed wetlands 

Contaminants are removed from CWs via physical, chemical and biological routes (Brix, 1994b). Physical 

removal relies upon sedimentation and entrapment of solids within the wetland bed. Chemical removal involves 

reactions which convert harmful contaminants into less dangerous compounds. Some of these reactions rely 

upon reducing conditions associated with the saturated wetland soils (Haberl et al., 2003) but many are driven 

by biological processes in the root region of the wetland vegetation (Headley and Kadlec, 2007). Table 2-3 

summarizes the primary contaminants and their mechanisms of removal from CW systems.  

 



Page | 35  

 

Table 2-3: The primary contaminants and their mechanisms of removal from CW systems (Vymazal, 2005, Brisson and 

Chazarenc, 2008) 

Contaminant Mechanism of Removal 

Organic Matter Aerobic and anaerobic degradation by bacteria attached to plant roots, rhizomes and 

media surfaces  

Suspended solids Filtration and sedimentation 

Nitrogen 

(ammonia/nitrate) 
Nitrification / denitrification and adsorption (if soil grain is sufficiently fine) 

Phosphorus 

(phosphate) 

Ligand exchange reactions in the presence of iron, aluminium or calcium hydrous 

oxides 

 

2.3.2. Removal of organic carbon 

The potential for degradation of organic compounds is measured by COD and BOD. COD is a measure of the 

oxygen required to oxidise organic matter by any means. BOD measures the amount of dissolved oxygen 

required by micro-organisms to metabolize organic compounds. Therefore, COD is representative of all organic 

matter present, while BOD is an indicator of the organic matter which can only be biologically decomposed. 

BOD and COD are widely utilised sum parameters for the removal of organic matter, but studies into the 

removal of specific types of organic compounds have also been conducted. Examples include hydrocarbons 

(petrochemical compounds), oil and grease, mineral oil, chlorinated volatile organic compounds, aromatic 

compounds, cyanide, glycols, atrazine (a herbicide), trinitrotoluene (TNT), RDX or cyclonite (an explosive) 

and general organics found in dairy farm, pig farm, slaughterhouse and olive mill wastewater plus landfill 

leachate. In these studies, emphasis has been placed on the selection of the particular plant species best suited 

to the removal of the specific organic compound (Haberl et al., 2003). CWs produce BOD5 due to the 

degradation of dead plant matter. As a result, there may be a residual BOD5 within the system and complete 

removal will seldom be achieved (Reed, 1993).  

Although the primary mechanisms of organic matter removal are biological, for example fermentation, aerobic 

respiration, anaerobic respiration and bio-augmentation (Table 2-3), other degradative processes such as 

volatilization, oxidation (photochemical), sorption by plants and sedimentation can also be effective. Research 

into the mechanistic details of the removal pathways for organic compounds is still ongoing. The removal 

efficiency for organic compounds in CWs is high, but will vary depending on the type of organic compound 

and the plant species selected. As such, the genetic modification of plants in such a way as to enhance the uptake 

of a specific organic compound could be one way to improve CW removal efficiency. With these factors in 

mind, and careful consideration of the environmental conditions, CWs can be designed for maximum removal 

efficiency of organic matter (Haberl et al., 2003).  



Page | 36  

 

2.3.3. Removal of nitrogen 

The primary mechanisms of nitrogen removal are microbial processes (Adler et al., 1996). Higher nitrogen 

removal efficiencies are achieved in planted, gravel-based systems (Yang et al., 2001, Brix, 1994a) with 

adequate oxygen, minimal algal growth and sufficient hydraulic residence time (Reed, 1993). Nitrogen is 

involved in many processes and reactions within a constructed wetland, yet often it is only transformed from 

one nitrogenous species to another. The processes that ultimately remove nitrogen from wetland waste water 

are volatilization, denitrification, anaerobic ammonia oxidation (ANAMMOX), adsorption, plant uptake 

coupled with biomass harvesting and burial (Vymazal, 2007).  

Ammonia volatilization 

Volatilization is the loss of ammonia (NH3) gas from the wetland surface to the atmosphere. This is an 

equilibrium reaction represented by Eq. 2-1: 

NH3(g) + H2O(aq) ⇌ NH4
+(aq) + OH−(aq)                                                           [2_1] 

A pH above 7.0 shifts the equilibrium towards the left, while a pH below 7.0 shifts the equilibrium towards the 

right. Hence, losses of NH3 by volatilization are greater at higher pH; particularly above 9.3 (Reddy and Patrick, 

1984). 

Nitrification 

Nitrification is the aerobic transformation of NH4
+ into NO3

- by nitrifying bacteria and is influenced by 

temperature, pH, inorganic carbon (IC) availability, the microbial population, the concentration of NH4
+ ions, 

the dissolved oxygen (DO) concentration and the bicarbonate (HCO3
-) concentration. Nitrification is a two-step 

process, represented by the overall Eq. 2-4: 

NH4
+ + 1.5O2 → NO2

− + 2H+ + H2O                                                            [2_2] 

NO2
− + 0.5O2 → NO3

−                                                                                        [2_3] 

NH4
+ + 2O2 → NO3

− + 2H+ + H2O                                                                [2_4] 

In Eq. 2-2, a nitrite (NO2
-) intermediate is formed by the action of ammonia oxidising bacteria. These bacteria 

draw their energy from inorganic carbon, carbon dioxide (CO2) or carbonate (CO3
2-) (Hauck, 1984, Paul and 

Clark, 1996). NO2
- is then transformed into NO3

- by nitrite oxidizing bacteria (Eq. 2-3), which may also derive 

energy from organic compounds in addition to NO2
- (Paul and Clark, 1996, Schmidt et al., 2001, Schmidt et al., 

2003). In low oxygen environments, only partial nitrification takes place. Eq. 2-2 proceeds in the same way, but 

a lack of oxygen means that NO2
- is transformed into nitrous oxide (N2O) or nitrogen gas (N2) rather than into 

NO3
- (Bernet et al., 2001). 



Page | 37  

 

Denitrification 

Denitrification is the anaerobic transformation of NO3
- to N2. Under anaerobic or anoxic conditions, nitrogen 

acts as an electron acceptor (Reddy and Patrick, 1984), but the rate of denitrification is fastest under anoxic 

conditions. The process, shown in Eq. 2-5, is induced by denitrifying bacteria which use organic carbon, such 

as formaldehyde (CH2O), as a source of energy (Hauck, 1984, Jetten et al., 1997, Paul and Clark, 1996). 

Denitrification is influenced by the DO concentration, redox potential, temperature, pH, soil characteristics, 

presence of organic matter, the concentration of NO3
- and the presence of free water on the wetland surface 

(Vymazal, 2005, Focht and Verstraete, 1977).  

6(CH2O) + 4NO3
− → 6CO2 + 2N2 + 6H2O                                                [2_5] 

Denitrification may also proceed in the presence of oxygen; in which case organic carbon is oxidized (oxygen 

acts as an electron acceptor) and CO2 and water (H2O) are produced as by-products (Reddy and Patrick, 1984).  

Anaerobic ammonium oxidation (ANAMMOX) 

ANAMMOX is the anaerobic transformation of NH4
+ and either NO3

- or NO2
- to nitrogen. The ANAMMOX 

process can be described by Eq. 2-6 and 2-7 (Mulder et al., 1995, van de Graaf et al., 1995). Although both NO3
- 

and NO2
- can act as electron acceptors in the ANAMMOX reaction, NO2

- is the key electron acceptor (Strous 

et al., 1997). 

5NH4
+ + 3NO3

− → 4N2 + 9H2O + 2H+                                               [2_6] 

NH4
+ + NO2

− → N2 + 2H2O                                                                    [2_7] 

Dissimilatory nitrate reduction (DNRA) 

DNRA, also known as nitrate ammonification, is the reduction of NO3
- to NH4

+ by nitrate-reducing or nitrate-

ammonifying bacteria. The process is influenced by the C:NO3
- ratio, the DO concentration and temperature 

(Vymazal, 2007). DNRA is described by three reactions, which are given below, with Eq. 2-8 being a form of 

bacterial cellular respiration (Megonigal et al., 2004).  

NO3
− + H2 → NO2

− + H2O + ATP                                                               [2_8] 

NO2
− + 3H2 + 2H+ → NH4

+ + 2H2O                                                          [2_9] 

NO3
− + 4H2 + 2H+ → NH4

+ + 3H2O                                                        [2_10] 

The NH4
+ produced may be taken up by the wetland plants (high levels of uptake, however, can lead to 

eutrophication), assimilated by micro-organisms or adsorb to various negatively charged surfaces (Vymazal, 

2007).  



Page | 38  

 

Assimilatory nitrate reduction 

Macrophytes, algae and heterotrophic bacteria take up inorganic nitrogen and convert it into organic nitrogen 

containing compounds, such as amino acids. By this process, they assimilate nitrogen. Soluble forms of 

nitrogen, such as NO3
- to NH4

+, are the preferred nitrogen containing species for this process (Vymazal, 2007).    

Ammonification 

Ammonification describes the transformation of organic nitrogen into NH3. Ammonification is biologically 

driven and influenced by temperature, pH, the C:N ratio, nutrient availability, the type of soil matrix and the 

presence of oxygen (Vymazal, 2007); with the rate of ammonification being fastest under aerobic conditions 

(Reddy and Patrick, 1984). Ammonification occurs via a number of intermediate steps, known as deamination 

steps, and may be either an oxidation or reduction process. Oxidative deamination is described by Eq. 2-11 

(Savant and DeDatta, 1982) and reductive deamination by Eq. 2-12 (Rose, 1976). 

amino acids → imino acids → keto acids → NH3(g)                                          [2_11]  

amino acids → saturated acids → NH3(g)                                                   [2_12] 

Nitrogen fixation 

Fixation is an enzyme (nitrogenase) catalyzed reaction by which N2 gas is converted to NH3 (Stewart, 1973) 

under anaerobic, or reduced, conditions (Buresh et al., 1980). Macrophytes, heterotrophic bacteria and 

cyanobacteria are all capable of N2 fixation (Johnston, 1991). 

Adsorption 

NH4
+ ions can adsorb to negatively charged sites on detritus, sediments and soils. This is an equilibrium reaction 

which establishes a balance with NH3 in the wetland water (see Eq. 2-1). The equilibrium can be shifted by a 

change in the concentration of NH4
+ and O2 in the water. For example, nitrification decreases the concentration 

of NH4
+ ions in solution and would drive desorption of ammonia from the sediment (Kadlec and Knight, 1996). 

Organic nitrogen burial 

Detritus is composed partly of organic nitrogen. Nitrogen locked up in peat and decaying organic matter, such 

as leaves, becomes permanently unavailable in the nitrogen cycle, but is only removed from the bed if the 

wetland is cleared of this debris. Organic nitrogen removal has only been reported to contribute to nitrogen 

removal in natural wetlands rather than CWs (Vymazal, 2007).   

2.3.4. Removal of phosphorus 

Phosphorus is largely removed by chemical adsorption; particularly onto iron, aluminium and calcium hydrous 

oxides (Adler et al., 1996). The removal of phosphorus may be ineffective, largely due to limited contact time 



Page | 39  

 

between the flowing water and the soil (Reed, 1993), but higher removal efficiencies have been reported for 

mineral soil containing wetlands as compared to gravel based systems (Yang et al., 2001). 

2.3.5. Removal of additional contaminants 

The removal of faecal coliforms (FC) is normally insufficient (Reed, 1993). Although high FC removal 

efficiencies, in the region of 90%, may be achieved (Neralla et al., 2000), this is largely dependent on favourable 

climatic conditions and levels of FC reduction that can be achieved in wetland systems often do not comply 

with regional discharge standards (Smith et al., 2005), (Kassim, 2007). In addition, removal of pathogenic 

microorganisms and viruses are seldom killed. Therefore, additional disinfection of the effluent may be required, 

such as chlorination (Neralla et al., 2000). 

 

2.4. Constructed wetland hydraulics 

CW hydraulics broadly describes flow behaviour within CW systems. The geometry, as well as various physical 

properties, can influence how water moves through a CW and both should be carefully selected during the 

design process in relation to site-specific loading and treatment requirements. Ultimately, CW performance and 

sustainability will be governed by how successfully designers can optimize and operators can control the 

hydrological regime and flow behaviour (Reed et al., 1995, Persson et al., 1999).  

2.4.1 Chemical engineering reactor theory 

Chemical engineering is the branch of engineering characterised by the commercialization of chemical reactions 

to turn raw materials into a desired product. Hence, the discipline of chemical engineering involves, 

predominantly, the design and operation of chemical reactors as well as the supporting processes to take the 

process from start (material feed) to end (separation and purification of desired product and adequate treatment 

of waste generated) (Levenspiel, 1999b).  

The wide variety of industrial chemical reactions are carried out in various types of chemical reactor. The choice 

of reactor depends on:  

i. the number and sequence of reactions (a single reaction, reactions in series, reactions in parallel, 

complex reactions)  

ii. whether the reactions are to be carried out in a continuous or non-continuous process 

iii. whether the reactions are exothermic or endothermic 

iv. the effect of temperature on the selectivity and reaction kinetics (which governs the distribution of 

products) 

v. whether a catalyst is required (Levenspiel, 1999b) 



Page | 40  

 

Table 2-4 summarizes the main classes of chemical reactors, their operational features and examples of where 

each may be used (Froment et al., 2011). Industrial reactors could also be designed as combinations of these 

types.  

Table 2-4: Classes of chemical reactors (Froment et al., 2011) 

Type of Reactor Operational Features 
Areas of Application (select 

examples) 

Batch reactor 

 Single vessel (may contain an agitator) 

 All reactants are fed into the reactor at once, 

after which the reaction is left to run to 

completion 

 Uniform composition and temperature 

throughout the reactor at any point in time 

 Composition changes with time as the 

reaction proceeds 

 Isothermal or non-isothermal 

 Specialty chemicals 

 Polymers 

 Pharmaceuticals 

 

Semi-batch reactor 

All characteristics of a batch reactor but 

 Reactants are fed into or removed from the 

reactor intermittently 

 Specialty chemicals 

 Polymers 

 Pharmaceuticals 

Plug flow reactor 

(PFR) 

 Tubular reactor 

 Fluid is assumed to move as a “plug” and 

with uniform velocity along the reactor 

 Fluid conditions (velocity, temperature, 

composition) are uniform at any cross section 

along the reactor at a point in time 

 Isothermal or non-isothermal 

 Adiabatic or non-adiabatic 

 Pressure drop can be large 

 May contain a solid packing material 

depending 

 Perfect plug flow (no intermixing between 

successive fluid elements) is an idealized 

case 

 Gasoline production 

 Oil cracking (olefin 

production) 

 Ammonia synthesis 

 

Mixed flow reactor 

or continuous stirred 

tank reactor (CSTR) 

 Reactor is a well agitated vessel 

 Continuous flow process 

 Assume concentration and temperature are 

uniform throughout the reactor (therefore the 

effluent has the physical and chemical 

properties as the contents of the reactor) 

 Opposite extreme to plug flow (perfect 

mixing is an idealized case) 

 Biazzi process (temperature 

control is critical) 

 Polymerization of butadiene & 

styrene (reaction requires 

constant composition) 

 Two-phase reactions 

 Ziegler catalytic reactions & 

Hercules Distillers process 

(catalyst must be kept in 

suspension) 

Fixed bed catalytic 

reactor 

 Tubular reactors packed with solid catalyst 

particles, over which the reactant-containing 

fluid stream flows 

 Steam reforming 

 Water-gas shift reaction 

 Carbon monoxide methanation 



Page | 41  

 

 Catalyst bed is fixed (catalyst particles are 

bound to a surface) 

 Isothermal or non-isothermal 

 Adiabatic or non-adiabatic 

 Pressure drop can be large 

 Reactor must be shut-down to regenerate 

catalyst 

 Ammonia, methanol & sulfuric 

acid synthesis 

 Petroleum refining 

Fluidized bed reactor 

 A portion of solid catalyst particles move 

against the flow of reactant-containing fluid 

 Suitable for highly exothermic reactions 

(require close temperature control) 

 Opportunity to recycle and regenerate 

catalyst in a continuous process 

 Catalytic cracking of gas oil 

(gasoline production) 

 Oxidation of naphthalene 

 Oxychlorination of ethylene 

