Waste stabilisation ponds (WSPs) refer to relatively shallow basin of water utilising the natural phenomena of waste degradation. The energy required to maintain the system is in renewable form and basically derived from the sun. The major biological components of the WSP system are algae and bacteria. Waste stabilisation in these systems is carried out by three major types of natural reactions namely physical, chemical and biological. WSP systems are basically classified into three major types according to the biological reactions occurring within the pond. Aerobic ponds posses a naturally aerated layer enriched with aerobic bacteria, whereas deeper oxygen deficient layers comprised of anaerobic bacteria are found in anaerobic ponds. Facultative ponds, characterised with aerobic layer at the surface, a middle facultative zone and an anaerobic bottom, are the most effective in waste decomposition.

The relationship occurring in facultative ponds is called the ‘Algae-Bacteria Symbiotic Reactions’. The top, aerobic layer of the ponds contains algal cells which, in the present of sunlight, perform photosynthesis and producing oxygen (O₂) gas, according to Eq. (1).

The O₂ gas produced in Eq. (1) is then used by suspended bacteria to oxidise the incoming organic waste, as shown in Eq. (2).

As long as the algae can provide an excess of O₂ above that required by the bacteria, an aerobic environment is maintained. Thus it can be seen that there is interdependence between algae and bacteria, with the algae supplying oxygen required by the bacteria, and the bacteria making available the CO₂ and inorganic nutrients required by the algae. Effluent of facultative ponds normally contains low dissolved organic matter, but high content of algal and bacterial cells which can be removed further in maturation or polishing ponds connected in series to the facultative ponds.

Aquatic weeds are prolific plants growing in water bodies which can create a number of problems due to their extensive growth and with high productivity. Since they exhibit spontaneous growth, they usually infest polluted waterways or water bodies, reducing the potential uses of these water courses. However, turning these aquatic weeds into productive uses, such as in wastewater treatment, in making compost fertiliser, and as human food or animal feed, some of the problems created by aquatic weeds may be minimised. Table 3 describes functions of the plant parts in waste-water treatment mechanisms.

The floating aquatic weeds that have been employed for wastewater treatment are water hyacinth—Eichhomia crassipesy, duckweeds (Woifia arrhiga) and water lettuce (Pistia stratiotes). Waste stabilisation in an aquatic weed pond is accomplished through the reactions listed in Eqs. (1) and (2), except O₂ gas in Eq. (1) is produced by the photosynthetic activity of the aquatic weeds and the produced O₂ is transferred to the roots and stems for the biofilm bacteria to biodegrade the organic matter (Eq. (2)). Reported values of O₂ release rates from the roots of emergent plants are 5–45 g O₂/m²/day, with a mean value of 20 g O₂/m²/day. Because there will be less algae growing in aquatic weed ponds, the aquatic weed pond effluent should contain less suspended solids and low organic matter content, suitable for discharging into water bodies or further reuse.

Wetland is an area which is inundated or saturated by surface or ground water at a frequency or duration sufficient to maintain saturated conditions and growth of related vegetation. Emerged aquatic weeds, such as cattails (Typha), bulrushes (Scirpusy) and reeds (Phragmites) are the major and typical component of the wetland systems. Wetland is a natural system where complex physical, chemical and biological reactions essential for waste-water treatment, exist. Constructed wetlands can range from creation of a marshland to intensive construction involving earth moving, grading, impermeable barriers or erection of tanks or trenches [7].

Two types of constructed wetlands have been developed for wastewater treatment, namely, free water surface (FWS) and sub-surface flow (SF). An FWS system consists of parallel basins or channels with relatively impermeable bottom and soil and rock layers to support the emergent vegetation, and the water depth is maintained at 0.1–0.6 m above the soil surface. An SF system, also called ‘root zone’, ‘rock-bed filter’ or ‘reed beds’, consists of channels or trenches with impermeable bottom and soil and rock layers to support the emergent vegetation, but the water depth is maintained at or below the soil surface. To reduce short-circuiting, the length to width ratios of constructed wetland units should be more than 2/1. Although it might appear that SF constructed wetlands could be subjected to frequent clogging problems, performance data reported so far have shown them to function satisfactorily with high degree of removal efficiencies [9]. A typical layout of a SF constructed wetland is given in Fig. 3.

The reactions responsible for organic matter degradation are similar to those of the aquatic weed ponds, but the wetland bed media will assist in the filtration and adsorption of solids and other pollutant compounds. Hence, constructed wetland can accommodate wide range of organic-rich wastewater and effluent of a well operated constructed wetland is normally of high quality suitable for discharge or reuse. Further, modification of constructed wetland system by changing media content or integrating earthworms into constructed wetland can be applicable for phosphorous removal and reduce biomass production [10–12].

As algal cells contain approximately 50% protein, Eq. (1) suggests that it is possible to culture algal biomass from wastewater. The data in Table 4 show that the potential of algal mass production of 70 tons/ha/yr, equivalent to algal protein production of 35 tons/ha/yr, the productivity which is much greater than those to be obtained from growing wheat, rice or potato. When considering the need to produce more protein for the ever increasing population growth, algal production is an aspect that cannot be ignored.

Fig. 4 shows a schematic diagram of high-rated algal ponds (HRAP) normally constructed to cultivate algal cells from wastewaters. The HRAP depth is 20–40 cm so that sunlight can penetrate the whole liquid depth, thus maximising algal production. In addition, organic matter will be stabilised in the HRAP according to Eq. (2).

The main problem associated with the production of algal protein is the small size of algal cells (approximately 1–20 μm) which causes difficulty in the harvesting of algal cells from the HRAP effluent and concentrating them in, to paste or pellet forms suitable for use as animal or human feeds. The available harvesting technologies, such as micro-straining, belt filtration and centrifugation are still not cost-effective and more research in this area is needed. Most algal species, except Spirulina, have rigid cell walls not easily digestible by non-ruminants and the high nucleic acid content may cause some health effects if ingested in large quantity.

The difficulty of algal harvesting mentioned in the previous section could be overcome using fish to graze on the algal cells or feeding the algal laden wastewater from HRAP into fish ponds. Herbivorous fish, such as Tilapia, grass carp and silver carp which feed mainly on plankton (algal cells) and other aquatic weeds are the desirable species. Table 5 compares the relative growth and feed utilisation of farm animals and fish, showing that fish have a better food conversion ratio (FCR) than farm animals. This is because fish are cold-blooded animals and, unlike other farm animals such as cattle or poultry, do not have to spend a lot of energy for movement.

The cultivation of waste-grown fish can be done by 1) rearing fish directly in waste stabilisation ponds (such as in maturation ponds) or 2) feeding effluent of facultative ponds or HRAP into fish ponds. Reported fish yields from various locations worldwide range from 1,000–10,000 kg/ha/yr depending on the operation modes, fish species and stocking density and climatic conditions.

Because fish can be harvested by nets, a much easier harvesting technique than those required for algal harvesting, the harvested fish can be readily used as animal or human foods. To avoid public health problems, these waste-grown fish may be put in clean water for 1–2 weeks to enable a natural process called depuration to remove some microorganisms from the fish body. Wastewater effluent to be fed to fish ponds must be treated so that its microbiological quality meets the World Health Organisation (WHO) standards for aquaculture reuse stated in Table 6. However, it is essential to investigate the accumulation of potentially toxic elements (PTEs) in the production system, which may contribute a food safety risk as some toxic elements have a great potential to accumulate in both edible and non-edible parts of aquatic plants [15].

The aquatic weeds such as water hyacinth or duckweeds grown in ponds and emergent plants (such as cattails or reeds) grown in constructed wetlands need to be regularly harvested, especially those under tropical conditions, In theory, the frequency of plant harvesting should be based on area doubling time, which is the time taken for the plants to double themselves. Under favourable conditions, the area doubling time of most aquatic weeds ranges between 2–4 weeks, depending on weather.

A water hyacinth pond (WHP) normally has plant density about 20 kg/m² (wet weight). For a pond with a surface area of 1 ha, if 50% of the water hyacinth plants are harvested once in 2 weeks, the amount of water hyacinth biomass available is 200 tons/mon or the biomass productivity of water hyacinth is 200 tons/ha/mon. Because aquatic weeds contain about 95% of water content, the dry biomass productivity of water hyacinth is approximately 120 tons/ha/yr (or annual protein production of 20 tons/ha/yr).

Because of the high moisture content, the harvested aquatic weeds should be dewatered prior to being reused. Table 7 shows the composition of some aquatic weeds. They can be used as feed for herbivorous fish, livestock fodder or raw materials to make compost fertilisers.

Although the biochemical reactions undergoing in the natural systems as outlined in the previous section are mainly governed by two types of suspended and biofilm bacteria, the design equations formulated in the past emphasized the role of the suspended bacteria only. An early study by Reynolds et al. [17] recognized the potential of biofilm bacteria growing on the surface of baffles submersed in the pond water in organic stabilization. Some previous researches on self-purification in shallow streams and rivers have shown biofilm bacteria growing in open-channel beds to be effective in biodegrading organic carbon compounds [18–22]. Recent studies on attach-growth WSPs found the biofilm bacteria growing on attached-growth media installed in the pond water to be effective in waste stabilisation [23–25]. From their laboratory experiments, Kilani and Ogunrombi [26] reported that WSP installed with baffles yielded better treatment performance than those without baffles; although they attributed the improved efficiencies to a reduction in dispersion number in the baffled ponds, the biofilm biomass, presumably growing on the baffle surfaces, could, together with the suspended biomass, contribute partly to the organic matter degradation. Because WSP normally have low flow velocities, but are subjected to higher organic loading rates than streams and rivers, biofilm biomass can be expected to grow at the bottom and side walls of the ponds. Hence, it can be concluded that the role of biofilm bacteria is important in the organic matter decomposition process and should be duly considered in the design equations.

There are numerous literatures indicating improved biochemical oxygen demand (BOD) or chemical oxygen demand (COD) removal efficiencies in WSPs implanted with aquatic plants. Orth and Sapkota [27] found the COD and SS removal efficiencies to increase more than 50% in a facultative pond implanted with water hyacinth when compared with a control facultative pond without water hyacinth plants. Reports from Brazil [28] and the United States [29] found BOD removal efficiencies in ponds stocked with water hyacinth to be about 20% better than those without the plants. It is therefore reasonable to hypothesize from these results that the improved organic removal efficiency in these WHP units was due to the activity of biofilm bacteria attached on the roots of the plants.

The media of the constructed wetlands provides comparatively higher number of sites for the biofilm bacteria than what is there in WSP or WHP. Further, the activity of suspended bacteria is obviously lower in constructed wetlands as the water depth is either less or absent. In wetlands, the path of the water is through the pores or microchannels of the soil/gravel-root matrix. Hence, the activity of biofilm bacteria plays a major role in the biochemical transformations occurring in the wetlands.

Realizing these facts, an integrated kinetic model that could incorporate the activities of both the suspended and biofilm bacteria, hydraulic regime in the system, and the effects of environmental factors in these natural treatment units was developed.

The traditional approach towards design and performance evaluation of the WSP, WHP, constructed wetlands has been to assume the occurrence of either complete mixing or plug flow conditions. The formulas for the completely-mixed and plug-flow conditions, assuming first-order kinetics, steady state conditions, and no effluent losses due to evaporation and seepage, are shown in Eqs. (3) and (4), respectively.

in which k is the first-order reaction rate which varies as temperature, C_e, and C₀, are the effluent and influent substrate concentration, and t is the hydraulic retention time.

The natural treatment systems are complex encompassing the existence of several reactions and hydraulic phenomena. The realization that none of the two extreme flow patterns really exists in practice, led to the development of models that could account for partial fluid mixing in these systems. The model defining the partial mixing of the fluid is termed as dispersed flow model and characterized by a parameter called the dispersion number. The dispersion number of flow, d, can be expressed as

in which D is the longitudinal or axial dispersion coefficient characterizing the degree of back mixing during flow, u is the flow velocity, and L is the length of fluid travel path from the inlet to outlet section. The dispersion coefficient is analogous to the molecular diffusion coefficient in Fick’s law and describes the concentration change due to back-mixing [30].

Wehner and Wilhelm [31] derived the following equation for chemical reactors which exhibit first-order kinetics and non-ideal mixing or partially mixed flow conditions under any kind of entrance or exit conditions:

in which

A conceptual illustration of organic matter (substrate) biodegradation in a treatment unit due to suspended and biofilm bacteria is shown in Fig. 5, in which x is distance from inlet along the pond length, y is the substrate travel distance in the biofilm, L_s is liquid sublayer thickness, L_f is thickness of biofilm, C_w is substrate concentration in bulk water, C_s is substrate concentration at the liquid sublayer and biofilm interface, and C_f is the substrate concentration in biofilm.

The substrate is utilized by the suspended bacteria in the bulk liquid flow and by the biofilm bacteria, with substrate transport through the liquid sublayer acting as a link between the two. With the assumptions mentioned elsewhere [32], a substrate mass balance in a control volume (bulk liquid and biofilm) under steady state condition and without lateral flow can be written as:

in which, z is fractional distance along the pond length is equal to x/L, x is distance from the inlet along the pond length (m), J is substrate flux into the biofilm (g/m²/day), r is substrate utilisation rate by suspended microbes (g/m³/day), as is specific surface area of the biofilm per unit volume of the treatment unit (m²/m³), and other terms are defined previously.

Considering first-order kinetics for the suspended and biofilm bacteria and assuming a linear variation of the substrate concentration across the liquid sublayer, Eq. (8) was modified to describe the substrate concentration as a function of fractional distance downstream of the ponds lengths as:

in which,

φ is characteristic biofilm parameter, D_w is diffusion coefficient in liquid sublayer (m²/day), D_f is diffusion coefficient in biofilm (m²/day), k_fs is first-order rate constant of suspended bacteria (1/day), and k_fa is first-order rate constant of biofilm (1/day).

Applying the boundary conditions,

C_w=C₀ at z=0 and dC_w/dz=0 at z=1.

Eq. (9) was integrated to obtain

in which, C_e and C₀ are the effluent and influent substrate concentrations respectively, and

where,

If kfb represents the term αβ/(α+β), Eq. (12) becomes

The relative importance of the biofilm bacteria compared to suspended bacteria in the organic removal for a natural system can be estimated by comparing the biofilm reaction rate to the sum of the both reaction rates [21]. Thus the percentage of total removal rate due to biofilm bacteria (p), determined from Eq. (14), is:

or

Eq. 10 is the model equation for the substrate degradation in natural treatment systems treating wastewaters. The substrate removal efficiency by suspended biomass only can be determined from Eq. (10) by making as in Eq. (12) equal to zero. Similarly, if the substrate removal efficiency by biofilm biomass only can be obtained from the same equation by making kfs equal to zero in Eq. (12). To determine C, from Eq. (10), several model parameters need to be determined either from the experiments or from the literature.

Polprasert and Agrawalla [32] applied the integrated model (Eq. (10)) to two full-scale facultative ponds located in central Thailand by comparing the computed effluent BOD₅ concentrations with the observed BOD₅ concentrations. The model was able to predict the effluent BOD₅ concentrations of these two ponds reasonably well, indicating the significance of these biofilm bacteria in organic matter degradation in facultative ponds (Fig. 6).

A design chart based on the proposed model for aiding in pond design or estimating additional attached growth area (baffles or other types of attached-growth media) to be installed in the pond water to improve efficiency of the pond performance was also developed (Fig. 7).

Polprasert and Khatiwada [33] applied the integrated model (Eq. (10)) to WHPs treating a low-strength anaerobic pond effluent with an account of the biofilm surface area contributed by the water hyacinth roots. The model was validated by comparing the predicted effluent BODs concentrations values with the observed effluent BOD5 concentrations of a facultative pond implanted with water hyacinth plants located in central Thailand. The predicted effluent BOD₅ values were found in close agreement with the observed effluent values (Fig. 8). Based on the model, Polprasert and Khatiwada [34] found the contribution of the biofilm bacteria attached to the roots of the water hyacinth plants in the WHPs directly related to the improvement on the effluent quality when compared to a similarly operated WSP as is demonstrated in Fig. 9. A design chart in Fig. 7 based on Eq. (10) can be used for WHPs, whereas other design considerations are summarized in Table 8 [4].

Apart from water hyacinth, duckweed is also an efficient aquatic plant used for treating domestic wastewater. The integrated kinetic model for organic and nutrient removal in terms of COD, BOD₅, NH₃-N, and TN can be applied in the duckweed ponds, while hydraulic retention time, organic loading rate, and stocking density are varied during operation period. The observed values from pilot-scale experiment at the Wageningen University, Netherland and the Birzeit University, Palestine and the integrated kinetic model finding developed from experiment at the Asian Institute of Technology, Bangkok illustrate high correlation [35].

In an attempt to apply the integrated kinetic model to predict organic matter removal in FWS constructed wetlands, the value of effective surface areas (where the biofilm bacteria grow) in a wetland bed needs to be determined. Because these effective surface areas include those on the roots and stems of the vegetation, the media soil matrix and litter, direct measurement of the area is impractical. Based on the data of some laboratory-scale FWS constructed wetlands having f 85–110 no./m², 10 cm water depth and 60 cm media depth data and with the aid of the integrated kinetic model, the effective specific surface area, as, was found to be approximately 4.40 m²/m³ (Fig. 10). Fig. 11 shows that the integrated kinetic model could predict organic matter (COD) removal in a FWS constructed wetland satisfactorily. A design chart in Fig. 7 based on Eq. (10) can be used for constructed wetlands, whereas other design considerations are summarized in Table 9 [36].