Wastewater Treatment, Plant Dynamics and Management in Constructed and Natural Wetlands

We determined the characteristics of reed stands at an intermittent lake in Slovenia and degraded and vital reed stands in Hungary. The disturbance in reed performance was measured through growth analysis, amino acid analysis in basal culm internodes, and photochemical efficiency of photosystem II (PSII) in leaves. Morphological parameters indicated higher disturbance in the development of degraded and intermittent reed stands in comparison to vital reed stands. Similarly, total free amino acid contents in basal culm internodes reflected temporary stress response in degraded and intermittent reed stands. On the other hand, potential photochemical efficiency showed undisturbed energy harvesting of all reed stands, even though actual photochemical efficiency revealed temporary disturbance of PSII. The most unfavourable condition for reed development seems to be degraded reed stand of Kis-Balaton wetland and littoral reed stand of intermittent Lake Cerknica.

The Komarnik accumulation lake was built to retain high waters in the Pesnica valley. Nowadays it is used as an unfertilised fishpond. In order to estimate the human impacts we have monitored changes in macrophyte community for 5 years and changes in water chemistry during two vegetation periods. The values of chemical parameters indicated the input of nutrients and different ions entering the system through the run-off from the surrounding areas and by the tributary. At low water level during summer period, oxygen was lacking in the whole water column. The bottom and the water column of the lake were completely colonised by macrophytes comprising 17 species of different growth forms, among which Trapa natans and Ceratophyllum demersum prevailed. The Komarnik accumulation lake revealed to be a resilient system, since inter-annual changes of water level affected only species abundance and not species composition.

The 30-year rate of organic carbon (C) accumulation, based on cesium- 137 (137Cs), was measured in 15 freshwater peatlands across a latitudinal gradient from southern Florida (26°N) to northern Minnesota (47°N) to identify relationships between climate (temperature) and C accumulation. Organic C accumulation was inversely related to mean annual air temperature (MAAT, °C) in acidic peatlands (pH < 5) (C, g m^–2 year^–1 = 199 – 7.94 × MAAT; r ² = 0.64, p ≤0.01), with greatest accumulation in the coldest climate. There was a weak but nonsignificant relationship between C accumulation and MAAT in circumneutral peatlands (pH > 5) (r ² = 0.41, p≤ 0.17). A regression model that incorporated both temperature and precipitation (rain factor, f = mean annual precipitation in cm/MAAT) was no more effective in predicting organic C accumulation than one with temperature alone (r ² = 0.57 for acidic peatlands, r ² = 0.36 for circumneutral peatlands). Across all sites, circumneutral peatlands sequestered less C (49 ± 11 g m^–2 year^–1) than acidic peatlands (88 ± 20 g m^–2 year^–1) regardless of temperature. Our findings suggest that, like terrestrial ecosystems, organic C accumulation in freshwater peatlands is linked to climate through the effects of temperature. Local factors such as pH, hydroperiod and nutrient enrichment also should be considered when assessing the potential of freshwater wetlands to sequester C.

The study was aimed at assessing buffering processes of the Eldoret Chepkoilel wetland in the Kenyan portion of the Lake Victoria Basin. Changing patterns of use and activities in the region of the wetland have exerted enormous pressure on land and water resources, including increased nutrient loads, increased erosion and thus increased sediment loadings to receiving waters. The Chepkoilel wetland was located a few kilometres north of Eldoret town. The wetland was sited in a shallow trough-like valley that lies at an elevation of between 2,110 and 2,140 m. The dominant wetland vegetation included a central band of dense papyrus (Cyperus papyrus) with narrow fringing stands of shorter emergent macrophytes, Cyperus spp. (C. rotundus, C. triandra, C. laevigatus). The section of the wetland described in the investigations was approximately 10 km long. The wetland received inputs from surrounding agricultural land, a flower farm and also a small domestic sewage treatment plant. The wetland received a permanent stream inflow from the Misikuri river. The Chepkoilel campus of Moi University was located to the west of the wetland, and effluent from the campus sewage treatment plant also discharged into the wetland. The monitoring data indicated significant nutrient-processing performance, reflecting a relatively long water-residence time within the system. Problems and approaches to monitoring the system are discussed.

Changes in the chemistry of bulk precipitation and stream water between 1999 and 2006 are described for small drained, wetland and forest catchments located within Šumava Mountains (Bohemian Forest) in south-west of the Czech Republic. The chapter is focused on interpretation of hydrochemical trends in selected time periods, especially on effect of deposition changes and landscape management. In bulk precipitation, the annual mean pH increased from 4.8 to 5.4, concentrations of base cations were relatively low and constant over time, nitrate moderately increased, conductivity and alkalinity decreased, while sulphate decreased (from 16.8 mg l^–1 to 7.2 mg l^–1) during the 7 year monitoring period. Catchments with wetland and forest appear to be more resistant to recovery in terms of water conductivity, alkalinity and base cations in comparison with the relatively fast response from the drained pasture. High nitrate, bicarbonate and base cations concentrations in the stream water discharged from the pasture indicate continuous acidification of the catchment, and they result most probably from fast mineralisation (oxidation) of soil organic matter.

The vast carbon (C) stores in peat soils may be seriously affected by different land-uses, or changes in the prevailing climatic patterns. Land use in peatlands usually includes artificial drainage. Any land-use mediated changes in C emissions from peatlands need to be estimated for greenhouse gas (GHG) reporting. This is not an easy task, since all factors that affect the dynamics of litter inputs and decomposition of organic matter are affected, and the changes may be different in different peatland types, or under different climates. This chapter describes an approach for estimating the C balance of peatland forests on site level that combines measured and modelled information on litterfall and decomposition. Further, we outline the most critical data needs. According to our estimates, the soils of boreal peatland forests may act as either net sinks or net sources of C depending on the ratio of litter inputs to decomposition outputs. The dynamics of below-ground litters, especially, and moss litters are most poorly known.

The ability of near infrared reflectance spectroscopy (NIRS) for the rapid determination of several chemical properties of plant litters was investigated. The chemical properties included fractions that potentially affect decomposition dynamics in peatlands: total carbon, total nitrogen, extractable (soluble) substances, holocellulose (sum of cellulose and hemicelluloses), sulfuric acid (H₂SO₄) insoluble lignin (Klason lignin), copper oxide (CuO) oxidation products of lignin: vanillin (V1), vanillic acid (V2), acetovanillone (V3), 4-hydroxybenzaldehyde (P1), 4-hydroxyacetophenone (P2), 4-hydroxybenzoic acid (P3), and ferulic acid (C2), as well as carbon (C) to nitrogen (N), and N to lignin ratios. To maximize variability, the samples included litters of nine plant species, representing five groups of plant litter: graminoid, deciduous foliage, conifer foliage, wood, and moss. For each quality parameter we compared (1) the model calibrated for all the various litter types (combined dataset) and (2) the model calibrated only for pine branch litter (branch dataset). Relationships were found between the chemical properties and near infrared (NIR) spectra using partial least squares (PLS) regression. Using both the combined and the branch datasets, very good NIR calibrations were possible for total C and N, ash content, nonpolar (dichlormethane) extractives (NPE), ethanol extractives (EE), and total extractives (TE) (sum of all soluble substances), holocellulose, C:N, and N:lignin ratios based on r ² that varied from 0.80 to 0.99. Using only the combined dataset very good calibrations were also possible for acetone extractives (AE), water extractives (WE), lignin, P2, P3, C2, V1, V2, and V3, with r ² from 0.81 to 0.97. The models combining all the different litter types performed better than the models constructed for the pine branch litter only, based on r ² and residual prediction deviation (RPD), the ratio of the Y-variable standard deviation to the root mean square error of prediction (RMSEP). However, RMSEP was generally smaller when a branch litter property was predicted by the branch model compared to the prediction using the combined model, indicating the potential of improving the NIR calibrations by constructing specific models for different groups of litter. NIRS proved to be an accurate and fast method for the determination of plant litter quality that might be highly relevant for decomposition and C dynamics in peatlands.

The purpose of this short-term pilot-scale (1 m³) experiment was to focus on the efficiency of leachate treatment during the initial period of a newly built peat filter. The initial start-up period of a filter is dynamic and differs for different types of peat and leachate. Reduction of biochemical oxygen demand (BOD), chemical oxygen demand (COD), and nutrient concentrations from leachate of different age and composition was targeted in experimental filters of three types of peat. Well-mineralised fen peat with undisturbed structure showed a significantly better reduction when treating methanogenic leachate with respect to the concentrations of COD, total nitrogen (N_tot), and ammonium-nitrogen (NH4-N), which were reduced by up to 36, 62, and 99%, respectively, compared to the poorly mineralised milled Sphagnum peat (with no reduction of COD, 52% reduction of N_tot, and 67% of NH4-N). Good results in the reduction of BOD and total phosphorus (P_tot) (50%) in well-mineralised Sphagnum peat were achieved when treating acidogenic leachate. Generally, even if a considerable amount of pollutants was removed in newly built peat filters (milligrams per metre per day) during the first days in operation, the target values were still above the Estonian limit for wastewater discharge into the environment. The best results were achieved for removal of P_tot, which was reduced below the target values.

The UK National Biodiversity Action Plan highlighted reed beds as a habitat of conservation concern and provided targets for the management, restoration and creation of reed beds throughout the UK, ideally in blocks greater than 20 ha in size. Previous water-use rate research by Aston University targeted small and fringe reed bed habitats. Extensive literature reviews carried out in 2000 identified a lack of information with respect to the water-use rates of larger reed beds. A research programme was developed utilising the lysimeter technique for determining evapotranspiration (ET) rates in large reed beds at three research sites. At each site, meteorological monitoring equipment and lysimeters were installed within the reed beds. Monitoring was carried out monthly throughout 2001 and 2002. Phenological data was collected from fixed quadrats within the reed beds and from the lysimeters during each site visit. This data was utilised to determine which of the lysimeters were ‘successful’, i.e. representative of the reed bed into which they where installed. From the ‘successful’ lysimeters, monthly crop coefficients (Kc) were derived using reference crop evapotranspiration (ETo) data provided by the UK Meteorological Office (MORECS Grass data). The results are compared with relevant published data and show that the findings are comparable with other studies.

It is argued that there is a need for a hydrological assessment to be undertaken at the design stage of wetlands, which are to be constructed for wastewater treatment, in order to ensure their long-term sustainability. A simple water budget provides a suitable tool for this purpose. The product of an appropriate crop coefficient (Kc), and the reference crop evapotranspiration (ETo) determined from meteorological data, is deemed a suitable procedure for determining the evapotranspiration component of the water budget. The UK and the Czech Republic are taken as case studies, representative of maritime and mid-continent climates, to illustrate a procedure for identifying regions where created wetlands are potentially at most risk from water deficits.

We studied factors affecting the accumulation, mobility and availability of metals in intertidal wetlands of the Scheldt estuary (Belgium), both in greenhouse experiments and under field conditions. The surface layer of the intertidal Scheldt sediments was found to be significantly contaminated with trace metals. The metal concentrations can be predicted from clay or organic matter concentrations. They were somewhat higher than predicted at sites within a range of a few kilometres from specific point-sources of metals, whereas they were lower than predicted at sites which are regularly subjected to flooding by high salinity water. In the deeper sediment layers, sulphides seem to play an important role in the metal accumulation. Salts significantly increased the metal mobility in the oxidised sediment layers in the brackish part of the estuary, which was especially observed for cadmium (Cd). The decomposition of stems and leaves of reed plants and willows in the upper sediment layer was found to both increase and decrease the metal mobility. The concentrations in the reed litter itself increased significantly during decomposition under field conditions. The hydrological regime to which metal-polluted sediments were subjected, affected the metal concentrations in the pore water to an important extent. Flooding increased the mobility of iron (Fe), manganese (Mn), nickel (Ni) and chromium (Cr) and decreased the mobility of Cd, copper (Cu) and zinc (Zn) in a calcareous substrate. Fluctuating hydrological conditions resulted in fluctuating metal concentrations in the pore water. Reduction and oxidation of Fe and Mn, decomposition of carbonates and the formation and re-oxidation of sulphides were hereby the main controlling processes.

This chapter will review the various theories of community development, particularly the concept of community participation. It will review various terms and will explain and evaluate the practical approaches underlying the terminology. The history of the Community Participatory Approach (CPA) will be presented. The chapter will also present the basic principles underlying this approach. It will also present the tools and techniques of CPA and their concern with reinforcing community structures, promoting community involvement, and building community capabilities and developing sustainability. The chapter will then proceed to review the technology of reed bed treatment systems (RBTS). Hybrid systems will be examined and the suitability of hybrid RBTS to community wastewater treatment will be presented. The principle considerations of reed bed design, construction and maintenance will be examined with mention of the newer developments and design modifications. The implementation of a RBTS in a small community will be examined from the point of view of CPA. The principles of RBTS design will be combined with CPA tools and techniques to show the usefulness of CPA to RBTS design, construction and maintenance.

A subsurface vertical up-flow followed by down-flow constructed wetland system was used to treat anaerobic digester effluent from pig-farm wastewater. The system consists of two beds connected in series and have equal size of 1.2 × 1.2 × 0.6 m (W × L × D). The first bed was planted with Typha angustifolia and the second bed was planted with Cyperus alternifolius L. The wastewater with NH₃-N and chemical oxygen demand (COD) concentration of approximately 300 mg l^–1 was fed to the first bed which performed as a vertical up-flow bed and then passed to the second vertical down-flow bed and recirculated to the first bed with the recycle ratio of 1:1 to enhance nitrogen removal. The system was operated with the hydraulic loading rate (HLR) of 10, 20, 30 and 40 cm day^–1 including recycled water to evaluate the capability of the system under high HLR. The results showed that the removal efficiencies of N and COD decreased when the system operated with higher HLR. The removal efficiency for NH₃-N and COD was maximum at 10 cm day^–1 (79 and 74% respectively). Nitrification was significant in vertical down-flow bed, denitrification in the up-flow bed at every HLR. The two-stage constructed wetland removed 96–98% of incoming total suspended solids (TSS) and the removal was not affected by high HLR. The removal efficiency of total phosphorus (TP) was low (24–27%) at all HLR.

This study compares the treatment performance of aerated and nonaerated horizontal flow constructed wetlands. Results are presented in a series of performance charts that relate effluent concentrations to influent mass loadings. A statistical approach has been developed that allows the application of confidence intervals to system performance. Performance data is presented for predicted median (50th-percentile) and 90th-percentile effluent concentrations. Aeration was found to improve removal of biochemical oxygen demand (BOD), total suspended solids (TSS), and to a lesser extent, total Kjeldahl nitrogen (TKN). Benefits of aeration include lower median effluent concentrations at a given influent mass loading, less variability in effluent quality, and lower background concentrations (C*). In contrast, aeration was observed to have no impact on total phosphorus (TP) or fecal coliform bacteria removal.

Constructed wetland Břehov for 100 PE was put in operation in October 2003. It is a typical horizontal subsurface flow system designed to treat municipal wastewater in the Czech Republic. Vegetated beds with a total surface area of 504 m² are filled with washed gravel (4–8 mm) and planted with bands of Phalaris arundinacea and Phragmites australis. During the period 2004–2006, the treatment performance and vegetation growth were regularly monitored. The average treatment effects for BOD₅, COD and suspended solids were high and amounted to 88%, 77% and 88%, respectively. The average outflow BOD₅, COD and TSS concentrations during the monitored period were 16.5 mg l^–1, 78.3 mg l^–1 and 19.2 mg l^–1, respectively. Removal of TN and TP amounted to 45% and 40%, with respective average outflow concentrations of 28 mg l^–1 and 4.3 mg l^–1. These results are quite commonly achieved in horizontal subsurface constructed wetlands in the Czech Republic. The results also indicated that during the warm period (May–October) treatment efficiency for nitrogen and phosphorus was higher than that during cold period of the year (November–April). Removal of organics and suspended solids was not affected by the season. Phalaris reached its aboveground biomass during the second growing season (2,265 g m^–2), while Phragmites increased its biomass over the 3-year period (2,532 g m^–2). The nutrient aboveground standing stocks were comparable with those observed in natural stands and did not exceed 5.3% and 6.2% of the inflow load of phosphorus and nitrogen.

Compared with other forms of wastewater treatment, horizontal flow reed beds require very little by way of operational and maintenance input. Unfortunately, in practice this frequently results in them receiving little or no attention at all. Tertiary treatment reed beds are not a “fit and forget” solution, but they are often treated this way because they are very forgiving and abuse-tolerant. Severn Trent Water Ltd. installed their first tertiary reed beds in the late 1980s and the company now has almost 350 units, mainly at village works, to ensure that the effluent from existing secondary treatment units is polished to a produce a high-quality effluent well within the standard. ARM Ltd. has been designing and constructing reed beds for 20 years and has units on more than 250 sites, a large number of these are on Severn Trent Water sites. After a number of years a few of these tertiary reed beds have deteriorated to the extent that they are close to failing to comply with the regulator’s requirements. Severn Trent Water Ltd. recognised this situation and has committed a budget for a programme of reed bed maintenance. Initially a budget was allocated to refurbish the reed beds in the poorest condition. Resources are now being directed to regular maintenance of all reed bed sites. Between November 2002 and October 2005, ARM personnel carried out more than 300 primary surveys of reed bed sites in order to determine the condition of the reed beds and to establish what if any remediation was necessary. Since September 2004, when the first part of this chapter was presented at the IWA Avignon conference, more than 200 secondary surveys have been carried out when staff visited sites to carry out remedial work. The work is being further extended by use of ground penetrating radar (GPR) to try to assess the degree of clogging in different beds and in different parts of the beds. The preliminary results from this study are included. A series of tertiary surveys will be made in the future when the sites are revisited as part of a rolling maintenance contract.

A two-stage constructed wetland (CW) system consisting of two vertical flow beds with intermittent loading operated in series has been investigated. The first stage uses a grain size of 1–4 mm for the main layer and has a drainage layer that is operated under saturated conditions; the second stage uses a grain size of 0.06–4 mm and a drainage layer with free drainage. The two-stage system was operated with an organic load of 80 g COD m^–2 day^–1 for the first stage (1 m² per person equivalent), i.e. 40 g COD m^–2 day^–1 for the whole system (2 m² per person equivalent). The system has been investigated using indoor pilot-scale CWs. The results of the indoor experiments are currently verified outdoor. Organic matter and ammonia nitrogen (NH₄-N) removal has been very high and has been shown to be the same as for a single-stage CW system (grain size for main layer 0.06–4 mm, free drainage, organic load 20 g COD m^–2 day^–1, i.e. 4 m² per person). A total nitrogen elimination rate of 58% could be reached for the two-stage indoor system compared to almost no nitrogen elimination for a single-stage system. The first results of the outdoor experiments are similar compared to the indoor experiments regarding organic matter, nitrogen, and microbial parameters. The nitrogen elimination reached in the first 3 months of operation was 63%. By using the two-stage CW system it is possible to obtain a higher effluent quality compared to a standard single-stage CW system with only half of the specific surface area requirement. Nitrogen removal efficiencies of about 60% and nitrogen elimination rates of more than 1,400 g N m^–2 year^–1 can be reached without recirculation.

The potential use of Myriophyllum spicatum L. for the removal of Co, Ni, Cu and Zn from industrial effluents was studied. The removal kinetics, tolerance and accumulation capacity of the submerged aquatic plant were assessed. Removal of Cu and Zn was similar and occurred rapidly with time whereas removal of Co and Ni was slower. Plant growth was not adversely affected during the 12 weeks of exposure to the wastewater. Cobalt, Ni, Cu and Zn concentrations of respectively 1,675, 1,529, 766 and 2,883 mg kg^–1 DM were observed in the biomass. M. spicatum is suggested as an efficient plant species for the treatment of metal-contaminated industrial wastewater.

The aim of this study was to quantify nitrogen removal in high loaded free water surface wetlands dominated by emergent vegetation. It was undertaken in two subsystems of the full-scale wetland Alhagen in Nynäshamn, Sweden. Time proportional samples were taken at the inlets and outlets for 2 weeks in September and November 2005, respectively, and the water flow was monitored. The samples were analysed for ammonium-N (NH₄ ^×-N), nitrate-N (NO₃ ^–-N), nitrite-N (NO₂ ^–-N) and total-N, and the mass nitrogen removal was calculated. Sediment cores were randomly collected to measure potential denitrification, and the result was related to the actual mass nitrogen removal. Zero total-N removal could be detected in the subsystem with 6 hours hydraulic retention time (HRT). In the one with 3–4 days HRT, the total-N removal rates were 0.6 g N m^–2 day^–1 in September and 0.2 g N m^–2 day^–1 in November. The potential denitrification rate was 8 times higher than the observed removal in September and 48 times higher in November. This deviation was likely related both to relatively high oxygen levels and to the amount of available organic carbon.

The objective of this study was to evaluate the effectiveness and contribution of Schoenoplectus fluviatilis (Torr.) (river bulrush) to phosphorus (P) removal from dairy-farm effluent in a cold climate constructed wetland. After 3 years of operation (1,073 days), both nonaerated wetland cell 3 (C3) and aerated cell 4 (C4) exhibited a sharp decline in dissolved reactive phosphorus (DRP) storage, indicating wetlands saturation. The quantities of DRP stored during the three growing seasons (433 days) represented only 10.0%(C3) and 17.7%(C4) of the total amount of DRP (435.06 ± 2.3 g m^–2, 97.02 kg) added to each cell (C3 and C4) over the entire 3-year period. However, of the total DRP retained by both wetland cells during 1,073 days of operation, the quantities stored during the three growing seasons (433 days) represented 50.3%(C3) and 36.50%(C4) of the total DRP retention. This indicated that vegetation had an important role in the overall DRP storage regardless of supplemental aeration. Overall, nonaerated C3 DRP mass removal efficiency during the 3-year period of investigation was low, averaging 19.9%. Aerated C4 DRP mass removal efficiency was 2.4-fold higher, averaging 48.4%. Belowground (BG) biomass had significantly higher (p < 0.001) P content than aboveground (AG) biomass, throughout the 3-year period of investigation.

Constructed wetlands can provide an effective and economical solution to landfill leachate treatment. In this chapter, performance of two constructed wetlands (CWs) for municipal leachate treatment, in Gdansk-Szadolki and in Gatka near Miastko (northern Poland) is discussed. The plant in Gdansk-Szadolki, consisting of two horizontal flow (HF) beds, has been in operation since 2001. The performance of the plant in the period 2004–2005 is discussed. Clogging problems and surface flow occurred in the system because of too-low hydraulic conductivity of the filter beds material. Neither the treatment effectiveness nor effluent pollutant concentrations meet the design criteria, though quite high efficiencies of polycyclic aromatic hydrocarbon (PAH) (over 70% at both beds), total nitrogen (N_tot) and ammonia nitrogen (N-NH₄) (61.7% for the bed I and 51.6% for the bed II) removal are obtained. The willow plantation in Gatka has been in operation since 1996. The leachate was transpirated to the atmosphere and there was no outflow from the plant. Prior to the investigations of the plant performance (in the period December 2005–November 2006), the polyethylene wells were installed in order to collect pore water samples. Good removal efficiency of N_tot and N-NH₄ (70–78% and 78–88%, respectively) was observed.

The municipal centralized treatment plant (a classic activated sludge technology) receiving the wastewater produced by the municipality of Jesi (Ancona province, Central Italy), was upgraded in 2002. A new “nitrification–denitrification” compartment and a hybrid constructed wetland (CW) were added to the system to enhance denitrification and provide final treatment polishing. The whole system treats about 19,000 m³ day^–1 (more than 60,000 PE) and a part of the effluent is reused in a nearby industrial area. All the new sections have been provided with an online monitoring system, in order to reduce as much as possible the energy consumption for the denitrification process, and relying more on the final wetland stage whenever it obtains sufficient treatment levels. The upgraded wastewater treatment plant has been operating since January 2003, and after the first year of operation, good development of the macrophyte communities has been observed. The hybrid CW system consists of an initial sedimentation pond (volume of 2,000 m³), a 1 ha horizontal subsurface flow system (HF) second stage, and 5 ha free water system (FWS) as a final component. It has taken almost 18 months since the CW start-up for denitrification to occur at considerable levels. Nitrates removal rates ranged from 25% to 98% during the hottest period of the year.

In order to understand the growth dynamics of the floating macrophyte Pistia stratiotes under natural conditions in a temperate climate, and the influence of higher levels of N and P concentrations in water on the specific growth rate, m, a number of experiments were carried out in two series. Series I: 29 determinations of m in the stationary growth stage using an oligotrophic medium were carried out. The experimental data was fitted by a function which rules cyclic processes in wetlands (R = 0.93), and a maximum m of 0.0395 day^–1 over the year was calculated. Series II: several batch experiments were carried out in Osaka 34° northern latitude, and in Montevideo 34°50' southern latitude. The transitory stage of growth dynamics for plants, which were transferred to a medium with higher concentrations of N and P, was considered. In this case, the Double Monod model with double inhibition affected by a constant term adequately explains the variations in m with R = 0.85 and 0.88 at 24°C and 29°C respectively.

In order to analyse the fate of organic matter and particles along the length of horizontal subsurface-flow constructed wetlands, two pilot wetlands were monitored in terms of BOD₅, COD and their fractions, TOC and particle counts in various sampling campaigns. Anaerobic biodegradability properties of the effluent organic matter were also measured by means of a methane-production test. The most abundant particles in the influent, at intermediate points along the length of the wetlands, and in the effluent were those comprised in the range of 0.7 to 2m. Most of the particles (close to 80%) were removed within the first quarter of the wetland length. Most organic matter was also removed within the first quarter (50% and 80% in terms of TOC and COD, respectively). Further organic matter removal along the rest of the length of the wetland only accounted for a 0–10% of the total removal. The organic matter of the effluents is difficult to biodegrade under anaerobic conditions, but it is readily biodegradable under aerobic conditions.

A wastewater-fed, partly indoor aquaculture plant consisting of 36 basins with a total area of 360 m², charged with effluent from a methanization plant processing organic household waste (Kompogas System), started operation in spring 1998 in Otelfingen, Switzerland. Due to high effluent concentrations on one hand (total organic carbon, total nitrogen, and total phosphorus concentrations being 670, 255, and 52 mg l^–1, respectively) and stringent wastewater regulations on the other hand, the aquaculture was split into modules and stocked with organisms of different environmental requirements. The modular structure, as opposed to traditional one-pond polyculture, allowed manipulation and better nutrient budgeting of the system. Research focused on nutrient-recycling efficiency of a wide array of modules, producing biomass such as algae, fish, zooplankton, and aquatic macrophytes as ornamental plants (Eichhornia, Pistia, Ipomoea, Lemna, Azolla) which were suitable for sale on the Swiss market. Different arrangements of modules, of water and wastewater flows and of nutrient concentrations were tested in 3 years, 1998–2000. Our research showed that aquaculture can be set up to either produce treated wastewater ready for discharge into surface waters or to maximize nutrient recycling through biomass production. In the first case, the recommended loadings are below 0.47 g N m^–2 day^–1 and 0.084 g P m^–2 day^–1, in the latter 4.3 g N m^–2 day^–1 and 0.31 g P m^–2 day^–1 still resulted in a 78% removal of nitrogen and 28% removal of phosphorus. Through weekly biomass harvest, up to 0.36 g N m^–2 day^–1 and 0.06 g P m^–2 day^– were converted into plants. With appropriate planning and plant selection and management, it was possible to increase the fraction of nutrients eliminated via primary production up to 40%, which is significantly higher than reported in the literature.

Filtration beds of subsurface horizontal-flow constructed wetlands (HF CWs) are generally considered as anoxic or anaerobic and, therefore it is assumed that the outflow concentration of dissolved oxygen (DO) is usually very low. During 2004 and 2005, nearly 60 HF CWs in the Czech Republic were visited and it was found that many systems really provided low DO outflow concentrations (<2 mg l^–1). However, a substantial number of systems surprisingly provided relatively high concentration of DO (>5 mg l^–1). We selected several systems where monitoring program included a broader selection of parameters, especially ammonia-N, nitrate-N, and sulfate, and set up more frequent measurements of dissolved oxygen. In our study, we also used some older data that were obtained during research projects in the past. We focused on nitrification and sulfate-reduction as processes occurring under strictly aerobic and anaerobic conditions, respectively. When evaluating the data, we found that in systems with very low outflow concentrations of DO, nitrification was frequently very limited but in some systems a substantial reduction of ammonia occurred. On the other hand, several systems with relatively high O₂ outflow concentrations provided nearly zero removal of ammonia. The similar unanimous results were observed for sulfate, i.e., high O₂ outflow concentrations were sometimes connected with high reduction of sulfate and on the other hand low O₂ outflow concentrations were not connected with sulfate reduction. We concluded that DO concentration at the outflow from HF CWs does not provide good information about the processes occurring in the filtration beds.

Two wetland systems in the Lake Victoria Basin, western Kenya, were investigated and monitored to assess buffering capacity and also to develop a model approach to evaluating larger-scale pollutant-buffering capacity of regional wetlands. The Dionosoyiet wetland was located in the highlands of the Rift Valley Province, at approximately 2,000 m altitude. It was located immediately adjacent the Kericho town centre, covered a 34 ha area and was set in a catchment of 23 km². The wetland was located in the upper reaches of the Sondu-Miriu river system which flows into Lake Victoria. The 560 ha Chepkoilel wetland, near Eldoret, had an agricultural catchment area of 210 km² with major inflows to the wetland contributed by the Sergoit-Misikuri river system. The catchment drained areas of mild slopes ranging up to 2,160 m above sea level. Water-quality investigations were undertaken in the wetlands from June 2004 to April 2005 for nutrients and suspended solids. Hydrology and water-quality modelling were performed utilising the LAVINKSWEB model. The model was adapted to incorporate a rainfall-runoff module based on the isochronal histogram technique and a partially stochastic prediction of water quality (TSS, TN and TP) based on incoming flow rates. Using the data gathered from June 2004 to April 2005 for calibration and earlier climatic data, modelling was performed to cover an 11-year period, from January 1994 to December 2004, and indicated that the wetland removed 43% TSS, 41% TP and 20% TN with average areal removal rates of 21.3 TSS, 0.038 TP and 1.03 TN (kg ha^–1 day^–1) for the Dionosoyiet system and considerably greater removal from the Chepkoilel system. The findings and model development show that in addition to being critical ecosystem diversity reservoirs and central community/agricultural-activity resources, the wetlands perform significant functions of water-quality improvement. The preservation of these wetlands and other similar wetlands is important in ensuring sustainable utilisation of water resources in the Lake Victoria Basin.

Constructed wetlands (CWs) have been successfully employed in both mining and domestic wastewater applications, yet the fundamental processes responsible for treatment are poorly quantified. Sulfur is common in CW influent streams and is highly reactive, redox-sensitive, and microbially active; therefore, it plays an important role in both desirable and deleterious processes in CWs. In this chapter we review the major sulfur transformations likely occurring in CWs, their interactions with other important processes, and their role in the treatment process. We also present two case studies on the influence of sulfate-reducing bacteria and sulfur-oxidizing bacteria on the performance of CW systems designed to treat mining-contaminated and municipal wastewater, respectively. In both cases there is a feedback between these microbial consortia and other microbes responsible for treatment. A better understanding of the important sulfur transformations in CWs will lead to better design and more confident performance expectations.