Reedbed technology for wastewater treatment:obtaining a better insight through modelling

Constructed wetlands: a green technology for integrated water management

The quality of our Flemish surface waters has improved significant during the last decade. Nevertheless, in a densely populated area like Flanders we are still confronted with polluted surface water. The main reason for the current degree of pollution of our rivers and canals are the numerous (untreated) discharges caused by the households, the industry and the agricultural sector. In completing the construction of a municipal wastewater treatment infrastructure in Flanders, the focus shifts from large scale to small-scale projects. As the degree of the connection to large-scale wastewater treatment plants (WWTPs) in urban areas increases, the necessity to treat the remaining pollution with small-scale wastewater treatment plants (SWWTPs) and individual treatment systems becomes more important. A quick and effective approach of the rural pollution is desired, and wastewater treatment with reedbeds as a way of integrated water management plays an important role in doing this. The different aspects of water management – like water supply, water pollution, water treatment, water drainage,… – are all strongly interrelated, which makes a well considered and integrated approach necessary. At the moment, Aquafin nv has 15 reedbed installations in operation, 13 are additionally planned. On average, these installations treat the municipal wastewater of 500 inhabitants.

The application of “green” water treatment technology is the typical example of the possibility for integrated water management. Small-scaleness, landscape fit-in, ecological added values and sustainability are just a few of the aspects that meet each other in the application of “green” wastewater treatment technology and that enhance the integrated character. Numerous studies during the last two decades have proven that plant systems are often the most cost-effective systems for the treatment of small-scale pollution. Reedbeds, as a consequence, make up an important group within the small-scale water treatment technologies who can tackle the problem of the diffuse water pollution at the source.

Principle of reedbeds used for wastewater treatment

The principle to use reedbeds as a way to treat wastewater is fairly simple at first sight. The raw wastewater is pumped to the reedbed (= influent) and flows over or through the soil of the wetland, after which it is conveyed to drainage pipes (= effluent). Depending on the way in which water can flow through the reedbed, one can distinguish free-water surface flow reedbeds (horizontal flow over an impermeable soil surface), subsurface flow reedbeds (horizontal flow through a permeable soil-root matrix) and vertical flow reedbeds (vertical downward infiltration through the reedbed soil).

Throughout the wetland passage, numerous pollutants will be captured, transformed or removed by different physical (e.g. filtration), chemical (e.g. precipitation) and (micro)biological (e.g. denitrification) processes on or inside the reedbed soil. The exact way in which pollutants are removed, the contribution of the different transformation and removal processes and the degree of their mutual and environmental interactions (e.g. reed plants, temperature,…) make it very complex to understand the reedbed technology in detail.

Reasons for scepticism ?

Although reedbeds are world-wide frequently used, some water managers and governments still remain sceptical toward this treatment technology. The reason for this can be found in the fact that reported treatment results of reedbeds are inconsistent and that insufficient clear information exists about this type of wastewater treatment. Both high and low treatment efficiencies (= removal percentage of a certain pollutant) are published. Operational experiences range from very positive to very negative.

A scientific based answer for these observed variations in construction, exploitation experience and efficiencies is hard to give. The research, with or without the aid of computer models, restricts itself to a so-called black box approach of the reedbed ecosystem. This means the treatment efficiency is being calculated based on concentration differences in the influent and effluent, without further attention to the underlying processes. As a consequence, this unclearness in operation and modelling causes insufficient insight in the role of the different treatment processes in a reedbed and its related possibilities to control the operation.

Seeking for answers: research on a pilot scale constructed wetland

Due to the importance of the above-mentioned problems the Department Applied Ecology and Environmental Biology decided to dedicate a thesis to this subject. The main objective of the thesis was to enhance the insight in the operation of the different removal processes and mechanisms that occur inside both subsurface and vertical flow reedbeds. The research consisted of 3 phases.

First of all a brief review was given of already acquired insights as well as existing uncertainties within the research for reedbeds by means of a literature study. A lot of literature exists about reedbed technology, but a lot of data is very fragmented and as a consequence it’s not easy to obtain straightforward a clear view on the different parts of these systems.
A second phase was an intensive monitoring campaign of 9 days executed in January 2001 on the pilot scale constructed wetland of Aquafin nv in Aartselaar (Belgium). This wetland comprises of a vertical flow followed by a horizontal flow reedbed. After monitoring of the pollutant concentrations in influent and effluent and by interpretation of the acquired data set, a first insight was acquired about the functioning of the researched wetland. By comparing with other research results from literature the understanding in some specific operational aspects could be increased.

Finally computer simulations were executed with 2 models based on the measured data set for the subsurface reedbed and the obtained insights from the literature study. The additional goal was to optimise these models in order to use them in the future for precise predictions. Altering certain parameters in the model (e.g. surface area and depth of the reedbed, temperature, soil porosity,…) makes it possible to foresee the influence on the effluent quality. From this, conclusions can be made regarding the dimensioning and operation regimes of new, similar reedbeds that can treat the wastewater even better.

Monitoring and modelling results

Measurements show good to very good results for the removal of most pollutants in the pilot scale constructed wetland, especially regarding the fact that the measuring campaign took place in the (cold) winter period. Total mean removal efficiencies for 8 of the 12 variables amount 80% or more. The limiting factor in the nitrogen removal was caused by the very bad functioning denitrification process in the subsurface reedbed. Especially temperature and reed growth seemed to have an influence on the efficiency of the nitrogen removal.

Execution of simulations revealed that the 2 applied computer models are powerful instruments to (i) predict accurately effluent concentrations of different pollutants in subsurface flow reedbeds, and to (ii) increase the insight in the relations between the different treatment mechanisms and problems. The ultimate objective of these models has to be the successful simulation and comparison of different operation scenarios in order to optimise the treatment capacity of the subsurface flow reedbed.

And the future…?

One of the main constraints that affect – sooner or later – every reedbed is the occurrence of clogging of the reedbed soil. According to different authors and designers clogging is one of the biggest disadvantages for an efficient operation of most reedbeds. Consequently further studies should concentrate on clogging phenomena in reedbeds. When it is possible to predict the evolution of the porosity of the reedbed over time with models, they can be powerful instruments to optimise the operation of the reedbeds in practice. In addition, further work should be directed towards the precise validation and optimisation of the developed modelling instruments (made e.g. in this research).

The existence of a universal model that can predict always and everywhere correct results for different types of reedbeds is still a far-off future. But when the studied models are optimally used, they will offer powerful tools to achieve a better dimensioning and optimal operational management of this “green” water treatment technology.