ARTICLE

Making constructed wetlands work better

By Clàudia Turon Planella
Granollers, Spain
















Short CV of Clàudia Turon Planella

This article summarizes the Ph.D. thesis of Clàudia Turon Planella. The full thesis can be downloaded from http://www.tdx.cat/ (enter Turon as authors name into the search mask)

PhD Directors:
Dr. Manel Poch Espallargas (University of Girona)
Dr. Joaquim Comas Matas (University of Girona)
Dr. Ulises Cortés (Technical University of Catalonia)

When trying to protect the receiving waters from wastewater impacts, the integrated management of urban wastewater systems (sewers and treatment plants) is crucial. In this line, several regulations have been developed, i.e. the Water Framework Directive in Europe and the Clean Water Act in the United States of America, specifying the type of technology and the treatment level.

Irrespective of technology and of the required treatment level, all types of Wastewater Treatment Plants (WWTPs) can be considered complex systems because they are ill-structured domains which are based on non-static and ill-defined purification processes. These facts (domain and processes) impair the reliability of a good treatment efficiency. Maintaining the quality of the effluent is a very important issue not only because of more and more restrictive environmental and social regulations but also for economical reasons.

Therefore, the special characteristics of WWTPs require more than a straightforward application of conventional numerical models to obtain optimal management when (1) the process state of the WWTP is far from its normal operational state and (2) reasoning with qualitative information is essential to dealing with problems. In this respect, Environmental Decision Support Systems (EDSSs) have generated high expectations as tools to tackle the complexity associated with WWTP management.

An EDSS can be defined as interactive, flexible and adaptable software, which links numerical models and algorithms with knowledge-based techniques, geographical information and environmental ontologies. EDSSs are developed to support environmental decision-makers in choosing between alternatives. They allow

1. to manage huge volumes of data;
2. to handle expert knowledge;
3. to tackle the uncertainty of data and knowledge;
4. to integrate both data and knowledge, through different models, into a software;
5. to accurately evaluate multiple alternatives;
6. to diagnose an abnormal situation and propose a procedure to deal with the problematic event and
7. to provide objective offline and online proposals.

My thesis [1] was developed within a project financed by the Catalonian Water Agency (Agència Catalana de l'Aigua), which aims to build an EDSS to support the definition of operation and maintenance protocols for small WWTPs. These facilities are more complex to manage than large WWTPs, because the scarce availability of economic, human and technical resources limits the correct functioning of these facilities. [For non-Europeans: Catalonia is an autonomous community of Spain, with the capital Barcelona. The editor.]

A common small-WWTP-technology is the Horizontal Subsurface Constructed Wetland (HSCW). Hence, the main objective of my thesis was to develop an EDSS to support the definition of operation and maintenance protocols for HSCWs (henceforth referred to as "EDSS-maintenance prototype"). Moreover, we also aimed to provide some guidelines to develop an EDSS which comprise all common small WWTP technologies (henceforth referred to as "EDSS-maintenance").

The first step to achieve our purpose was the selection of a methodology to develop the "EDSS-maintenance prototype". We chose the five-step methodology proposed by Poch et al. [2] because (1) it was successfully applied in the construction of seven EDSSs in the water management framework and (2) it provides a certain flexibility to acquire and integrate data and knowledge required to solve water environmental problems, as well as represent this information through different models.

The development of the "EDSS-maintenance prototype" started with the analysis of the environmental problem. A comprehensive literature research allowed characterizing the complexity associated to small WWTPs and the HSCW technology. Moreover, this analysis allowed to define the requirements for developing the "EDSS-maintenance prototype".

The information was acquired from different sources:

1. literature research
2. a questionnaire distributed to 13 small WWTPs based on HSCW technology and
3. onsite visits.

All these data and knowledge were analysed using categorization and was then organized in terms of: problem, modes, effects, causes, measures and actions. As a result of this work we identified and characterized 18 problems.

The second step in the development of the "EDSS-maintenance prototype" was the selection of a methodology to represent the acquired data and knowledge. After an analysis of the gathered information and the available representation methods we chose the Rule-Based System (RBS) methodology.  In this step Java and Drools were also selected as the adequate software tools to build the "EDSS-maintenance prototype".

The following third phase was the representation of the data and knowledge acquired according to the RBS methodology. First of all, the information about of HSCWs was organized in 37 decision trees and one matrix. Following this, it was simply converted into IF-THEN rules by traversing each branch from the root to the leaf or crossing from the columns to the rows.

An important step in the development of "EDSS-maintenance prototype" was the evaluation process which aimed to verify whether we have built the system "right" and validate whether we have built the "right" system. To perform this evaluation we proposed a methodology based on four steps:

1. verification
2. laboratory validation with historical data
3. laboratory validation with experts and
4. field validation, which includes two mathematical functions to quantify how accurate is the EDSS.

Along with the development of the "EDSS-maintenance prototype" several recommendations to develop the full "EDSS-maintenance" were made. These suggestions were based on the five-step methodology of Poch et al. [2] and the experience gained during the development of the "EDSS-maintenance prototype".

The first objective of these recommendations was to improve data and knowledge acquisition (quantity and quality). Afterwards these recommendations were focused on the analysis and representation phase of this knowledge, and finally they conclude with some advice for the "EDSS-maintenance" evaluation step.

[This section was added by the editor, Andreas Schoenborn]

The "EDSS-maintenance prototype" was applied to the planned HSCW at Hostalnou de Bianya, a community of 75 inhabitants without any industry, located in the North-East of Catalonia, which wil be served by a HSCW of 600m² built in 2007. It will add one to the 13 HSCW currently existing in Catalonia, with about 400 to be expected to be built in the near future.

In concrete:

1. Data on 48 parameters were collected that served as database for the operation and maintenance protocol of the Hostalnou de Bianya HSCW (e.g., length of sewers in m, number of HSCW units and their size, bottom slope, species of macrophytes planted, distance to receiving water and so on)

2. These data were entered into "EDSS-maintenance" by using a graphical user interface

3. From the community characteristics, the WWTP characteristics and the design parameters of the HSCW, "EDSS-maintenance" identified 17 potential problems and suggested carrying out a number of controls to monitor and prevent these problems. As an example, the suggested measures are shown in Fig. 3. Some of them need to be done weekly, others monthly.

4. From these suggestions, "EDSS-maintenance" automatically generated a monitoring notebook that can be used by the responsible maintenance person to define when the control needs to be done by whom and leaves space for taking notes of results and write down observations.

5. Finally, "EDSS-maintenance" generated an operating manual specific for Hostalnou de Bianya, that lists the potential problems, their causes, the monitoring program, the preventive and the corrective actions with their respective effects. This information is organized in so called "technical cards".

The striking advantage of this approach is that authorities and/or constructors of HSCW can easily generate a site-specific manual together with a monitoring tool, based on systematic and objective procedures. These tools allow laymen, which in most cases will have to maintain the constructed wetland, to control the facility and gain some insight into what can be done if problems occur. This can be very helpful to guarantee and even improve the long-term performance of HSCW.

Claudia Turon concludes her thesis with a set of recommendations for the full version of "EDSS-maintenance" that still needs to be developed and that should eventually be able to deal with all types of small-WWTP in Catalonia.

The annex of her thesis summarizes the results of the questionnaires given to HSCW users in Catalonia (13 HSCW).

References

[1] Turon Planella, C., 2006, EDSS-maintenance prototype - an environmental decision support system to assess the definition of operation and maintenance protocols for horizontal subsurface constructed wetlands, PhD thesis, Universitat de Girona and Laboratori d'Enginyeria Química i Ambiental (LEQUIA)

[2] Poch M., Comas J., Rodriguez-Roda I., Sànchez-Marrè M. and Cortés U., 2004. Designing and building real environmental decision support systems. Environmental modelling and software, 19, 857-873.