Combining constructed wetlands and engineered treatment for water reuse, report WP3, Deliverable D3.1.
In this report, the treatment efficacy of four demonstration sites combining constructed wetlands with engineered pre- or post-treatment processes for wastewater treatment is evaluated focusing on the achievement of effluent quality suitable for water reuse. Special focus is given on the performance of disinfection processes and their combination with constructed wetlands targeting water reuse applications for treatment of primary effluent and polishing of secondary effluent. Monitoring results of the demonstration sites are compared to five existing legally binding national water reuse regulations of European countries, highlighting similarities and differences between these regulations. Results are furthermore compared to the EU-level water reuse standards proposed by the European Commission in May 2018: “Proposal for a Regulation of the European Parliament and of the Council on minimum requirements for water reuse” (COM337, 2018). The first part of this report focuses on the comparison of the application of water reuse in the EU and the different national regulations in Cyprus, France, Greece, Italy and Spain – countries, which incorporated water reuse standards into their national laws. Water reuse legislations vary significantly among the EU member states. Different reclaimed water uses associated with different water quality classes and varying levels of detail in definitions are considered in each regulation. The number of classes defined in the regulations varies from 1 class including 3 categories of reuse purposes in Italy to 12 classes including 24 categories of reuse purposes in Spain. The allocation of a reuse purpose to the relevant class in the different regulations may change when looking at the level of definition of the regarded reuse purpose. For example, differences in individual definitions for use types of agricultural products, such as irrigation of a “crop consumed processed” and a “vegetable consumed cooked”, may lead to the inclusion or exclusion of the same reuse purpose into different classes in some of the regulations. The same is true for restrictions of irrigation types, which can differ regarding temporal or spatial restrictions. The number of water quality parameters which are restricted by each national regulation also differs considerably, ranging from six parameters regulated by the French water reuse legislation to 55 parameters regulated in Italy. In certain cases, the number of restricted parameters can increase up to 80 (Greek reuse regulation for WWTP > 100,000 p.e.) or even 90 in Spain (when requested by regional government depending on external regulations concerning the protection of the receiving environment). Apart from defined water reuse classes, regulated parameters and relevant limit values, the national reuse regulations also differ with regard to compliance requirements, which further complicates evaluations. While some regulations specify a percentile of samples required to comply with the set limit values (e.g. 80% of annual samples need to meet the limit), others require the annual mean to comply with the limits. In addition, sometimes maximum allowed deviation limits for samples exceeding the limit values are defined. As these specifications may not only vary among different regulations but also for different parameters in the same regulation, as well as among different quality classes for the same parameter in the same regulation, an evaluation of monitoring results of the different demonstration sites in regard to the national water reuse regulations is challenging and might become confusing. The proposal of the European Commission for an EU-level regulation on water reuse includes 4 water quality classes and 4 restricted quality parameters (with two additional for certain reuse purposes). However, water reuse in this proposal is only limited to agricultural irrigation. In contrast to national regulations, the EC proposal includes performance criteria for unrestricted irrigation on top of effluent quality limits. The variability of standards and definitions for water reuse across European countries poses a barrier for the wide application of reclaimed water, resulting in an underdevelopment of the water reuse sector in Europe. The second part of the report provides a comparison of the monitoring results of four AquaNES constructed wetlands (CW) demonstration sites in Greece and Germany with European water reuse regulations. Because of the regulatory heterogeneity described above, a direct comparison of the different European water reuse regulations with monitoring data of the demonstration sites is only possible for well-defined cases, as the allocation to the relevant class in the different regulations may change when looking at the level of definition of the regarded reuse purpose. Therefore, three specific reuse cases have been defined (for details see 3.1): - restricted irrigation (irrigation of beans using drip irrigation), - unrestricted irrigation (irrigation of tomatoes using any irrigation methods) and - urban irrigation (irrigation of a public park). For both Greek sites, monitoring results were evaluated regarding respective water reuse classes of these use cases for all national legislations, while for both German sites, evaluation was only done in respect to the standards proposed by the European Commission. The two Greek sites, Antiparos and Thirasia wastewater treatment plants (WWTP), are both located on the Cyclades island group of the Aegean Sea, and are full-scale WWTPs subjected to significant season fluctuations in the hydraulic and pollution loads between summer and winter periods. The combination of a two-stage CW with chlorination-disinfection realized at Antiparos WWTP results in water quality suitable for “restricted irrigation” according to the French and Greek regulation as well as to the EU-level proposed regulation (COM337, 2018). TSS and electrical conductivity (E.C.) have been identified as the two main parameters limiting possible reuse options. Before implementation of reconstruction measures in clogged wetland beds and pond, and managerial changes for optimization of plant performance (restriction of sewage trucks per day during peak season) some limits for “restricted irrigation” were exceeded. This was mainly due to elevated TSS concentrations and temporarily due to elevated concentrations of E. coli resulting from insufficient chlorination at peak flows that exceeded the design capacity of the plant. Different constructional and managerial improvements in this plant were found to improve and equalize the performance of the plant under peak and low flow conditions in summer and winter periods. However, high values for E.C. in WWTP effluent would prevent application in countries with reuse legislations that include this parameter (i.e. Cyprus, Italy, Spain). The Thirasia WWTP combines primary treatment and photocatalysis before horizontal subsurface flow (HSSF) CWs with subsequent ultrafiltration and chlorination. The quality of treated effluent meets the requirements for the defined case of “restricted irrigation” only according to the French regulation and the EU-level proposal. Parameters limiting the effluent’s suitability for reuse are more variable among the three defined reuse purposes and among the different reuse regulations compared to the Antiparos WWTP. The only parameter exceeding the Greek limits for “restricted irrigation” is total nitrogen. Performance of the HSSF CW regarding total nitrogen (TN) removal is not optimal, thus, the average concentration of total nitrogen in WWTP effluent (50 mg/L, n=24) exceeds the limit of class 3 of the Greek reuse regulation (45 mg/L). However, values since August 2018 show an improved removal of TN that always meets the limit (mean: 34 mg/L, n=11). Further analyses are suggested to ensure the sufficient removal of TN to reliably meet the Greek limit for water reuse. Testing different dosages of titanium dioxide (TiO2) in the photocatalysis stage led to the conclusion that adding the catalyst does not considerably improve the removal of relevant parameters, and therefore is economically unfeasible. Similar and relevant removal for BOD5 and COD (~60%). and TN (~30%) were found regardless of TiO2 dosage, even without addition of the catalyst and associated chemicals. Thus, it is recommended to run this stage as aeration stage with sedimentation. In the two German sites (Schönerlinde and Erftverband), polishing stages were tested at pilot scale after full-size WWTPs. Effluent quality was evaluated for compliance with the proposed EU-level water reuse quality standards. In Schönerlinde, the combination of ozonation with two CWs differing in substrate composition (sand or lava gravel with biochar) was demonstrated. Regarding E. coli, most of the removal was accomplished during ozonation (>2 log units), which also achieved removal of various micropollutants (see D3.2). The subsequent removal in both wetland types was similar, reaching a further reduction of E. coli by about 0.5 log units and resulting in effluent quality that meets class B limits according to the proposed limits of COM337. When ozonation was not in operation, the conventional wetland (with sand as substrate) still achieved a similar effluent concentration for E. coli (2.7 logreduction), demonstrating the robustness of this combination for water reuse purposes. TSS and turbidity were well removed by CWs reaching the best class A limit for these parameters. Overall, the combination of ozonation with CWs for polishing of WWTP effluent is a good option to achieve a very good effluent quality suitable for water reuse, with the potential to reach class A quality suitable for irrigation of crop that is consumed raw with further reduction of E. coli by about 0.5-1 log units. At Erftverband, a full-scale system is built at WWTP Rheinbach for flexible treatment of combined sewer overflow (CSO) during storm events, and polishing of WWTP effluent during dry weather. Three pilot-scale retention soil filters (RSF, specific form of vertical flow CWs for the treatment of rain water and/or wastewater) were tested for >3 years with one system containing an additional layer of activated carbon, and one RSF being subjected to simulated CSO events. Regarding E. coli, only class C limit is achieved (mean log removal in wetlands about 1.5). During CSO events with high peaks of E. coli in the influent of the RSF, effluent quality does not meet the requirements for any reuse purpose defined in the EC proposal, even though a log removal of about 2.5 is achieved. A temporary disinfection during heavy rain events would be necessary in order to provide effluent suitable for water reuse. BOD5 and TSS do not limit water reuse according to the EC proposal, thus, a sufficient disinfection would allow water reuse even for class A reuse purposes. Overall, systems, which include a combination of CWs with some sort of technical system with disinfection capabilities, achieved class B effluent quality according to the proposed EU-level standards. The Erftverband site containing a natural treatment stage without an additional disinfection achieved class C quality when not subjected to CSO events. Thus, effluents of all sites would be suitable for the following reuse purposes defined in the EC proposal: (a) food crops consumed raw, where the edible portion is produced above ground and is not in direct contact with reclaimed water; (b) processed food crops and (3) non-food crops including crops to feed milk- or meat-producing animals. Whereas in class B the irrigation method is unrestricted, in class C only drip irrigation is allowed. The combination of CWs with disinfection treatment processes for wastewater treatment in small communities is a promising option for the wider application of water reuse, at least for restricted irrigation purposes.