Experiences from the full-scale implementation of a new two-stage vertical flow constructed wetland design.

This paper describes the results of the first full-scale implementation of a two-stage vertical flow constructed wetland (CW) system developed to increase nitrogen removal. The full-scale system was constructed for the Bärenkogelhaus, which is located in Styria at the top of a mountain, 1,168 m above sea level. The Bärenkogelhaus has a restaurant with 70 seats, 16 rooms for overnight guests and is a popular site for day visits, especially during weekends and public holidays. The CW treatment system was designed for a hydraulic load of 2,500 L.d(-1) with a specific surface area requirement of 2.7 m(2) per person equivalent (PE). It was built in fall 2009 and started operation in April 2010 when the restaurant was re-opened. Samples were taken between July 2010 and June 2013 and were analysed in the laboratory of the Institute of Sanitary Engineering at BOKU University using standard methods. During 2010 the restaurant at Bärenkogelhaus was open 5 days a week whereas from 2011 the Bärenkogelhaus was open only on demand for events. This resulted in decreased organic loads of the system in the later period. In general, the measured effluent concentrations were low and the removal efficiencies high. During the whole period the ammonia nitrogen effluent concentration was below 1 mg/L even at effluent water temperatures below 3 °C. Investigations during high-load periods, i.e. events like weddings and festivals at weekends, with more than 100 visitors, showed a very robust treatment performance of the two-stage CW system. Effluent concentrations of chemical oxygen demand and NH4-N were not affected by these events with high hydraulic loads.

amoA-encoding archaea in wastewater treatment plants: a review.

Recent evidence from natural environments suggests that in addition to ammonia-oxidizing bacteria, ammonia-oxidizing archaea (AOA) affiliated with Thaumarcheota, a new phylum of the domain Archaea, also oxidize ammonia to nitrite and thus participate in the global nitrogen cycle. Besides natural environments, modern data indicate the presence of amoA-encoding archaea (AEA) in wastewater treatment plants (WWTPs). To further elucidate whether AEA in WWTPs are AOA and to clarify the role of AEA in WWTPs, this paper reviews the current knowledge on this matter for wastewater engineers and people in related fields. The initial section coveys a microbiological point of view and is particularly based upon data from AOA cultures. The later section summarizes what is currently known about AEA in relation to WWTPs. Based on the reviewed data, future research pathways are proposed in an effort to further what is known about AEA in wastewater treatment systems.

Long-term behaviour of a two-stage CW system regarding nitrogen removal.

In the first two years of operation a nitrogen removal efficiency of 53% and a high average elimination rate of 1,000 g N m(-2) yr(-1) could be observed for a two-stage vertical flow (VF) constructed wetland (CW) system. The two-stage system consists of two VF beds with intermittent loading operated in series, each stage having a surface area of 10 m2. The first stage uses sand with a grain size of 2-3.2 mm for the 50 cm main layer and has a drainage layer that is impounded; the second stage sand with a grain size of 0.06-4 mm and a conventional drainage layer (with free drainage). The two-stage VF system was designed for and operated with an organic load of 40 g COD m(-2) d(-1) (i.e. 2 m2 per person equivalent). Data from the following years of operation showed that from the third year nitrogen elimination increased and stabilized. The median values of the nitrogen elimination rate in the first five years of operation have been 3.51, 2.76, 4.20, 3.84 and 4.07 g N m(-2) d(-1), the median value of the last three years being 3.8 g N m(-2) d(-1) and 1,380 g N m(-2) yr(-1), respectively, and the nitrogen removal > 60%. It can be assumed that the vegetation as well as the biofilm development in the two-stage VF CW system plays the major role for the enhancement of the nitrogen elimination rate.

Comparison of single-stage and a two-stage vertical flow constructed wetland systems for different load scenarios.

Constructed wetlands (CWs) are known to be robust wastewater treatment systems and are therefore very suitable for small villages and single households. When nitrification is required, vertical flow (VF) CWs are widely used. This contribution compares the behaviour and treatment efficiencies of a single-stage VF CW and a two-stage VF CW system under varying operating and loading conditions according to standardized testing procedures for small wastewater treatment plants as described in the European standard EN 12566-3. The single-stage VF CW is designed and operated according to the Austrian design standards with an organic load of 20 g COD m(-2) d(-1) (i.e. 4 m(2) per person equivalent (PE)) The two-stage VF CW system is operated with 40 g COD m(-2) d(-1) (i.e. 2 m(2) per PE). During the 48 week testing period the Austrian threshold effluent concentrations have not been exceeded in either system. The two-stage VF CW system showed to be more robust as compared to the single-stage VF CW especially during highly fluctuating loads at low temperatures.

Experiences with a top layer of gravel to enhance the performance of vertical flow constructed wetlands at cold temperatures.

In a first phase of this study it was shown that the Austrian effluent standards for organic matter could not be met in winter for vertical flow (VF) beds designed for and loaded with 27 g COD.m(-2).d(-1) (3 m2 per person equivalent). The aim of this second phase of the study was to investigate, if the performance of a constructed wetland can be enhanced, i.e. if the effluent requirements can be met, when an additional gravel layer (15 cm, 4-8 mm) is added on top of the main layer of the VF bed. The hypothesis was that this top layer would increase the thermal insulation and consequently the temperatures in the filter bed during cold periods, thus resulting in higher removal efficiencies during winter. Two VF beds were operated in parallel; one bed with such a 15 cm top layer, one without. Otherwise the construction of both beds was identical: surface area of about 20 m2, 50 cm main layer (grain size 0.06-4 mm, d10=0.2 mm; d60=0.8 mm), planted with common reed (Phragmites australis). The beds were intermittently loaded 4 times per day with mechanically pre-treated wastewater (hydraulic loading: 47 mm.d(-1); median value of the influent concentration: 505 mg COD.L(-1)). Despite a better performance during the first winter, the bed with additional top layer showed in general a very unstable performance. It is assumed that the main reason for this was that the oxygen transfer was reduced by the additional top layer so far that suspended organic matter could not any longer be degraded in between loadings. Therefore clogging of the filter occurred.

Use of faecal pollution indicators to estimate pathogen die off conditions in source separated faeces in Kathmandu Valley, Nepal.

As the introduction and promotion of dehydrating toilets progresses, the safety of handling and reuse of their biosolids remains a question. A detailed study to understand the storage conditions and the fate of selected faecal indicators was conducted on four urine diverting dehydrating toilet units, using ash as a major additive, in Kathmandu Valley, Nepal. Presumptive Escherichia coli, total coliforms, enterococci and different fractions of Clostridium perfringens were investigated under field storage conditions. In addition, chemo-physical and chemical (carbon, nitrogen, phosphorous content) parameters were investigated. Observed temperature was low in all the four toilets with a median of 24.0 degrees C, which was in the same range as the ambient temperature. pH was below the desired range of >9 and moisture level was very high (>60%). No single factor of the studied chemo-physical and chemical parameters could be found by statistical analysis to have accounted for the reduction of the indicators in any of the toilets. By time series analysis of the investigated strata in the faecal heaps (n=96), the determined reduction rate showed increasing persistence characteristics for E. coli, coliforms and enterococci with respective average log(10) reduction of -0.4, -0.3 and -0.2 per month (p<0.001). No significant reduction was observed for the different fractions of C. perfringens determined for the non-pasteurised and pasteurised fraction at 60 degrees C and 85 degrees C. 72% of randomly selected and analysed samples (n=36) were found to contain helminthes eggs. The used 6 months storage time did not prove sufficient to reach appropriate safety levels for handling and reuse of the biosolids.

A two-stage subsurface vertical flow constructed wetland for high-rate nitrogen removal.

By using a two-stage constructed wetland (CW) system operated with an organic load of 40 gCOD.m(-2).d(-1) (2 m2 per person equivalent) average nitrogen removal efficiencies of about 50% and average nitrogen elimination rates of 980 g N.m(-2).yr(-1) could be achieved. Two vertical flow beds with intermittent loading have been operated in series. The first stage uses sand with a grain size of 2-3.2 mm for the main layer and has a drainage layer that is impounded; the second stage sand with a grain size of 0.06-4 mm and a drainage layer with free drainage. The high nitrogen removal can be achieved without recirculation thus it is possible to operate the two-stage CW system without energy input. The paper shows performance data for the two-stage CW system regarding removal of organic matter and nitrogen for the two year operating period of the system. Additionally, its efficiency is compared with the efficiency of a single-stage vertical flow CW system designed and operated according to the Austrian design standards with 4 m2 per person equivalent. The comparison shows that a higher effluent quality could be reached with the two-stage system although the two-stage CW system is operated with the double organic load or half the specific surface area requirement, respectively. Another advantage is that the specific investment costs of the two-stage CW system amount to 1,200 EUR per person (without mechanical pre-treatment) and are only about 60% of the specific investment costs of the singe-stage CW system.

Bacterial carbon utilization in vertical subsurface flow constructed wetlands.

Subsurface vertical flow constructed wetlands with intermittent loading are considered as state of the art and can comply with stringent effluent requirements. It is usually assumed that microbial activity in the filter body of constructed wetlands, responsible for the removal of carbon and nitrogen, relies mainly on bacterially mediated transformations. However, little quantitative information is available on the distribution of bacterial biomass and production in the "black-box" constructed wetland. The spatial distribution of bacterial carbon utilization, based on bacterial (14)C-leucine incorporation measurements, was investigated for the filter body of planted and unplanted indoor pilot-scale constructed wetlands, as well as for a planted outdoor constructed wetland. A simple mass-balance approach was applied to explain the bacterially catalysed organic matter degradation in this system by comparing estimated bacterial carbon utilization rates with simultaneously measured carbon reduction values. The pilot-scale constructed wetlands proved to be a suitable model system for investigating microbial carbon utilization in constructed wetlands. Under an ideal operating mode, the bulk of bacterial productivity occurred within the first 10cm of the filter body. Plants seemed to have no significant influence on productivity and biomass of bacteria, as well as on wastewater total organic carbon removal.

Investigation of bacterial removal during the filtration process in constructed wetlands.

In this study, bacterial removal efficiencies of planted and unplanted subsurface vertical flow constructed wetlands are compared. Indicator organisms such as faecal coliforms (Escherichia coli, total coliforms) and enterococci, and a number of heterotrophic bacteria (heterotrophic plate counts) have been analysed from the influent and effluent of the constructed wetlands as well as at different depths (water and substrate samples). Furthermore dry matter content and total organic carbon (TOC) have been analysed and correlated. The investigated systems show a high removal rate for indicator organisms (a log removal rate of 2.85 for HPC, 4.35 for E. coli, 4.31 for total coliforms and 4.80 for enterococci was observed). In general no significant difference in the removal efficiency of planted and unplanted vertical flow beds could be measured. Only enterococci measured in the substrate samples of the main layer of the filter could a statistically significant difference be observed.

Characterisation of microbial biocoenosis in vertical subsurface flow constructed wetlands.

In this study a quantitative description of the microbial biocoenosis in subsurface vertical flow constructed wetlands fed with municipal wastewater was carried out. Three different methods (substrate induced respiration, ATP measurement and fumigation-extraction) were applied to measure the microbial biomass at different depths of planted and unplanted systems. Additionally, bacterial biomass was determined by epifluorescence microscopy and productivity was measured via (14)C leucine incorporation into bacterial biomass. All methods showed that >50% of microbial biomass and bacterial activity could be found in the first cm and about 95% in the first 10 cm of the filter layer. Bacterial biomass in the first 10 cm of the filter body accounted only for 16-19% of the total microbial biomass. Whether fungi or methodical uncertainties are mainly responsible for the difference between microbial and bacterial biomass remains to be examined. A comparison between the purification performance of planted and unplanted pilot-scale subsurface vertical flow constructed wetlands (PSCWs) showed no significant difference with the exception of the reduction of enterococci. The microbial biomass in all depths of the filter body was also not different in planted and unplanted systems. Compared with data from soils the microbial biomass in the PSCWs was high, although the specific surface area of the used sandy filter material available for biofilm growth was lower, especially in the beginning of the set-up of the PSCWs, due to missing clay and silt fraction.