Enveloping odour emissions
Odour management is a key part of Reading's new treatment works. Thames Water's senior consultant Ian Cranshaw and Dr Pat Coleman from Black & Veatch Contracting reportSince the early 1960s Reading's population has grown from 200,000 to around 285,000 today.
The existing Manor Farm WwTW was last significantly extended in 1963. At that time, the site was beyond the urbanised area.
Things have changed dramatically, however, and the site has seen encroachment of both domestic and commercial development. Subsequent investment since privatisation has delivered many improvements including sludge dewatering and odour control measures. However, the Manor Farm WwTW will not meet the stringent new effluent consent that comes into force in 2005. Although the works could have been upgraded, there were numerous difficulties and it was decided to construct a new works on a nearby brownfield site.
Reading Borough Council formerly owned the new site at Island Road and old sewage sludge lagoons and elements of municipal and other wastes occupied this derelict site.
The decision to build a new WwTW provided Thames Water with the opportunity to remediate a brownfield site and replace a works no longer in keeping with the neighbourhood. The new works will satisfy the new consent and produce a biosolids product for disposal to agricultural land. Through its architectural design the new facility will blend in with future development in the area. The new WwTW buildings also act as an envelope to capture emissions, hide unit processes and minimise noise.
From the outset the design presented challenges in the areas of effluent and sludge quality, and a broad spectrum of environmental issues at the forefront of which was odour management. The design team recognised these challenges could only be met if odour management was integrated with the architectural and process design from the start. The new works' aim is to prevent an odour nuisance to current and future neighbours.
The project team agreed with Reading District Council planning team that the design would achieve five odour units at the nearest receptor as 98 percentile.DESIGN AND PROCESS
It could have been acceptable to design the site to meet this condition at the nearest current receptor, currently represented by an adjacent waste transfer station. However, the design team chose to apply the standard at the works' boundary. The first step in the design was to identify what contributes to odours deemed offensive by the public. The conventional approach is to control emissions of volatile organic compounds (VOCs) and other gases, typically hydrogen sulphide (H2S) generated from the reduction of organic wastes in the absence of oxygen. These processes can both occur in the sewerage network and in the works. However, this does not account for all types of odour considered offensive.
Volatile compounds generated by aerobic activity have a sweet earthy odour. In certain concentrations, people find these as offensive as odours caused by H2S or mercaptans.
The team assessed the activated sludge process as a source of odour by taking olfactometric measurements at Thames Water's Hogsmill WwTW. Both Hogsmill and the new Reading works are configured as a modified three-stage Bardenpho process for biological nutrient removal. Although the odour emission per unit area from an activated sludge plant is low, the quantity of odour discharged is significant because of the large surface area of the process. The team recognised the need to include odour emissions from the aerobic processes in the model.
The group adopted this approach to ensure a comprehensive level of odour mitigation could be achieved.
Odour dispersion was modelled using the Complex-1 atmospheric dispersion model. The model produces the probability of exceeding a certain hourly average concentration at locations where people may be exposed. When an annoyance criterion is applied to these results, an odour unit contour can be plotted that identifies the area in which odour annoyance may occur. The design used the model's results to set a discharge standard for the odour abatement stacks and to locate the stacks within the site so they have the least impact on the site's neighbours. The basis of the design also takes into account a precedent set by case law (Newbiggin-by-Sea, 1993).
This precedent accepted that in the absence of air quality standards for WwTWs in the UK, it is reasonable to plan using the five ouE/m3 - 98 percentile of 1-hour average concentration contour. When the odour concentration exceeds five units at this contour, then odour annoyance may become an issue for a normal population. LEGAL points The design also took into account the uncertainty in the law relating to UK odour nuisance. Two processes of consultation are currently under way which may result in the imposition of an even tighter standard.
The first is from the Environment Agency (EA) which has set an out a classic IPPC approach to odour management in its paper H4 - horizontal guidance notes for odour. This paper is not directed with intent at the regulated water industry.
However, elements of the strict regime it sets out for odour, (including standards at the receptor as low as 1.5 OUE 98 percentile), could be embraced by the regulator in any future legislation pertaining to the regulated water industry. The second consultation is from Department for Environment, Food and Rural affairs (Defra). This paper sets out several alternative forms of control from a voluntary code through to a full IPPC approach. A likely outcome is that the water industry will need to agree common approaches to setting standards. To reinforce this, the government may extend the regulators' current odour responsibilities.
Public acceptance of a new WwTW for Reading depends on the how well odours are managed. For this reason, the team started at the outset to engineer a full design looking at odour and airflow minimisation, containment, treatment and dispersion. The approach taken was to design out odour production at the outset.
Bolting on an odour solution after the fact was not a solution that would deliver the robust system Reading requires. Because the effluent from a large brewery arrives at the works via a separate rising main, the team chose to screen this sewage to 2mm and bypass it around primary treatment to the bioreactor. This is because preliminary and primary treatment would strip odorous fermentation by-products to atmosphere.
Instead, by contacting the brewery effluent with the return activated sludge, the biomass assimilates the odorous components in this stream before they can be stripped to atmosphere. Acknowledging there will be emissions from this stage of the process, the entire activated sludge plant is covered. The head space of the unaerated fractions are ventilated to the inlet of the main process air blowers.
This foul air stream, diluted by the major flow of process air, receives biological treatment as it passes through the aerated fraction as fine bubble diffused air. The aerated lanes' head space is extracted and the air is dispersed via a high-level stack. The extraction fans step up and down with the process air blowers.
Process selection for both the sludge and effluent streams was also influenced by the need to minimise odour at source and where necessary to secure containment and extraction.ODOUR CONTROL
The two main characteristics in any plant selected was their ability to accept odour control covers and once covered to offer an acceptable level of operability and maintainability. Ideally, all serviceable components were to be located outside the odour control cover. The adoption of a design for enhanced treated sludge removed the need for secondary digestion for stabilisation purposes.
However, a strategic sludge store was still needed to support the intermittent nature of the disposal route to agriculture. In early stages of design development, this featured as an open cake pad.
The odour modelling activity concluded that irrespective of location within the site the pad would need containment and odour treatment.
This need was met by the provision of cake silo storage. The high aspect ratio of the silos minimised the requirement for ventilation and odour treatment. The works produces three types of materials that must be taken off-site for re-use or disposal: screenings, grit and biosolids. Screenings and grit must comply with the requirements of the Landfill Directive. The grit and screenings are cleaned, dewatered and stored in an odour controlled environment.
One of the project aims was to maintain the practice of recycling treated sludge to agriculture. To sustain this route, the product must be free of Salmonella and have been treated to destroy 99.9999% of the pathogens. The Safe Sludge Matrix refers to this as being an enhanced treated product. An enhanced treated biosolids product is less likely to cause odours when stockpiled prior to being spread on agricultural land. At many UK works, up to 40% of H2S emissions are generated by poor design or by poor process control of the sludge stream. To prevent this happening at Reading, the time it takes for the sludge to pass from the primary tanks and the bioreactor to the silos is tightly controlled.
All equipment chosen for this project, including the gravity sludge thickening belts and dewatering centrifuges, have a track record in odour containment. The thickening and dewatering technologies chosen for this project also produce high quality sludge liquor. All liquors are kept separate and are pumped back to the works as they are generated. The production of high quality fresh liquor reduces the risk of uncontrolled odour emissions. Entraining the liquors in the effluent stream flow downstream of the inlet screens further reduces this risk. Despite the team's best efforts to minimise odours, the remaining odours must be contained and not allowed to enter the environment without treatment. The design is based on odour generation rates observed at other Thames Water sites.
The design team built an odour mass and flow balance model to help select the optimal odour management solution. Because of recent failures in the structural supports of GRP odour containment covers at other sites in the UK, a corrosion risk assessment was undertaken.
The outcome was the crude sewage reception chamber and digester limpet box concrete are protected against corrosion. The sludge, sludge liquor and aeration tanks have flat concrete covers and therefore not at risk. All unit processes in the inlet, sludge and dewatering buildings are under local control.
The building envelope, also odour controlled, acts as a second level of containment. The storm tanks are buried, covered and odour controlled. This is not the approach taken in the lamella building. The team visited a number of sites in the UK where lamella settling tanks are covered. In most cases, the covers hinder operation's ability to maintain and operate the settlers. For this reason, the lamella tanks at Reading are not covered and the lamella building air space is extracted to the odour control plant at six air changes per hour.Finding solutions
Dispersion modelling confirmed that air collected from the unit processes and buildings as well as exhaust from appurtenances burning biogas could not enter the atmosphere untreated without causing an odour nuisance.
The abatement solution that provided the best whole life cost consisted of:
- maintenance iron dose into the inlet works to reduce the sulphide emissions in the inlet works and the biogas (reducing sulphides in the boiler and combined heat and power engine exhausts),
- chemical scrubber with standby regenerable carbon filters for the inlet and lamella buildings and the sludge building. The carbon units offer equivalent performance to the wet chemical units. The whole life cost is least when the carbon units are used as standby units. The provision of carbon filters rather than a second chemical system also prevents odour emissions if the chemical supply to the works is interrupted,
- biological treatment in the bioreactor aeration lane of the foul air off gas from the unaerated lanes. Other options, including biofilters and further use of the bioreactor, were rejected because of either land take, ducting or risk. For example, a low technology biofilter along the site perimeter was rejected because of the amount of ducting required.
Because the buildings are used for containment, a traffic light system linked to the status of the ventilation system and air monitors will warn operators if they are at risk should they enter a building.
An undetected hydrocarbon spill entering the works poses a risk to a covered works like Reading. For this reason, hydrocarbon monitors will be installed in the pump station's wet wells feeding the works, within the inlet works odour ducting, in the inlet works building drainage sump and at the inlet to the bioreactor.
The design team zoned the odour control ductwork for hydrocarbon emissions. On detection of a spill, the operator can use a rapid acting penstock to divert contaminated flow to the blind storm tanks.
Because of the risk of line blockages caused by struvite or carbonate scale, the digester and digested sludge storage tank limpet boxes and the joining pipework are extracted to the odour control system.
The ventilation rate is such that the methane gas concentration is diluted to 20% of its lower explosion limit.
The successful operation of the plant will require diligence in operation and an enhanced level of operator expertise with respect to the technologies employed. For this reason, the commissioning team includes part of the future operations team. A comprehensive package of online monitoring will provide management information regarding the performance of the abatement plant.
Auditable records will be secure through the site-wide Scada system. Hydrogen sulphide has been selected as an appropriate marker gas.
Online measurement of H2S at the stacks will give an indication of performance and quickly highlight any deterioration in abatement performance. Key mechanical plant items will be monitored with alarms routed out of hours to an off site-manned control and co-ordination centre.
The design process for Reading has recognised odour as one of the key drivers from concept through to operation.
The team has thought beyond classical sewage and sludge septicity and taken a site-wide approach encompassing both the main process streams and residuals such as grit and screenings.
The mass balance approach taken in the ventilation design has allowed the team to appreciate a hierarchy in odour sources and realise minimal odour generation by design.
The strategy for minimisation supported by local containment and ventilation has enhanced the working environment and promoted the use of architecturally appropriate structures. The span of odour abatement technologies employed is a clear demonstration that there is no panacea for odour.
This observation highlights the fact that any future document attempting to describe best practicable means will need to allow a good measure of flexibility in design.
The practice of benchmarking the odour dispersion design against equivalent processes has given the team a high level of confidence in the design. Thames Water, with its alliance partners Black & Veatch and Taylor Woodrow, seized a major opportunity in setting standards for odour at Reading during the conceptual design phase.
It has allowed the team to select process solutions that are not only amenable to the design and build element but those which present a sustainable operation