International Conference on Integrated Water Management, Perth 2011 1
Assessing decentralised wastewater treatment technologies:
Correlating technology selection to system robustness, energy
consumption and fugitive GHG emission
Meng Nan Chong1*, Angel Ho2, Ted Gardner1,3, Ashok Sharma4, Barry Hood3
1 CSIRO Land and Water, Ecosciences Precinct, Dutton Park Queensland 4102 2 Macao Environmental Protection Department, Taipa Macao SAR 3 Departments of Environment and Resource Management, Ecosciences Precinct, Dutton Park Queensland 4102 4 CSIRO Land and Water, Highett Victoria 3109
*Corresponding author: Dr Meng Nan Chong
CSIRO Land and Water, Ecosciences Precinct, Dutton Park Queensland 4102
Tel: +61 7 3833 5593, Fax: +61 7 3833 5501, Email: [email protected]
Key words: Decentralised wastewater, Energy consumption, Water-energy nexus, Greenhouse gas emissions,
Membrane bioreactor
Abstract
The projected population growth of 1.5 million in South-East Queensland (SEQ) by
2031 is expected to pose a serious challenge to the treatment capacities of existing
sewage treatment plants. New sewage infrastructures need to be planned and
implemented to accommodate the future treatment demands, and will include either
conventional centralised systems, or a resilient suite of decentralised technologies. In
the past, decentralised technologies have been largely viewed as an alternative for
remote or specialised boutique developments such as ecovillages. However, there is
an emerging demand for the adoption of decentralised technologies in SEQ based on
technical and cost minimisation criteria, as well as sustainability grounds. A major
limitation to the wider uptake of decentralised technologies is the lack of both
technical and scientifically-credible information on process selection using criteria
such as system stability, energy consumption and environmental sustainability. In this
study, we compared two different types of decentralised systems in SEQ, and assessed
their system robustness to shock loads (for MBR system only), energy consumption
and fugitive greenhouse gas (GHG) emissions. Both systems were designed to
produce Class A+ recycled water suitable for toilet flushing and external irrigation
water use. These systems are: (i) a holding wetwell, immersed membrane bioreactor
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 2
(MBR) with anoxic and aerobic zones, ultraviolet irradiation (UV) and chlorination
and; (ii) communal septic tanks, anoxic and aerobic bio-filtration, microfiltration, UV
and chlorination. To evaluate the stability of MBR to different shock loads, we used a
modelling approach whereby an industry standard activated sludge BioWin® model
was calibrated and validated before conducting a series of “virtual” dynamic shock
loads experiments. We found that the MBR system was relatively robust to hydraulic
shock loads with tolerance up to 1.5 times of the design dry weather daily flow
without violating the plant’s licence requirements. However, the stability of
nitrification process in MBR was significantly affected when the total nitrogen load in
the influent increased by 30% whilst maintaining the constant inlet wastewater flow
rate. Once upset, it took approximately 12 hr for nitrification behaviour to recover.
Such sensitivity did not occur with carbonaceous (COD) shock loads. For the energy
consumption study, we found that the specific energy requirement (kWh/kL of treated
sewage) for the MBR system was 6.1 kWh/kL, which was substantially higher than
that for the other decentralised aerobic bio-filtration system (1.9 kWh/kL). We also
used a mass balance approach to estimate the fugitive GHG emissions (CH4, N2O)
and concluded that electrical energy consumption data alone could substantially
underestimate the overall GHG footprints for the decentralised systems. When the
estimated CH4 fluxes were added to the initial low electrical energy consumption, the
communal septic tanks with aerobic bio-filtration system generated a carbon dioxide
equivalent footprint similar to that of the MBR system.
1. Introduction
Significant and continued population growth in most of the major urban regions in
Australia, and in particular, the South-East Queensland (SEQ) region has placed an
increasing pressure on wastewater service providers to develop a response plan to
cope with their obligations. By 2031, it is expected that the urban population in SEQ
will grow by more than 1.5 million people (DIP, 2009). In the conventional “end-of-
pipe” paradigm for wastewater services, it is expected that all these urban residential
developments spread over about 32,767 ha of new urban areas can be connected to the
existing major wastewater treatment infrastructures by expanding the existing
centralised sewage collection, conveyance and treatment systems. If we utilise the
“business-as-usual” solutions for new urban growth areas, there will be a high capital
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 3
cost and pumping energy penalties associated with long conveyance distance, as well
as operational difficulties due to the age and capacity of the current sewage treatment
plants.
In view of this, decentralised technologies covering a range of wastewater treatment
technologies could be an “ideal” solution to accommodate urban growth by providing
location specific sewage treatment options that can outweigh most of the negative
aspects associated with centralised systems (Tjandraatmadja et al., 2009). In the past,
decentralised technologies have been largely viewed as an alternative for remote or
specialised boutique developments such as ecovillages. With the emerging suite of
decentralised technologies, however, conventional technology such as septic tank can
be combined with advanced decentralised wastewater treatment technologies to
deliver treated sewage quality of up to Class A+ recycled water. The advanced suite of
decentralised treatment technologies include attached biological media (biological
activated carbon, biofilters), adsorption processes using sand or clay materials,
membrane technologies such as membrane bioreactor (MBR), microfiltration (MF),
ultrafiltration (UF) and reverse osmosis (RO) and tertiary disinfection treatments such
as UV sterilisation and chlorination. The availability of such an advanced suite of
wastewater technologies has certainly enhanced the wider uptake of decentralised
systems, as it allows for “fit-for-purpose” application with greater flexibility in
process selection and matching specific end-uses. However, a major limitation to the
wider uptake of decentralised technologies is lack of both technical and scientifically-
credible information on process selection using criteria such as system stability,
energy consumption and environmental sustainable (i.e. the total associated GHG
emissions).
In this study, we compared two different decentralised systems in SEQ and correlated
their system behaviour to shock loads (for MBR system only), energy consumption,
and fugitive greenhouse gas (GHG) emissions. These decentralised systems are
located at Capo di Monte (Mount Tamborine) and Currumbin EcoVillage (Currumbin
Valley) and were designed to produce Class A+ recycled water for toilet flushing and
external irrigation use. A 6-day diurnal wastewater quality sampling was conducted
for the MBR plant at Capo di Monte (CDM) in order to calibrate a commercially
available activated sludge BioWin® model. Using the calibrated model, the impacts
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 4
of various shock loads including hydraulic, nitrogen and carbonaceous COD on the
decentralised MBR operation were evaluated. In addition, we also measured the
specific energy consumption for process equipments using data-logged smart water
and energy meters. We complemented this energy data with a mass balance approach
to estimate the fugitive GHG emissions from certain treatment components such as
communal septic tanks, MBR and aerobic bio-filtration processes.
2. Site, Process Description and Monitoring Equipments Setup
2.1. Capo di Monte Sewage Treatment Plant (CDM-STP)
Capo di Monte (CDM) decentralised sewage treatment plant, located at Mount
Tamborine, was built within a 4.3 ha urban residential development that comprises of
46 detached and semi-detached residential dwellings and a large community centre.
This urban development was built as a “retirement village” theme that caters for
“over-50s” individuals, and each dwelling was designed with one to two bedrooms.
The main reason for the adoption of decentralised technologies was the absence of a
centralised sewer collection system. The CDM-STP was designed for a hydraulic
capacity of 11,000 L/day based on estimated peak sewage flows. Fig. 1 shows a
schematic of decentralised wastewater system. The CDM-STP involves a raw sewage
holding wet-well followed by submerged flat sheet MBR (Kubota) that incorporates a
raked screen, anoxic/aerobic treatment zones, alum dosing in aerobic zones (for
phosphorus removal) and tertiary disinfection that includes UV disinfection and
chlorination. A submersible pump in the aerobic MBR zone allows for a return
activated sludge (RAS) stream back to the anoxic zone. Excess activated sludge from
the anoxic zone is pumped out on a fortnightly basis to a Gold Coast regional sewage
treatment plant for further treatment. Class A+ recycled water is produced from this
STP, and used for household toilet flushing and external irrigation via a dual
reticulation system. A vegetated buffer zone of 6,000 m2 is available for land
application of excess treated wastewater to prevent direct discharge into the local
waterway.
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 5
Fig.1: Schematic of decentralised wastewater system at Capo di Monte. RAS flow is
the return activated sludge stream.
2.2. Currumbin Ecovillage STP
Currumbin Ecovillage (CEV) STP is situated at the Currumbin Valley, Gold Coast
and comprises 110 residential lots that range from 400 to 1,600 m2 with an extensive
proportion for communal open areas (80:20 of open-to-living space). The main reason
for the uptake of decentralised technologies was due to the unavailability of access to
a centralised sewer network. The CEV-STP has a design capacity of 51,000 L/day for
raw sewage treatment. Fig. 2 shows the schematic for the treatment processes, as well
as its wastewater flow lines. The wastewater is collected at each household and
conveyed to the STP using a combination of gravity and sewer pumping. The initial
anaerobic treatment is performed by three in-series septic tanks with a BioTube® filter
installed in the last tank to remove carry-over solids. The sewage effluent is then
treated in a secondary process of aerobic bio-filtration and denitrification. An Orenco
Advantex® Textile Filter (AdvanTex AX100) is used for the simultaneous aerobic
degradation and nitrification of carbonaceous and nitrogen compounds in the primary
treated effluent. A proportion of the treated effluent from the textile bio-filters is
recycled back to an anoxic/recirculation tank to allow a denitrification process to
occur. This recycling ratio is a crucial process parameter and is currently set at a 5:1
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 6
ratio (Xavier, 2008). This recycling ratio means that only one sixth of the wastewater
flow in a full pumping cycle is diverted for subsequent downstream treatment, whilst
the remaining flow fraction is recycled continuously to ensure sufficient BOD
reduction is achieved. The diverted effluent is treated to a Class A+ recycled water via
microfiltration (with an effective pore size of 0.2 µm) follows by UV disinfection and
chlorination. The Class A+ recycled water produced from the CEV-STP is stored in a
large storage tank before being reticulated to the households for toilet flushing and
external irrigation use.
Fig. 2: Schematic of decentralised wastewater system at Currumbin EcoVillage.
3. Results and Discussion
3.1. Influence of various shock loads to MBR operational
Membrane bioreactor (MBR) is an emerging technology for wastewater treatment that
is capable of transforming various types of wastewater into high quality treated
effluent, equal or exceeding almost every discharge requirement. Unlike conventional
activated sludge process, MBR usually comes in a small physical size, produces less
activated sludge, and achieves higher biomass concentration for organic
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 7
mineralisation (Gander et al., 2000). All these characteristics made MBR an attractive
technology option for decentralised wastewater treatment. To date, there are only a
few published studies that discuss on the potential application of MBR for small-scale
decentralised wastewater treatment (Gander et al., 2000). Most of the current design
knowledge and guidelines on MBR plants are applicable to large scale centralised
WWTPs. Thus, there exists an imperative to close the knowledge gaps on design and
implementation for small-scale MBR plants.
Abbeglen et al. (2008) also discussed that the conventional MBR design cannot be
applied directly to a decentralised system before detailed considerations are resolved.
This is due to the wastewater flows being subjected to high fluctuation in volume and
composition, and shorter residence times. The decentralised MBR needs to be
designed with a certain “buffering capacity” for wastewater flow rates and other
potential nutrient and pollutant perturbations. Such fluctuations in influent wastewater
characteristics, however, can be dampened by the provision of flow equalisation or
buffer tank prior to treatment in the decentralised MBR system (Sipma et al., 2010).
Membrane bioreactorInfluent Effluent
WAS
Anoxic
Alum addition
Fig.3: Simplified process schematic in BioWin® simulation model. WAS represents
the waste activated sludge stream.
To understand and evaluate the impacts of various shock loads to the decentralised
MBR operation directly (neglecting the upstream buffer in this instance) at CDM, we
set-up and calibrated the dynamic activated sludge system model in BioWin®
(EnviroSim Associated Ltd, USA) based on the simplified process schematic in Fig.
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 8
3. The bio-kinetic model used was the ASM1 (Henze et al., 1987) for organic matter
and nitrogen, with the parameter set from Vanrolleghem et al. (1999). As phosphorus
is removed via alum dosing at CDM, the biological P-removal process is not an
important part of our simulations. Model calibration included both the steady and
dynamic state simulations, which involved the matching of sludge production from
measured plant data with the modelled data sets (dynamic simulation was achieved
via matching of COD values).
Table 1 shows the summary of license requirements, measured influent wastewater
quality at CDM-STP, and its comparison with the common values from centralised
WWTPs. It is evident that the wastewater composition for the decentralised systems
exhibited wide variations in COD, BOD and total nitrogen concentrations compared
to those of the centralised treatment plants. This is almost certainty due to the small
connected population whereby variations from individual households are not buffered
very well compared with the large connected population in a centralised sewage
system.
Table 1: Summary of license requirements, measured influent wastewater qualities at
CDM-STP and its comparison with common values from centralised WWTPs.
Wastewater Parameters Units
CDM-STP License Limits
CDM-STP Influent
Values Range
CDM-STP Average Values
Common Values Range at
centralised WWTPs
CODtotal mg/L - 590 - 1060 825 314 - 438*
BODtotal mg/L 10 240 - 430 335 120 - 190#
Ntotal mg/L 10 69 - 140 105 87 - 94*
Ptotal mg/L 5 14 - 27 21 - Suspended solids (TSS) mg/L 10 120 - 260 190 144 - 207* Volatile Suspended solids (VSS) mg/L - 120 - 180 150 125 - 168* From *Pollice et al. (2004) & #Freeman et al., (2009)
From the BioWin® simulation result, we found that the MBR system was relatively
robust to hydraulic shock loads with tolerance up to 1.5 times of the design dry
weather daily flow without violating the licence requirements at CDM. In this
instance, the susceptibility of the MBR to different hydraulic shock loads was found
to be highly dependent on its effective “working” volume, hydraulic retention time
(HRT) and solid biomass concentration (a function of sludge retention time, SRT).
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 9
Since the “working” volume for the MBR volume is constant, the MBR robustness
was found to be highly dependent on both the HRT (1-2 hr) and SRT (approximately
200 d) used. Wastewater flows fluctuation of more than 50% is uncommon (as
observed from the measured diurnal flow patterns) and thus the dynamic simulation
has been constrained at this threshold limit.
When a stepwise increase in nitrogen shock loadings was applied to the model, we
found the MBR operational stability was impacted at N-loads greater than 30% from
its average value listed in Table 1. This was simulated assuming the inlet wastewater
flows remained within normal operational range. From the model simulation, it was
found the high susceptibility to increasing N-loads is due to the low biomass
concentration and substrate utilisation rate as estimated using the default ASM1 bio-
kinetic parameters that affects both the autotrophic and heterotrophic growth
processes. Further work needs to be carried out to determine whether the current
default bio-kinetic parameters can be applied to the specific MBR system at CDM.
Once the nitrification process was upset, the overall MBR system took approximately
12 hr to re-establish steady-state operation.
In contrast, it was identified that there are no net impacts of carbonaceous COD shock
loads (590 – 1060 mg/L) on the MBR operation. This is probably due to the high
concentration of mixed liquor suspended solids (MLSS) of 24,000 mg/L within the
current MBR system that can cope with the COD load variations. However, high
concentrations of non-biodegradable COD can accumulate within the MBR and
subsequently impair its operation. Thus, a further analysis on the COD fractionation
(dissolved, particulate and non-biodegradable components) in the local wastewater
needs to be carried out in order to understand its effects on the subsequent
decentralised MBR operation.
Currently we are also simulating both the steady-state and dynamic-state wastewater
treatment conditions for the decentralised system at CEV. The CEV-STP presents a
challenge as BioWin® cannot be used for process modelling and as we are using the
empirical National Research Council (NRC) and first-order formulation methods to
simulate the impacts of shock loads on its process operation (Crites and
Tchobanoglous, 1998).
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 10
3.2. Energy requirements of decentralised technology
Decentralised technology is regarded as being less resource intensive, more
environmentally benign and more precautionary in terms of its ability for (1)meeting
location specific solutions; (2)targeting costly augmentations to centralised systems
and (3)avoiding the financial risks inherent for new large wastewater infrastructures
(Fane et al., 2006). However, there is still a lack of credible scientific data on
sustainability issues such as energy use and carbon footprints which are needed to
facilitate its wider adoption. In this study, we monitored the energy use of various
components of the decentralised wastewater technologies for energy-related carbon
emission (Section 3.2) and also estimated fugitive GHG emissions for direct carbon
emission (in Section 3.3).
Fig. 4: Comparison on specific energy use for wastewater pumping and treatment at
our studied sites to other recycled water schemes in Australia.
Fig. 4 shows the specific energy use (kWh/kL) for our monitored decentralised
wastewater systems that produced Class A+ recycled water. The CDM-STP was found
to consume 6.1 kWh/kL whereas, the CEV-STP has a much lower total specific
energy requirement of 1.9 kWh/kL. Result in Fig. 4 also shows the specific energy
requirements for both decentralised systems are higher than the centralised
wastewater treatment facilities in Pimpama-Coomera (Gold Coast) , but are similar to
the energy requirement of the Western Corridor purified recycled water (PRW),
Tugun RO desalination plant and Sydney desalination plant (Hall et al., 2009;
Kenway et al., 2008; Australia Institute Ltd., 2005). Such a comparison suggests that
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 11
decentralised systems have a real prospect to deliver alternative urban water resources
at a better energy cost, if the decentralised systems can be properly selected,
configured and operated.
3.3. Estimation on GHG emissions from decentralised technology
As indicated earlier, there is a deficiency in the current national carbon accounting
methodology for GHG emissions from decentralised wastewater systems. To date,
virtually no information is available on the GHG emissions from different
decentralised wastewater systems. Energy-related GHG emissions are the
predominant source of GHG emissions from wastewater treatment (Kenway et al.,
2008). However, the non-energy related GHG, often referred to as fugitive GHG
emissions such as methane (CH4) and nitrous oxide (N2O), are also of significance
owing to their high global warming potential (GWP). Both CH4 and N2O are reported
to have a GWP of 25 and 298 times of carbon dioxide equivalent (CO2-e)
respectively, over a 100 years period (Foley et al., 2009). Due to uncertainties in their
magnitude and the lack of standard measurement protocols to quantify the fugitive
GHG emissions, there is limited information available to use as an informed selection
guide for sustainable decentralised technologies.
Previous studies by Kinnicutt et al. (1910) and Winneberger (1984) found a high CH4
emission potential from decentralised septic systems. However, conversion of their
data into a statistical distribution that can be used to predict CH4 emission was of
limited value due to the low sample numbers. Consequently, other CH4 predictive
models were investigated to predict CH4 emission rates, such as the Inter-govermental
Panel on Climate Change, IPCC models (1996, 2007), the Sasse model (1998) and the
Foley model (2009). We found that the estimated CH4 emission rates using these
proposed models are highly variable owing to the different underlying assumptions
made. Table 2 summarises many of the relevant studies and the assumptions for their
CH4 models, along with the magnitude of CH4 emission rates that were estimated
using the respective models. Summary results show that the current IPCC model
yields the highest estimate on CH4 emission rate compared with Sasse (1998) and
Foley models (2009).
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 12
Table 2: Summary of key assumptions of each CH4 emission models from sewage,
along with some field measurement values from the literature.
Reference Model assumption CH4 estimate
(g CH4/capita.d)
Kinnicutt et al. (1910) *Measured value 10.1
Winneberger (1984) *Measured value 14 – 18
Sasse (1988) *Estimated value assuming 25% CH4 dissolved 18
IPCC (2007) *Estimated value based on 50% BOD is converted anaerobically 25.5
Foley et al. (2009) *Estimated value based on that 40% of solids are removed as septage 11
Leverenz et al. (2010) *Measured value 11
Similarly, there is also limited information available on nitrous oxide emission from
wastewater treatment plants. Only recently have Foley and Lant (2009) reported on an
off-gas measurement method to quantify the direct GHG fluxes from seven full-scale
wastewater treatment plants (WWTPs) across Australia. In their study, they found that
the magnitude of N2O emissions from WWTPs ranged from 0.006 – 0.253 kg N2O-N
per kg N denitrified. Kampschreur et al. (2009) also reviewed N2O emission during
biological nutrient removal processes and found that the nominal direct N2O
emissions lay between 0-4% of the total nitrogen loads in the influent wastewater.
The N2O emission rate during wastewater treatment can be estimated using a mass
transfer kinetic expression (Eq. 1), with an empirical volumetric mass transfer
coefficient kLa that can be tailored to the process conditions studied (Eq.2) (Foley et
al., 2009).
[ ] }N]O[NNON{akVTrTr *
S2R2LRRN,N2OWWTPN2O, ∑ −−∑ −×== −
(Eq. 1)
0.86
g
0.49
L
R
L34,500υ
D
Dak ×
=
−
(Eq.2)
where TrN2O-N,R is the N2O emission rate from individual treatment vessels; VR is the
volume of reactor zone, [N2O-N]R is the dissolved N2O-N concentration in the reactor,
[N2O-N]S is the saturation N2O-N concentration in water at atmospheric conditions
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 13
(2.57x10-4 g/m3), kLa is the volumetric mass transfer coefficient, DR is the depth of the
reactor, DL is the depth of the lab stripping column (0.815 m), and υg is the superficial
gas velocity of the reactor.
Based on the literature, we have assumed that the major fugitive GHG emissions in
our study are sourced from the anoxic/aerobic zones in the MBR system at CDM STP,
and the communal septic tanks, aerobic bio-filtration and denitrification processes at
CEV STP. Both the CH4 and N2O emission rates are preliminary estimates based on
the Sasse (1998) and Foley (2009) models respectively. Table 3 shows the overall
GHG emissions from the two decentralised wastewater systems that include the
measured energy-related GHG and the estimated fugitive GHG emission values. The
GHG emissions from other small components (e.g. landfill, irrigation, etc) were
estimated based on the first-principle solid and mass balance approach. Results from
Table 3 indicate that the potential of fugitive GHG emissions from CEV can
significantly exceed the high energy-related GHG measured for CDM. The overall
GHG emissions from CEV is estimated at 7.06 kg CO2-e per kL of treated wastewater
compared with 5.96 kg CO2-e per kL for CDM (i.e. a reversal of magnitude when
only energy-related GHG was considered).
If the current IPCC protocol were utilised, the estimation uncertainty associated with
CH4 emission at CEV can be increased by an additional 40% from the current values.
Fig. 5 shows a stochastic log-normal distribution of the estimated CH4 emission from
the communal septic tanks at CEV-STP using the IPCC model. In this instance, a log-
normal distribution is used because of the assumption that CH4 emission from septic
tanks always exist, and have unlimited positive emission potential. It was estimated
that in the 90% probability distribution range, the communal septic tanks have a
methane emission potential of 6.69 to 15.18 kg CO2-e/kL. When these estimated
values are added to the energy-related GHG, the total GHG emissions for CEV-STP
increases to 8.83 to 17.32kgCO2e/kL.
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 14
Table 3: Overall GHG emissions from the decentralised wastewater systems.
Estimated GHG emissions
(kg CO2-e per kL)
Components
CDM CEV
Current average daily wastewater flows (kL/d) 9.3 50.5
Energy related GHG emissions from imported electrical power 5.59 1.81
CH4 emissions from identified decentralised process 0 4.92
N2O emissions from identified decentralised process 0.23 0.22
Landfill disposal of screens, grit and bio-solids 0.01 0
Effluent disposal for irrigation 0.02 0.03
Dissolved CH4 in raw sewage 0.08 0.08
Chemical and fuel consumption 0.03 0
Total GHG emissions 5.96 7.06
Fig. 5: Stochastic log-normal distribution of estimated CH4 emission from the
communal septic tanks at CEV STP using IPCC model
It should be noted that our estimates of fugitive GHG emissions from both
decentralised systems are based largely on assumptions, predictive models and
literature values from centralised WWTPs. The “true” behaviour on fugitive GHG
emissions from wastewater treatment facilities might be different across treatment
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 15
capacities and scales, and might well be affected by the influent wastewater
characteristics and local conditions at the decentralised systems. Hence, our results
should be used only as a preliminary guide for the selection of sustainable
decentralised technology for wastewater treatment. Further direct measurement of
CH4 and N2O is needed in order to validate the modelled fugitive GHG values for
their use in assessing the sustainability of decentralised wastewater technologies.
4. Conclusion
This work has provided a new insight into understanding the operational stability and
GHG impacts of various decentralised wastewater technologies. From the outcomes
of this study, it can be concluded that MBR operated at a decentralised scale offers an
excellent treatment option in terms of final treated effluent qualities (i.e. meeting the
license requirements), system robustness, and resistance to various hydraulic and
pollutant shock loadings. When the MBR is coupled with an upstream primary
holding tank, the susceptibility to various shock loads was enhanced owing to its
buffering capacity. However, the utilisation of MBR at CDM comes at the expense of
high specific energy use (kWh per kL of treated sewage) which disadvantages the
MBR in terms of energy-related GHG emissions. In comparison, the decentralised
CEV-STP provides an effective solution to treat the sewage effluent to Class A+
recycled water with a much lower specific energy-related GHG emissions. However,
when the fugitive GHG emissions from the communal septic tanks at CEV are
included, the high CH4 emission potential along with its uncertainties makes the CEV
system as operated relatively unattractive. Taken in total, our findings have provided
some useful technical insights into the decentralised technologies selection to achieve
sustainable operations in future urban developments. Further studies are currently
under way to model and optimise the wastewater treatment efficiency and energy use
in order to reduce the total GHG emission. Significant efforts will also be directed to
measuring and validating the fugitive GHG emissions from our studied sites.
Acknowledgement
The authors would like to thank the management and staff of Capo di Monte
(especially Mr Gary McOrmish) and Currumbin Ecovillage for their valuable
feedback, information, cooperation and assistance. Special thanks also due to Mr
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 16
Peter Tardrew (CEV), Mr Dominic Xavier and Mr Ivan Bragg (Sustainable Solutions
International P/L, Brisbane) in providing technical assistance and information on the
decentralised system operation at CEV.
This work was funded by the Urban Water Security Research Alliance under the SEQ
Decentralised Systems project.
References
Abegglen, C., Ospelt, M., Siegrist, H. (2008) Biological nutrient removal in a small-
scale MBR treating household wastewater Wat. Res. 42, pp.338-346.
Australia Institute Limited (2005) Greenhouse implications of the proposed Sydney
desalination plant. Australia Institute Webpaper July 2005.
Crites, R.W., Tchobanoglous, G. (1998) Small and decentralized wastewater
management systems. The McGraw-Hill Companies Inc., USA.
Department of Infrastructure and Planning (2009) Broadhectare Study, South East
Queensland: Executive Summary:
http://www.dip.qld.gov.au/resources/factsheet/PIFU/seq-executive-summary.pdf
Fane, S. and Mitchell, C. (2006) Appropriate cost analysis for decentralised water
systems. Enviro ’06 Conference and Exhibition, May 9-11 2006, Melbourne.
Foley, J., Lant, P. (2009) Direct methane and nitrous oxide emissions from full-scale
wastewater treatment systems, The University of Queensland, Australia. Water
Services Association of Australia (WSAA) Occasional Paper, No.24.
Freeman, B., Jones, N., Bartels, C., Sunano, S., Itonaga, T. (2009). Operation of a
submerged hollow fiber membrane bioreactor for wastewater treatment on
meeting reclamation criteria. Available online:
http://www.membranes.com/docs/papers
Gander, M., Jefferson, B., Judd, S. (2000) Aerobic MBRs for domestic wastewater
treatment: a review with cost considerations. Sep. Purif. Technol. 18, pp.119-
130.
Hall, M., West, J., Lane, J., de Haas, D., Sherman, B. (2009) Energy and greenhouse
gas emissions for the SEQ water strategy. Urban Water Security Research
Alliance Technical Report No.14.
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 17
Henze, M., Grady, C.P.L. Jr., Gujer, W., Marais, G.v.R., Matsuo, T. (1987) Activated
sludge model no.1. IAWPRC Scientific and Technical Report No. 1. London,
UK.
IPCC (1996) Climate change 1995: The science of climate change. Intergovernmental
panel on climate change. Houghton, J.T., Meira Filho, L.G., Callander, B.A.,
Harris, N., Kattenberg, A., Maskell, K. (eds.), Cambridge University Press,
Cambridge, United Kingdom.
IPCC (2007) Climate change (2007) synthesis report. Contribution of working groups
I, II and III to the fourth assessment report of the Intergovernmental Panel on
Climate Change [Core writing team, Pachauri, R.K. and Reisinger A. (eds.)].
IPCC, Geneva, Switzerland.
Jefferson, B., Laine, A., Brindle, K., Judd, S., Stephenson, T. (1998) Towards a
sewerless city in the Proceedings of Water Environment 98: Maintaining the
flow, London.
Kampschreur, M.J., Temmink, H., Kleerebezem, R., Jetten, M.S.M., van Loosdrecht,
M.C.M. (2009) Nitrous oxide emission during wastewater treatment. Wat. Res.
43(17): 4093-4103.
Kenway, S.J., Priestley, A., Cook, A., Seo, S., Inman, M., Gregory, A., Hall, M.
(2008) Energy use in the provision and consumption of urban water in Australia
and New Zealand. A national flagships report CSIRO.
Kinnicutt, L.P., Winslow, C.E.A., Pratt, R.W. (1910) Sewage disposal. John Wiley
and Sons, New York.
Kishino, H., Ishida, H., Iwabu, H., Nakano, I. (1996) Domestic wastewater reuse
using a submerged membrane bioreactor. Desalination 106(1-3): 115-119.
Leverenz, H.L., Tchobanoglous, G., Darby, J.L. (2010) Evaluation of greenhouse gas
emissions from septic systems, WERF Report, IWA Publishing, London, UK.
Pollice, A., Laera, G., Blonda, M. (2004) Biomass growth and activity in a membrane
bioreactor with complete sludge retention. Wat. Res. 38: 1799-1808.
Sasse, L. (1998) DEWATS, Decentralized wastewater treatment in developing
countries, Borda, Delho, India.
Sipma, J., Osuna, M.B., Emanuelsson, M.A.E., Castro, P.M.L. (2010) Biotreatments
of industrial wastewaters under transient-state conditions: process stability with
fluctuations of organic load, substrates, toxicants, and environmental
Accep
ted pa
per
International Conference on Integrated Water Management, Perth 2011 18
parameters. Crit. Rev. Env. Sci. Technol (In press) DOI:
10.1080/10643380802039329
Tjandraatmadja, G., Cook, S., Ho, A., Sharma, A., Gardner, T. (2009) Drivers for
dencentralised systems in South East Queensland. Urban Water Security
Research Alliance Technical Report No. 13.
Vanrolleghem, P.A., Spanjers, H., Petersen, B., Ginestet, P., Takacs, I. (1999)
Estimating (combinations of) activated sludge model no. 1 parameters and
components by respirometry. Wat. Sci. Tech. 39(1), pp.195-214.
Winneberger, J.H.T. (1984) Septic-tank systems: a consultant’s toolkit. Volume II –
The septic tank. The Butterworth Group, Ann Arbor, Michigan, USA.
Xavier, D (2008) Personal communication to authors. Sustainable Solutions
International P/L, Brisbane Australia.
Accep
ted pa
per