Wastewater reclamation is an upcoming approach that will significantly affect wastewater treatment systems. Despite the fact that technology for treating wastewater to an effluent water quality that meets various quality standards for reuse is already available and applied, the reuse of water is not just a simple and straightforward road. Significant additional energy and civil infrastructure is required to treat the water to a standard that allows it to be safely reused. The total impact of treating and reusing water may be higher than the environmental benefits. Thus, it is crucial the life-cycle impacts from upstream and downstream processes of various reuse technologies, i.e. production of chemicals, energy use, eutrophication, sludge handling, etc. The present paper provides a comprehensive evaluation considering different reuse purposes, treatment technologies and plant size. The results of this study suggest that all these factors are highly significant for the environmental impact of wastewater treatment systems for non-potable applications.

INTRODUCTION

Wastewater reclamation, the reuse of treated wastewater, has been identified as the most significant approach to meet current and future demands of different water using sectors such as human requirements for drinking water, consumption for agricultural and industrial use and the environment, which all consume substantial quantities of water (World Health Organization 2006; ACWUA 2010; National Research Council 2012; US Environmental Protection Agency 2012). While access to fresh water is getting more costly due to environmental pollution, climate change and increased demand on water resource, the use of water for non-potable purposes can be based on reclaimed wastewater. In order to safely reuse water, solids and pathogens need to be removed for most reuse applications. Furthermore, micropollutants and emerging contaminants may need to be removed in other reuse applications. Since this cannot be achieved with traditional secondary treatment alone, additional tertiary and disinfection steps are required. Besides the efficiency of a process to reduce targeted substances, the environmental impact of the wastewater treatment process itself, in terms of the energy and the use of chemicals necessary to operate the process, has been discussed by several authors (Pillay et al. 2002; Lundie et al. 2004; Kennedy & Tsuchihashi 2005; Tangsubkul et al. 2005; Friedrich et al. 2007; Memon et al. 2007; Ortiz et al. 2007; Muňoz et al. 2009; Zhang 2009; Pasqualino et al. 2010).

The use of energy, specifically the use of electricity, is the dominating environmental aspect in many of these studies. This is not unexpected, since in all the referenced studies electricity is partially or entirely generated from fossil energy resources. As such, it has been identified that the environmental profile of wastewater reclamation is highly dependent on the source of electricity.

The relative importance of the actual construction of treatment systems on the overall environmental impact depends on the type of materials, and the size and the service life of the plant. Results reported indicate that the construction phase may contribute from a few percent up to as much as 30% of the impacts from a water treatment plant (e.g. Pillay et al. 2002; Coté et al. 2005; Tangsubkul et al. 2005).

The total environmental impact of a treatment system is a function of several parameters, such as the quality requirements on (intended use of) the treated water, the treatment technology and the size of the treatment plant. Various studies have compared the environmental impacts caused by different technologies to supply water for specified purposes such as irrigation, urban use or industrial use (Stokes & Horvath 2006; Tangsubkul et al. 2005; Memon et al. 2007; Ortiz et al. 2007; Muňoz et al. 2009; Pasqualino et al. 2010; Hancock et al. 2012). In all studies except Ortiz et al. (2007), the treatment technology is the only parameter that is varied within the studies. In some of the publications (Stokes & Horvath 2005; Muňoz et al. 2009; Pasqualino et al. 2010; Hancock et al. 2012), one of the technology options is water supply entirely from a natural source, i.e. no reclamation of water. Tangsubkul et al. (2005) monitored, but did not control, the water quality of the technologies studied. Results and conclusions from these studies suggest that treatment systems with the least environmental impact, evaluated with a normalisation procedure, also provide the poorest water quality. The reuse of treated wastewater for irrigation purposes would thus be considered one of the preferred reclamation targets. This is due to the fact that almost two-thirds of all freshwater is used for irrigation (Frenken & Gillet 2012) but also because the required treatment process is considered simpler and thus more environmentally benign, e.g. nutrients do not have to be removed completely. The intended reuse purpose certainly determines the water quality criteria and therefore the required treatment technology.

As already indicated, there have been a limited number of papers involving several aspects at the same time. No paper has taken into account the influence of all three parameters, water quality (or intended water use), treatment technology and plant size, and studied the environmental impact of water treatment under well-specified conditions. These conditions include one common well-characterised wastewater as a starting point and a specified geographical setting, using pilot-plant experiments and full-scale designs as a data source and life-cycle assessment (LCA) methodology with a complete set of impact assessment categories as an assessment tool.

The goal of the present study, therefore, was to evaluate different wastewater treatment systems for sustainable reuse of treated wastewater for different plant sizes and reuse purposes. The starting point was to assess ad improve the environmental profile of eight treatment trains that combine the secondary treatment (sequencing batch reactor, SBR) with different state-of the-art tertiary treatment technologies and disinfection steps. The environmental impact assessment and comparison of the treatment trains was done using LCA with a set of environmental key performance indicators (KPIs). The set of indicators made it possible to analyse the environmental implications of choices of technology for the purpose of water reclamation.

MATERIALS AND METHODS

System assessed

The presented environmental assessment is based on a combined experimental, design, modelling and assessment study of several technologies for secondary and tertiary treatment of municipal wastewater with the target of wastewater reclamation (Baresel et al. 2015). The work comprised the following consecutive steps:

  1. Pilot-plant studies of the treatment systems at the R&D facility Hammarby Sjöstadsverk, Sweden

  2. Use of the results to construct mass balances of the treatment systems

  3. Design of full-scale treatment plants

  4. Environmental assessment of the modelled full-scale systems.

All treatment systems had a primary screening step followed by a biological treatment with a continuous inflow sequential batch reactor (SBR or ICEAS™ from Sanitaire, Xylem). The SBR, operated in various modes, was followed by eight variations of tertiary treatment trains.

The sludge treatment consists of pre-thickening, aerobic sludge stabilisation step, and dewatering. Aerobic sludge stabilisation is selected, as this is the preferred sludge treatment in all targeted regions for wastewater reclamation. Dahlgren et al. (2014) describes implications of other sludge stabilisation methods and energy recovery possibilities on the overall assessment. The general setup of the pilot system used, the system modelled and the system used for an environmental impact assessment are presented in Figure 1.

Figure 1

System setup of the studied system with different model boundaries.

Figure 1

System setup of the studied system with different model boundaries.

The eight treatment trains are divided into three reuse categories, defined by the intended use of the reclaimed water and the technical complexity of the treatment, i.e. agriculture reuse (AG), groundwater recharge (GW) and industrial reuse (I; Table 1). The agriculture reuse partly reclaims the nutrients (nitrogen and phosphorus) in the water to the levels of 5 mg/l of ammonium and 2 mg/l of total phosphorus. While the groundwater recharge quality requires more stringent nutrient removal, less solids have to be removed. The industrial reuse decreases the level of solids even more (see also Supplementary material for more details, available online at http://www.iwaponline.com/wst/072/373.pdf). Both industrial and groundwater recharge are also evaluated with micropollutant removal as an additional treatment goal (lines GW2, GW3 and I2 in Table 1).

Table 1

Classification of the treatment systems

Train(1) Treatment train Process Description 
AG1 SBR (P-NIT) + RGSF + UV SBR(2) SBR in incomplete nitrification mode 
AG2 SBR (P-NIT) + DF + UV (P-NIT) (dissolved oxygen = 1 mg/l) 
GW1 SBR (NDN) + RGSF + UV + Cl SBR SBR in full nitrogen removal mode, 
GW2 SBR (NDN) + DF + Ozone + BAF + UV + Cl (NDN) nitrification/denitrification (dissolved oxygen = 2 mg/l) 
GW3 SBR (NDN) + Ozone + BAF + UV + Cl BAF Biologically active filter, Leopold, Xylem 
I1 SBR (NDN) + pUF + UV + Cl DF Disc filter, 10 μ mesh, Nordic Water 
I2 SBR (NDN) + sUF + Ozone + Cl Ozone Ozone treatment, Wedeco, Xylem (ozone dose 5–10 mg/l) 
I3 SBR (NDN) + sUF + UV + Cl pUF Pressurised ultra-filter, Xiga 55, X-flow (0.2–0.8 bar) 
RGSF Rapid gravity dual media filter, Leopold, Xylem 
sUF Submerged ultra-filter, GE ZW 1000 
UV Treatment with UV irradiation, Wedeco, Xylem (UV dose 12–20 mJ/cm2
CL Sodium hypochlorite treatment 
Train(1) Treatment train Process Description 
AG1 SBR (P-NIT) + RGSF + UV SBR(2) SBR in incomplete nitrification mode 
AG2 SBR (P-NIT) + DF + UV (P-NIT) (dissolved oxygen = 1 mg/l) 
GW1 SBR (NDN) + RGSF + UV + Cl SBR SBR in full nitrogen removal mode, 
GW2 SBR (NDN) + DF + Ozone + BAF + UV + Cl (NDN) nitrification/denitrification (dissolved oxygen = 2 mg/l) 
GW3 SBR (NDN) + Ozone + BAF + UV + Cl BAF Biologically active filter, Leopold, Xylem 
I1 SBR (NDN) + pUF + UV + Cl DF Disc filter, 10 μ mesh, Nordic Water 
I2 SBR (NDN) + sUF + Ozone + Cl Ozone Ozone treatment, Wedeco, Xylem (ozone dose 5–10 mg/l) 
I3 SBR (NDN) + sUF + UV + Cl pUF Pressurised ultra-filter, Xiga 55, X-flow (0.2–0.8 bar) 
RGSF Rapid gravity dual media filter, Leopold, Xylem 
sUF Submerged ultra-filter, GE ZW 1000 
UV Treatment with UV irradiation, Wedeco, Xylem (UV dose 12–20 mJ/cm2
CL Sodium hypochlorite treatment 

(1)Reuse applications. AG: Agriculture, GW: Groundwater, I: Industry.

(2)Modified SBR called ICEAS™, SBR with continuous inflow and a pre- and main reaction zone.

Within each reuse category, environmental impacts were studied as functions of two independent variables: technical design (provided in Table 1) and plant size. The influence of plant size was studied at three levels: (i) 20,000 pe (person equivalents); (ii) 100,000 pe; and (iii) 500,000 pe, where one person consumes 230 l/day.

Environmental assessment methodology

Each treatment train was assessed by an attributional LCA according to the ISO standard 14044:2006 (ISO 2006), which comprised the treatment from the influent water to the reclaimed water. The system was divided into three parts: the pilot part of the system, the modelled part, which also included the piloted part, and the peripheral part, which describes the supply of energy and other commodities (Figure 1). The modelled part consists of the core processes of each treatment train. The inflows (untreated wastewater, energy, chemicals, materials for construction, machine work, etc.) and outflows (reclaimed water, sludge, direct emissions from the process, etc.) are calculated with a mass balance from pilot-plant data, supplemented with other data, e.g. full-scale references, as necessary. Analytical pilot data are based on standard analytical methods performed by accredited laboratories. Nitrous oxide (N2O) emissions were continuously measured from the secondary treatment. The entire core process was then condensed into an input–output module, which was used as the core module of an assessment model in LCA. The LCA–software GaBi 6.3 Professional (PE International, Leinfelden-Echterdingen, Germany) was utilised. More detailed description of the method and underlying information is provided in the Supplementary material.

The functional unit was defined as 1 m3 of reclaimed water delivered by a treatment train for the intended purpose, and meeting or exceeding the specified quality requirements for this purpose. Different reclamation trains were thus considered to deliver the same function for a certain reuse category as long as the reclaimed water fulfilled at least the minimum requirements for the intended purpose.

The system boundary, as indicated in Figure 1, comprises the complete system of all sewage treatment processes including related onsite sludge handling. Inputs across that boundary include the untreated sewage influent and the natural resources, which are necessary to generate energy, produce material commodities and construction materials and services, and to transport materials to the site of the plant. Outputs are the reclaimed water at the outlet from the plant and the sludge after onsite stabilisation ready for transport to disposal. Decommissioning and disassembly of the plant and scrapping of the equipment were not considered in the system. It follows from the definition of the functional unit that the use of the reclaimed water was not part of the system. Differences in water quality may cause different environmental impacts in the usage phase, but such differences are not considered, as long as the minimum effluent quality requirements are fulfilled. This simplification allows for the comparison of the different treatment options which otherwise would be affected by the environmental impact of the effluent (and sludge).

As KPIs, maximum potential impacts (midpoint indicators) were used, as they are defined and calculated in the LCA methodology for selected impacts (see for instance Guinée et al. 2001). This means that the performance indicators used measure physical or chemical effects that have the potential to cause damage. The indicators do not describe the actual damages as such or the extent to which they actually occur at a given location. The Supplementary material provides a detailed description and definition of the environmental key performance indicators used in this study.

The time horizon for the assessment was defined as 100 years, i.e. the surveyable time, except for toxicity potentials. This had the following implications for the impact assessment:

  • Nitrous oxide (N2O) was regarded as a greenhouse gas only. Its eutrophication potential was thus not considered.

  • So-called Ecoinvent long-term emissions to freshwater (used in the Ecoinvent database 2.2 (2007)) were excluded.

For toxicity potentials, values directly available in the GaBi-tool, which are the potentials integrated to infinite time, were used. For persistent compounds in the environment, the difference between the calculated toxicity potentials at the surveyable time and at infinite time may be considerable. For example, the Aquatic Ecotoxicity Potentials (AETPs (t/s)) for marine waters may be significant depending on hydrogen fluoride (HF) emissions to air and on the time horizon to which the effect is integrated. An infinite time frame caused an anomalous impact of HF on Marine AETP that could not be supported by direct contact with the Institute of Environmental Sciences (CML, University of Leiden, The Netherlands (M. Huijbregts, Department of Environmental Science, Radboud University, The Netherlands, personal communication) and its value was therefore modified to 4.1·103 kg 1,4-dichlorobenzene equivalents per kg HF. This was considered more realistic when considering the other potentials with a 100 years period of integration as fluoride has an extremely long residence time in the marine environment. Nonetheless, the fate and effects of fluorides are extremely difficult to predict in the aquatic environment with simple models, like USES-LCA (Huijbregts 2014).

Spain was selected as a model country for implementation of the water reclamation systems. This implies that electricity supply was modelled considering the Spanish energy-grid. Commodities like chemicals were assumed to be produced in Europe and the manufacture was as far as possible modelled with average European data. However, no specific site in Spain was selected. Logistics were modelled using the assumption of 300 km transportation by truck, with some exceptions. Details are provided in the Supplementary material.

The core of the treatment system, i.e. the pilot-studied and modelled processes, was described by data, which pertain to best available technology in 2013. The most recent data were used for the peripheral processes. This means that the models of the peripheral processes reflect average technology during the period 2000–2010. The Spanish electricity is the average mix for the year 2012 (see also Supplementary material).

Data collection

Operational data for the core system were specific data from the experiments and the mass-balance modelling in Matlab (Baresel et al. 2015). Data on materials and construction of the equipment were specific design and engineering data from existing or designed installations. Upstream data to describe the peripheral processes (chemicals, energy and construction) was collected from relevant literature or from life-cycle inventory databases as modules from suitable databases: Ecoinvent 2.2 and ProfDB. See the Supplementary material (online at http://www.iwaponline.com/wst/072/373.pdf) and Baresel et al. (2015) for more details on the data used.

Data were selected to meet the following criteria in the specified order:

  1. Plausibility

  2. Consistent with the geographical boundaries

  3. Consistent with the temporal boundaries.

Data have as far as possible been chosen to fit the actual substance, but in case of missing data (data gaps) these have been filled in with analogues or approximations.

Inventory methodology

The inventory of core process equipment comprised all construction materials and their origin in natural resources, at the time of the inventory. Certain data, such as use of energy and chemicals, were received from the modelling of the treatment trains and considered for the complete reclamation train. All design data for each type of Xylem equipment were provided by Xylem Inc. and relevant data are provided in Baresel et al. (2015). For the treatment using other producer equipment, the operation guidance was provided by the supplier. This was then transformed into input–output lists for each equipment unit and used as input into GaBi-modules. Transport of construction materials to the site and required replacement of major parts of the equipment during their projected service life were included. This excludes material expenditure for daily routine maintenance, like lubricating oils, fuses, ordinary light bulbs, paint and putty. Replacement of UF-membranes and of UV-lamps was included. For each part of the equipment with a specified service life, the quantity of material was multiplied by a factor = 20/(service life of that particular equipment part). This recalculated quantity of material was then entered into the input–output lists. Each input to the GaBi-modules is thus a fraction of number of pieces of equipment, calculated as 1/(m3 of water delivered during the service life of 20 years). Generic data from the databases Ecoinvent 2.2 and ProfDB were used for this (see the Supplementary material). Average up-to-date European data were used from Ecoinvent. In some cases, average German data or average global data were used when this was considered to be of better quality or more appropriate.

RESULTS AND DISCUSSION

Figure 2 shows absolute values for each of the investigated wastewater reclamation treatment–trains (AG1 to I3) for all of the considered environmental impact categories (KPI) and for three different plant sizes: 20,000, 100,000, and 500,000 pe. The cell shading further provides information about how the environmental impacts of each treatment train changes relative to the first train shown for each plant size, i.e. AG1.

Figure 2

Environmental KPIs for all investigated treatment trains, absolute values. The full colour version of this figure is available online at http://www.iwaponline.com/wst/toc.htm.

Figure 2

Environmental KPIs for all investigated treatment trains, absolute values. The full colour version of this figure is available online at http://www.iwaponline.com/wst/toc.htm.

It can be seen that more stringent effluent qualities (GW and I) do generally imply an increasing environmental impact, here as increase in KPI-value if considering each plant size. The KPI Global Warming Potential (GWP), however, indicates a significant decrease of this impact when going from wastewater reclamation for irrigation to industrial reuse or groundwater recharge (from category IV to categories I–III; from green colour to blue in online version). This is because significantly larger amounts of nitrous oxide (N2O) are emitted from the process in the less advanced AG-treatment mode: on average 1.7% compared to 0.1% of the total nitrogen load in nitrification/denitrificationode. This is due to incomplete nitrogen removal compared with a complete nitrification/denitrification for groundwater recharge and industrial applications, an observation that is also supported by other recent studies (Chandran 2010; Graaff et al. 2010; De Foley et al. 2011). The emission of N2O outweighs lower energy consumption required to accomplish partial nitrogen removal from the wastewater. This was also indicated by the dominance analysis presented in Table 2. In addition, in the case of lower N2O emissions, energy consumption becomes the dominating effect, resulting in an increased GWP with increasing effluent quality (GW and I lines). However, for some KPIs, e.g. Terrestric Ecotoxicity Potential (TETP) and Abiotic Depletion (AD) of elements, higher effluent quality implies a very significant increase of impacts from treatment processes for all plant sizes, in many cases more than 200% (Figure 2 category VIII; red in online version).

Table 2

Largest sources of impacts for the different KPIs and the investigated treatment trains

  Agricultural use
 
Groundwater recharge
 
Industrial use
 
KPI AG1 AG2 GW1 GW2 GW3 I1 I2 I3 
GWP N2N2Energy Energy Energy Energy Energy Energy 
AP Energy Energy Energy Energy Energy Energy Energy Energy 
EP Energy Energy Energy Energy Energy Energy Energy Energy 
POCP Polymer Polymer Polymer Polymer Polymer Polymer Polymer Polymer 
HTP Energy Energy Energy Energy Energy Energy Energy Energy 
FAETP Energy Energy Energy Energy Energy Energy Energy Energy 
MAETP Energy Energy Energy Energy Energy Energy Energy Energy 
TETP Energy Energy NaOCl NaOCl NaOCl NaOCl NaOCl NaOCl 
AD, elements Energy Energy NaOCl NaOCl NaOCl NaOCl NaOCl NaOCl 
AD, fossil Energy Energy Energy Energy Energy Energy Energy Energy 
  Agricultural use
 
Groundwater recharge
 
Industrial use
 
KPI AG1 AG2 GW1 GW2 GW3 I1 I2 I3 
GWP N2N2Energy Energy Energy Energy Energy Energy 
AP Energy Energy Energy Energy Energy Energy Energy Energy 
EP Energy Energy Energy Energy Energy Energy Energy Energy 
POCP Polymer Polymer Polymer Polymer Polymer Polymer Polymer Polymer 
HTP Energy Energy Energy Energy Energy Energy Energy Energy 
FAETP Energy Energy Energy Energy Energy Energy Energy Energy 
MAETP Energy Energy Energy Energy Energy Energy Energy Energy 
TETP Energy Energy NaOCl NaOCl NaOCl NaOCl NaOCl NaOCl 
AD, elements Energy Energy NaOCl NaOCl NaOCl NaOCl NaOCl NaOCl 
AD, fossil Energy Energy Energy Energy Energy Energy Energy Energy 

GWP – Global Warming Potential; AP – Acidification Potential; EP – Eutrophication Potential; POCP – Photochemical Ozone Creation; HTP – Human Toxicity Potential; FAETP – Freshwater Ecotoxicity Potential; MAETP – Marine Ecotoxicity Potential; TETP – Terrestric Ecotoxicity Potential; AD – Abiotic Depletion (elements; fossil).

Furthermore, most KPIs increase with increasing effluent water quality while the plant size is kept constant, whereas most environmental impacts are reduced as the plant size increases. This is true not only for each impact category at each effluent quality, i.e. each cell with increasing plant size, but also for differences in each impact category with increasing effluent quality, i.e. each row with increasing plant size. In Figure 2, this means that the number of category I to III (blue) (<95%) and category IV (green) cells (±5%) in the matrix increases with increasing plant size and that absolute impacts for certain treatment trains for industrial and groundwater-recharge applications become close to the ones for irrigation reuse (AG). However, the wastewater treatment train represents a more advanced process with a higher reduction of various substances. Furthermore, and even more significant, the impact of several KPIs of large plants (500,000 pe) becomes lower than or similar to the impact of treatment line AG1 at the smallest plant size (20,000 pe). As energy consumption is a major impact for almost all KPIs, the decrease of several KPIs when increasing the plant size can be explained by decreased energy consumption per cubic metre of treated water. This is due to increased size efficiency of larger treatment systems; for example water can be moved or mixed more efficiently the larger the volume (see also Supplementary material, available online at http://www.iwaponline.com/wst/072/373.pdf).

Observed differences are caused by various materials used, different energy and chemicals required for the various processes, and other process-related characteristics. The impact of construction materials had an average of 8% of the total impact for all trains, at all plant sizes and for all KPIs. The impact of construction materials was in general largest on the KPI AD of elements, but never exceeded 37% of the total impact. No single technology (e.g. DF, RGSF, see Table 1) can be identified leading to a minimum environmental impact throughout all KPIs and for all plant sizes. For agriculture reuse, for example, the use of a disk filter seems to imply a higher impact (more than 5% difference) for only two KPIs, POCP and AD of elements. As seen in Table 2, POCP is largely influenced by the use of polymers (due to emission of volatile organic carbon released during the manufacturing process), as polymers are used for disk filter operation. When targeting groundwater recharge, the combination of dual media filter (RGSF) and UV provides the lowest environmental impacts at all plant sizes (GW1). For industrial reuse of treated wastewater, submerged ultrafiltration (sUF) in combination with UV provides the lowest impacts (I3). For the treatment trains that include ozone treatment and thus removal of micropollutants, the line with sUF and ozone (I2) has the lowest impact. The highest effluent quality is achieved with the lines GW2 and GW3 that do have the largest environmental impact for all sizes. In addition, comparing these two lines with one another, it is possible to conclude that the addition of the disk filter in order to decrease ozone dosages did not improve the environmental impact, but on the contrary it increased the impact without improving effluent quality.

Table 2 shows which factor is contributing most for each impact category for the investigated treatment trains. The same trends can be observed for the three studied plant sizes. Clearly, from all materials used to construct and operate the plant, use of energy (electricity) from fossil sources is the single most important source of impact for most of the KPIs. It is dominated by energy used for secondary treatment in the SBR and the aerobic sludge stabilisation. Therefore, tertiary treatment steps with disinfection, being only small consumers of energy, have only a small impact on the overall environmental impact, even though those steps upgrade the water quality to non-potable water reuse standards. For some KPIs, instead, the use of chemicals dominates because of either the use of sodium hypochlorite (NaOCl) as a final disinfection step or the use of polymers to enhance flocculation. For global warming, the emission of the N2O from the secondary treatment step is slightly more dominating than energy consumption when incomplete nitrogen removal is applied (AG lines).

CONCLUSIONS

The main conclusion from the presented environmental impact assessment is that increased environmental impacts caused by higher quality targets with more advanced treatment processes become less significant with increasing plant size. This implies that higher quality targets do not automatically imply an increase of environmental impacts as earlier studies have indicated, as they do not consider production scale. Instead, poorer water treatment can increase the environmental impact if considering only the treatment process and not downstream effects of, for example, substances in the effluent. This implies, for example, that agriculture reuse of treated wastewater may not be the most favoured reuse approach. Despite its less stringent effluent quality to recycle nutrients, the impact of the treatment as such can imply a higher environmental impact than treatment systems for reuse applications requiring higher effluent qualities such as groundwater recharge and industry. Therefore, it is important to consider that downstream impacts, i.e. after the actual water and sludge treatment, have not been considered. These may add additional impacts, both negative (toxicity, heavy metals, etc.) and positive (nutrient, decreased utilisation of fresh water), to the assessment but are not quantifiable in the same way as for the technical system, and would thus make a comparison of different technical system more difficult.

The study shows that the size of wastewater reclamation facilities plays a crucial role and can imply less environmental impact for advanced wastewater treatment for reuse. The results indicate that there is a clear trend for all impact categories considering increasing plant sizes. This is true for every single impact category analysed in this assessment. As process efficiency increases with increasing plant size, the overall environmental impact of various direct and indirect emissions from the treatment decreases. This behaviour, however, is not linear but strongly determined by equipment sizing and modular-based treatment processes. This is explained by the fact that some equipment is only available in certain sizes and creates a nonlinear size efficiency effect. Small-scale facilities have to face less efficient systems while larger plants can profit from multiplier effects of using module components. The actual treatment process in large plants producing high-quality effluent may thus have a lower environmental impact than smaller plants using simpler processes and lower effluent quality.

The main contributing factor in the current analysis was the use of energy for nutrient removal and sludge stabilisation. This study, being one of the first using actual measurements of nitrous oxide emissions from biological nitrogen removal processes in environmental impact assessment, indicates that the energy savings when targeting lower effluent criteria are outweighed by the increased release of nitrous oxide to the atmosphere.

The study shows that targeting sustainable wastewater reclamation systems by using environmental impact assessment needs to consider effluent water qualities, plant sizes and various treatment technologies.

REFERENCES

REFERENCES
ACWUA
2010
Wastewater Reuse in Arab Countries
.
Arab Countries Water Utility Association, Amman, Jordan
.
Baresel
C.
Dahlgren
L.
Lazic
A.
de Kerchove
A.
Almemark
M.
Ek
M.
2015
Reuse of Treated Wastewater for Non-Potable Use (ReUse) – Final Report
.
IVL Swedish Environmental Research Institute
,
Report B2219, Stockholm, Sweden
.
Chandran
K.
2010
Greenhouse Nitrogen Emission from Wastewater Treatment Operations – Interim Report
. .
Dahlgren
L.
Almemark
M.
Rahmberg
M.
Andersson
S.
Baresel
C.
Lazic
A.
2014
Life cycle assessment linked with process models for the development of water reclamation processes
.
SETAC Europe 24th Annual Meeting
,
Basel, Switzerland
.
de Graaff
M. S.
Zeeman
G.
Temmink
H.
van Loosdrecht
M. C. M.
2010
Long term partial nitritation of anaerobically treated black water and the emission of nitrous oxide
.
Water Research
44
,
2171
2178
.
Ecoinvent database 2.2
2007
Swiss Centre for Lifecycle Inventories. http://www.ecoinvent.org (accessed September 2014)
.
Foley
J.
Yuan
Z.
Keller
J.
Senante
E.
Chandran
K.
Willis
J.
Shah
A.
van Loosdrecht
M.
van Voorthuizen
E.
2011
N2O and CH4 Emission from Wastewater Collection and Treatment Systems
.
Technical report
,
Global Water Research Coalition
,
London
,
UK
.
Frenken
K.
Gillet
V.
2012
Irrigation water requirement and water withdrawal by country. AQUASTAT, http://www.fao.org/nr/water/aquastat/water_use_agr/index.stm (accessed September 2014)
.
Friedrich
E.
Pillay
S.
Buckley
C. A.
2007
The use of LCA in the water industry and the case for an environmental performance indicator
.
Water SA
33
(
4
),
443
451
.
Guinée
J. B.
(final ed.)
2001
Handbook on Life Cycle Assessment. Operational Guide to the ISO Standards
(
Gorrée
M.
Heijungs
R.
Huppes
G.
Kleijn
R.
van Oers
L.
Wegener Sleeswijk
A.
Suh
S.
de Haes
H. A. U.
de Bruijn
H.
van Duin
R.
Huijbregts
M. A. J.
, eds).
Institute of Environmental Sciences – Leiden University (CML)
,
Leiden, The Netherlands
.
ISO (International Organization for Standardization)
2006
ISO 14044 Environmental Management – Life Cycle Assessment – Requirements and Guidelines
.
European Committee for Standardization
,
Brussels
,
Belgium
.
Kennedy
L. A.
Tsuchihashi
R.
2005
Is water reuse sustainable? Factors affecting its sustainability
.
The Arabian Journal for Science and Engineering
30
(
2C
),
3
15
.
Lundie
S.
Peters
G. M.
Beavis
P. C.
2004
Life cycle assessment for sustainable metropolitan water systems planning
.
Environmental Science & Technology
38
(
13
),
3465
3473
.
Memon
F. A.
Zheng
Z.
Butler
D.
Shirley-Smith
C.
Lui
S.
Makropoulos
C.
Avery
L.
2007
Life cycle impact assessment of greywater recycling technologies for new developments
.
Environ. Monit. Assess.
129
,
27
35
.
Muňoz
I.
Rodriguez
A.
Rosal
R.
Fernandez-Alba
A.
2009
Life cycle assessment of urban wastewater reuse with ozonation as tertiary treatment
.
Science of the Total Environment
407
,
1245
1256
.
National Research Council
2012
Water Reuse: Potential for Expanding the Nation's Water Supply Through Reuse of Municipal Wastewater. National Research Council
,
Committee on the Assessment of Water Reuse as an Approach for Meeting Future Water Supply Needs, Water Science and Technology Board, Division on Earth and Life Studies
,
Washington, DC
.
Pasqualino
J. C.
Meneses
M.
Castells
F.
2010
Life cycle assessment of urban wastewater reclamation and reuse alternatives
.
Journal of Industrial Ecology
15
(
1
),
49
63
.
Pillay
S. D.
Friedrich
E.
Buckley
C. A.
2002
Life cycle assessment of an industrial water recycling plant
.
Water Science and Technology
46
(
9
),
55
62
.
Stokes
J.
Horvath
A.
2006
Life cycle energy assessment of alternative water supply systems
.
International Journal of LCA
11
(
5
),
335
343
.
Tangsubkul
N.
Beavis
P.
Moore
S. J.
Lundie
S.
Waite
T. D.
2005
Life cycle assessment of water recycling technology
.
Water Resources Management
19
,
521
537
.
US Environmental Protection Agency
2012
Guidelines for Water Reuse (No. EPA/600/R-12/618)
.
Office of Wastewater Management, Office of Water
,
Washington, DC
.
World Health Organization
2006
Guidelines for the Safe Use of Wastewater, Excreta and Greywater: Wastewater and Excreta Use in Aquaculture
.
World Health Organization
,
Geneva, Switzerland
.
Zhang
Y.
2009
Sustainability Oriented Feasibility Model for Construction Decision Making: Water Recycling Cases in Buildings
.
MSc Thesis
,
University of Toronto
,
Toronto, Canada
.

Supplementary data