Comparative life cycle assessment of aerobic treatment units and constructed wetlands as onsite wastewater treatment systems in Australia

Wastewater treatment is performed in two types of systems depending on the treatment capacity. Centralized systems are also referred to as conventional wastewater treatment plants (WWTPs), and are usually used in high population density areas where sewerage system is present. Meanwhile decentralized systems or the onsite wastewater treatment systems (OWTSs), are normally applied in low population density areas where a sewerage system does not exist (Leigh & Lee 2019; Ergas et al. 2021).

Septic tanks, composting toilets, and the aerobic treatment units (ATUs) are among the most commercialized OWTSs in developed countries (Dubois & Boutin 2018). Particularly, the ATUs are the most popular technologies in Australia. ATUs are technologies that have similarities to centralized systems that use activated sludge or biofilm bed reactors; and have acceptable removal efficiencies according to regulations provided by the Department of Building Property and Development of Queensland, Australia (Government of Queensland 2021). Conversely, nature-based alternatives such as constructed wetlands (CWs), are often considered as lower efficiency OWTSs than the ATUs (Magar 2016). Consequently, the CWs are in most cases categorized as a ‘polishing’ processes associated with septic anaerobic systems; hence these units are used as the last step prior to the release of treated wastewater to the environment (Rodriguez-Dominguez et al. 2020; Sylla 2020; Koottatep et al. 2021).

Due to the commonly held perception that CWs are likely to produce lower quality discharge than the more advanced technologies, studies comparing different OWTS often assign higher environmental impacts to nature-based systems (Garfí et al. 2017; Yang et al. 2018). Meanwhile, the ATUs have a contrary perception due its more advanced process for treatment. One example of these criteria for selection of technologies can be seen in Queensland, Australia, where regulatory bodies focus on the effectiveness of the OWTS such as the ATUs, in removing key pollutants from domestic wastewater. However, this selection criteria does not consider the overall environmental impact of the employed technology in any of the reference documents for approval and commercialization (Government of Queensland 2021).

With the purpose of achieving a better comprehensive understanding of the environmental implications of different wastewater treatment technologies, Life Cycle Assessment (LCA) can be used as a tool to provide useful and essential guidance for identifying the potential environmental implications of selecting one or another OWTS technology (Garfí et al. 2017). Previous LCA studies have shown that some advanced technologies for wastewater treatment may be outperformed in their environmental performance by conventional or low technological alternatives, despite the improved treated wastewater quality to be discharged (Yıldırım & Topkaya 2012; Corominas et al. 2013; De Feo & Ferrara 2017; Diaz-Elsayed et al. 2017; Garfí et al. 2017; Magar & Magar 2017).

Currently, there is a lack of LCA studies which compare the environmental performance of nature-based systems such as the CWs to ATUs. Energy usage of ATUs and the construction materials for CWs have been identified as significant contributors to the environmental performance of the system (De Feo & Ferrara 2017). Currently, LCA studies in active systems such as the ATUs have shown important changes in the environmental performance when the source of electricity generation (coal, gas, oil, solar) is altered for one or another specific location, this is due to the permanent and high demand of electricity for its operation (De Feo & Ferrara 2017; Mirra et al. 2020). Conversely, amendments of the construction material to better target specific pollutants or to improve the quality of the discharge have resulted in inconsistent impacts on the overall environmental performance of the CW systems (Cao et al. 2021; Deng et al. 2021; Diaz-Elsayed et al. 2017; Yang et al. 2018; Shen et al. 2020; Zhou et al. 2020). The availability of new materials that may be used in the substrate can enhance the removal efficiencies of different pollutants in CWs and shrink the gap in the treated wastewater quality currently existing between CWs and ATUs. Concerning this, the use of biochar, a prefabricated material based on carbon, has shown promising results in improving the efficiency of CWs for domestic wastewater treatment, specifically as a substrate material (De Rozari et al. 2015; Deng et al. 2021). However, it has not yet been established if the improved pollutant removal efficiency will enhance the overall environmental performance. Particularly, it is not clear if the environmental benefits gained through the improved wastewater quality overcome the burdens associated with the manufacturing and use of biochar. LCAs which examined the wastewater treatment systems focused mainly on centralized WWTPs (Tangsubkul et al. 2005; Foley et al. 2010a, 2010b). Consequently, there is little information about the environmental implication of the use of biochar and how its use in CWs would compare to systems with high removal capacity such as the commercialized ATUs which limits the ability of decision making about the selection of better OWTSs from an environmental perspective. Therefore, the present study aims to evaluate the performance of CWs and ATUs from an LCA perspective in the Australian context and provide an objective assessment of their environmental impacts.

The LCA was conducted using SimaPro v.8.0.4.30 modelling software, based on the International organization for standardization ISO 14040:2006/AMD 1 :2020 (International Organization for Standardization 2020), which includes the following steps:

In the present study, the diagnosis and comparative analysis was performed among the ATUs and the CWs as OWTS alternatives, having two construction variations for the ATUs (concrete and HDPE), and three for the CWs with different biochar contents in the substrate. The ATUs constructed with reinforced concrete as the main material for the structure are referred to C.ATU. ATUs constructed with HDPE as the main material for the structure are referred to as P.ATU. The three configurations of CW are: CW.BC0, corresponding to CWs with 0% v:v content of biochar in the substrate; CW.BC10: corresponding to CWs with 10% v:v content of biochar in the substrate; and CW.BC20: corresponding to CWs with 20% v:v content of biochar in the substrate.

The scope includes the inputs and outputs from ‘cradle to grave’. The system boundaries are shown in Figure 1. Additionally, the functional unit was set as all the input/outputs associated with 1 m³ of treated wastewater, considering a lifetime of the treatment systems of 20 years.

For this section, the design and inventory data for both technologies, ATUs and CWs, and their construction types are detailed in the Appendix A (Table A1–A5).

The grouping of the inventory data for both technologies was separated in three subgroups: Construction and installation; operation and supervision; and waste management after 20 years of use (Technology wastes).

Treated wastewater was the only ‘product’ of concern being used for the diagnosis and comparison between the two technologies. Therefore, allocation procedure was not required due to the lack of co-functionality of the system (Lopsik 2013; Sabeen et al. 2018).

The ATU inventory detailed in Appendix A (Tables A3–A5) was obtained from the analysis of approved certificates for companies installing OWTPs in Queensland, Australia. This included the ATUs as the core treatment unit sized for 10 people equivalent (P.E), also referred to as a 2 m³/day treatment capacity of domestic wastewater. In Figure 2, the scheme of the processes for the ATUs is presented.

As shown in Figure 2, the core unit of the ATU had four chambers: primary, aeration, settling, and disinfection/pump chamber. After the wastewater is treated, it is sent to the absorption trenches unit for final filtration prior the discharge to the environment, as is required by the local regulations (Government of Queensland 2021). For this purpose, the absorption trenches were sized according to the sizing procedure based on the Australian standards for domestic treatment systems (AS/NZS.1546.1 2008), where the final quality of treated wastewater from the core unit was a determining factor for sizing the filter.

The CWs inventory data, shown in Appendix A (Tables A3–A5), were set according to designs from pilot scale experiments performed in Queensland, Australia (De Rozari et al. 2015,  2018). In brief, the CW system consists of a vertical subsurface flux constructed wetland core treatment unit, with a pre-treatment process performed in a settling/pump chamber. Both process units, as in the design of the ATUs, were sized for 10 P.E.

Once treated, the wastewater was also sent to an absorption trenches unit (Figure 3). The absorption trenches were sized according to the final quality of wastewater following the Australian standards for domestic treatment systems AS/AZN1546.1.

For complementing the LCI for both technologies, 10 assumptions were made as shown in Table 1.

LCIA was performed to evaluate the significance of the environmental impacts calculated from the LCI. From the published literature, the experiences across different regions of the world in LCAs for centralized WWTPs and OWTSs show evident variation in the models used for its calculation. In wastewater treatment, the most used LCIA methods in order of higher to lower frequency are the CML 2,001 baseline, Eco-Indicator 99, Eco-Points 97, EDIP, EPS2000, ReCiPe, Impact 2000, and the e-balance method (Corominas et al. 2013; Renouf et al. 2015; Sabeen et al. 2018).

For the present study, the CML 2001 method was selected due to the higher number of experiences and recommendations from LCA guidelines specifically for Australia. CML 2001 includes the following impact categories: Eutrophication potential (EP); Terrestrial Eco toxicity (TETP); Marine eco toxicity (METP); Freshwater eco toxicity (FAETP); Human toxicity (HTP); Acidification potential (AP); Photochemical oxidation (PO); Ozone depletion potential (ODP); Global warming potential (GWP); Fuel depletion potential (FDP); as well as Abiotic depletion potential (ADP) (Corominas et al. 2013; Renouf et al. 2015).

To compare the OWTS alternatives, the overall environmental performance (OEP) was calculated following the process detailed in LCA experiences from Australia (Lu & El Hanandeh 2019), where one stage normalization was performed (Appendix B).

The procedure for calculating the OEP using the characterization score of each of the 11 impact categories was performed by normalizing the impacts to the Australian context. Each of the category scores was divided by the Australian per capita values for the corresponding impact category (Eq.B1) (Bengtsson & Howard 2010; Lu & El Hanandeh 2019). Normalization results are shown in Table B.1. Then, the 11 normalized scores were summed to obtain a unique score, known as the OEP; lower values of OEP indicate better performance.

In Table 2 the LCIA characterization results and the OEP for the ATUs and CWs are presented. The ATU systems show a higher OEP score than CW systems. Differences in the OEP score among the three CW systems were observed, being higher as the content of biochar increases in the substrate from 10% to 20%. In this case, the lowest OEP score corresponds to the CW with no biochar in its substrate (CW.BC0).

Table 2 shows that the GWP impact associated with the CWs, specifically the configurations that include biochar as complement of substrate (CW.BC10 and CW. BC20), is lower than the ATUs.

The contribution analysis (Figure 4(a)) revealed that the main contributor to the GWP impact category is the technology wastes generated at the end of life of the CWs after 20 years of operation. The disposal and treatment of spent substrate materials, such as sand and gravel used in high mass quantity, is a particular process contributing to the GWP because of the energy consumption. In contrast, Figure 4(a) also shows that the inclusion of 20% of Biochar in the substrate (CW.BC20) has a negative value of GWP during the construction of CWs. The reason for this is the carbon content in biochar that is stored in the soil, which in turn offsets the net greenhouse gas emissions of the technology.

For the ATUs, the highest contribution to GWP comes from the emissions of CO₂-eq generated by the operation of the systems. Precisely, this is the permanent requirement of electricity for powering pumps and blowers for the treatment process. This is consistent with previous studies which pointed out energy consumption during the operation phase as being a hot spot (De Feo & Ferrara 2017; Mirra et al. 2020). This is especially so in Australia because the generation of electricity in Australia is highly dependent on fossil fuels (approximately 88% of fossil fuels: 68% coal and 20% natural gas) (Australian Government 2019).

From Table 2, the two construction types of the ATUs show higher HTP score compared to that obtained from the three types of CWs. Moreover, in Figure 4(b), the operation of the ATUs contributed more impact on HTP, mainly because of the permanent demand of electricity. Conversely, the construction phase is the main contributor to HTP in the case of CW, which is mainly related to material production.

As mentioned in the GWP analysis, coal and natural gas are used in higher proportions for the generation of electricity in Australia. According to the Toxic Release Inventory (TRI), the specific use of coal involves emissions of three principal categories englobing nonvolatile and volatile highly human toxic substances such as, mercury, selenium, beryllium and barium (Rubin 1999; Färe et al. 2010; Ozturk & Dincer 2019). In the case of the CWs, this technology showed a higher contribution to HTP in its construction phase, due to the relatively high requirement of materials such as PVC pipes, which involves emissions of carcinogens such as vinyl chloride and dioxins during its manufacture (EPA 2021).

The CWs (CW.BC0; CW. BC10; CW. BC20) have higher ODP score than the ATUs as shown in Table 2. Figure 4(c) shows that the construction phase of CWs represents the highest contribution for the ODP; further analysis reveals that the transport of construction materials such as sand, gravel, coir peat, concrete, and others, was the input with the higher contribution for this impact category. This result implies that the emissions from the transport of large quantities of materials requires combustion of fossil fuels with the specific release of nitrous oxide, pointed out as one of the gases with evidenced adverse effects for the ozone layer (Chmielewski 1999).

ATUs are much lighter systems and require only the transport of approximately two-thirds of the materials compared to CWs for the construction of the system. Therefore, they result in lower emissions due to transportation and in turn lower ODP impact. However, the electricity used for the operation of this technology represents its highest contribution to the ODP impact of the ATUs. Of particular concern is the use of natural gas for power generation, comprehending the emissions of gases such as chlorofluorocarbons (CFC), halons, and hydrochlorofluorocarbon (HCFC), used as fire suppressants and coolants in the gas pipeline distribution system, being substances with adverse effects for the ozone layer (Purohit et al. 2020).

The EP impact is slightly higher for the ATUs than the CWs, as shown in Table 2. From Figure 4(d), the operation is the main phase that contributes to the eutrophication impact category for ATUs and CWs (approximately 95%). The higher impact found in ATUs is not due to discharge of treated wastewater and its quality, being the same for both technologies and its configurations. In this case, the difference is set as a consequence of NO₂ emission from the use of fossil fuels such as coal and natural gas for the generation of electricity (Zhengfu et al. 2010; Korpela et al. 2017; Purohit et al. 2020). In this context, the emissions from the ATUs are higher due to the continuous demand of electricity.

The ATUs have a higher toxicity impact to aquatic ecosystems (FAETP and METP) than the CWs, as shown in Table 2. Figure 4(e) and 4(f) shows that, in the case of the ATUs, approximately 90% of the ecotoxicity impact can be attributed to the operation phase. The principal cause can be traced to the emissions from mining processes for coal used in power plants, which involves direct and indirect emissions of heavy metals and acid drainage to water bodies which have toxic effects on aquatic ecosystems (Zhengfu et al. 2010). Meanwhile, the main contribution for the FAETP and METP in the case of the CWs is the construction phase, with PVC pipes and the pump devices being the main contributors for both impact categories. The emissions from the production of PVC pipes imply fossil fuels as prime matter, having the background emissions from the mining and refinement processes (Saeki & Emura 2002). Furthermore, the emissions related to the mining and refinement of metals, and its further transformation to components of the pump device, includes emissions of substances with heavy metal content reaching water bodies by direct and indirect pathways (Hu et al. 2014).

For the TETP (Table 2), the CW technology has higher impact scores compared to the ATUs. Particularly, due to coir peat used as a substrate component in the CW.BC0; CW.BC10; and CW.BC20 (Figure 4(g)). Despite being a ‘waste’ product made from the coconut crops, coir peat processing for substrate applications involves the use of large quantities of water and chemicals for removal of substances such as salts, tannins and phenolic compounds (Areapeat 2020). In consequence, it may generate a bulk of liquid solutions with variable toxicity that reaches the soil. Furthermore, this material may be transported long distances to the target location, comprising emissions related to fossil fuel utilization (Stucki et al. 2019).

In the case of the ATUs, the operation is the phase with the highest contribution to the TETP (Figure 4(g)). Further analysis of the two ecotoxicity impact categories in ATUs and CWs point to electricity consumption as a main contributor due to heavy metal release (mercury, vanadium, cadmium, etc.) during the coal extraction and its subsequent use for power generation (Färe et al. 2010; Zhengfu et al. 2010).

From Table 2, the three CWs variants show higher ADP impact than the ATUs. The construction phase had the highest contribution to ADP. Figure 4(h) shows the construction phase of the CWs as being the phase with the highest requirement of abiotic resources. The mineral resources to produce pumping devices, followed by the materials used for the fabrication of the geomembrane are identified as inputs with highest contribution to ADP in the case of CWs. The construction phase was also identified as the main contributor to ADP in the case of ATUs, Figure 4(h). Drilling down into the results revealed that pumps and blowers are the highest contributors to ADP in the case of ATUs.

The CWs with all variants have a higher FDP impact than ATUs, as shown in Table 2. The FDP worsen with the higher inclusion of BC. Further inspection of the results revealed that the main contributor to this surprising outcome is the waste treatment at the end-of-life stage followed by the construction phase (Figure 4(i)).

Disposal of PVC pipes and geomembrane, as well as other inert wastes require treatment prior to landfilling. For example, spent biochar (after being used) may be considered as hazardous waste due to potential biological contamination and inorganic pollutant loads (such as heavy metals adsorbed by biochar), thus requiring special treatment. Furthermore, transformation of large quantities of waste material where the use of fuels is crucial for powering the processes is a significant contributor. Conversely, the operation phase is the main contributor to FDP in the case of ATUs, due to the permanent requirement of electricity.

CWs has higher AP impact than ATUs as shown in Table 2. Upon inspection of Figure 4(j), the construction phase of the CWs, specifically the production of coir peat, and the PVC pipes for the core treatment unit, causes direct and indirect release of SO₂ and NO_x, agents with a strong acidification effect on the environment. In case of the ATUs, the operation phase is the main contributor to AP with electricity consumption being the main culprit.

Compared with the impacts caused by the ATUs, the CWs showed higher PO (Table 2), being the scores attributed to the emissions associated with the technology wastes at the end of its lifetime service of the CWs (Figure 4(k)). Materials such as the PVC and geomembrane, manufactured from fossil fuels, may have important relevance together with the solid waste treatment process fed with fossil fuels, generating along with the primary air pollution, a secondary type of pollution, influenced by the sunlight reaction over gases such as nitrogen oxides and volatile organic compounds, generating other pollutants.

For the technology of the ATUs, as in most of the impact categories of the CML 2001 method, the operation remains as the main contributor to the PO (Figure 4(k)), due to the emissions related to the fossil fuels combustion for power generation in Australia.

The higher contribution in the LCIA for both construction types of ATUs (C:ATU, H.ATU), comes from its operation; specifically, electricity consumption for water pumping and air injection, which contributes at least 50% for most of the impact categories with the exception of the ADP, ODP, and EP. Conversely, the LCIA of the three variants of the CWs (CW.BC0; CW.BC10; CW.BC20), had no evident dependence on the operation of its electronic devices, instead the materials used for the construction weighed more heavily.

As pointed in the contribution analysis for the 11 impact categories, the current generation of electricity in Australia is highly dependent on fossil fuels (Australian Government 2019). The current scenario dubbed 2019 AU, is shown in Table 3 together with two scenarios constructed in line with the objectives for energy supply for Australia; where national policies stipulate that by 2030 the use of renewable energy should increase to 50% of the total supply (Karp 2019).

In detail, the 2019 AU energy scenario represents the business-as-usual case, where sources of renewables account for an approximated 12% (Australian Government 2019). This scenario has already been analyzed and the results shown in the previous comparative OEP section. In this section, the impact of achieving the national goal of 50% renewable energy is investigated. Two pathways to achieve the target are considered: the case of the 2030 AU (50%) scenario assumes an increase of the renewable share to 50% with equal distributed increment among all the renewable sources (hydro energy, ethanol from wood, photovoltaic); and the 2030 S.AU energy scenario, assumes the increase of renewable energy to 50% through increasing the share of photovoltaic electricity.

To evaluate the influence of energy sources on the environmental performance of the ATUs and CWs and compare it with current scenario (2019 AU); the OEP of each OWTS alternative was calculated for the 2030 AU and 2030 S.AU scenarios; the results are presented in Table 4.

Increasing the share of renewables in the electricity grid reduces the environmental impacts of the CWs and the ATU systems. However, it has higher effect on the ATUs. The 2030 AU pathway is likely to return greater environmental benefits than the 2030 S.AU as evident from the percentage reduction of the OEP in Table 4.

A move towards more renewables share can potentially reduce the ATU overall impact by more than 35%. Meanwhile, the change in the OEP of the CWs is less than 5% (Table 4). Despite the significant decrease in the overall impact of the ATUs, the CWs remain as the technology with better environmental performance.

The energy demand is the principal obstacle limiting the ATUs from achieving better environmental performance compared to the CWs. Moreover, operational costs due to electricity consumption is another factor to consider in the selection of technologies. For the present case, the Australian electricity price rates of electricity to residential customers are around $0.34/kWh (Globalpetrolprices 2020). The electricity consumed to treat 1 m³ of wastewater in ATUs is estimated to be 1.053 kWh/m³ which corresponds to an approximate of 0.358 $/m³. This cost is higher than the average prices of energy required for centralized WWTPs (0.46 kWh/m³) (Gu et al. 2017), corresponding to approximately 0.156 $/m³. Conversely, the electricity requirement for the operation of CWs is less than half of that required by the ATUs (0.42 kWh/m³) which translates to 0.143 $/m³.

The percentage of biochar used in the substrate was the principal variant among the three CWs. Due to the high content of carbon in the biochar, its use had a positive influence on balancing the CO₂ emissions from the construction stage. The CW.BC10 and CW.BC20 had lower scores and even negative values (net sequestration) to the GWP when compared with the CWs built with conventional substrate elements (CW.BC0). However, the biochar also influenced adversely and at different magnitude the HTP, ODP, EP, FAETP, METP, TETP, ADP, FDP, AP, and PO impact categories. Most of those adverse effects were identified in the construction and technology waste stages; as a result a slight increase in the OEP scores of the CW.BC10 and CW.BC20 over CWBC0 was noticed as shown in Table 2.

Variation in the production of biochar detailed in Table 5 were assumed to be negligible. However, the fixed carbon in the biochar and its further release in the form of CO₂ to the atmosphere can influence the GWP and the OEP of CWs alternatives; principally, due to the differences in the environmental exposure of the material along its life cycle. Furthermore, as pointed in the LCI assumptions in Table 1, 50% of the carbon in the biochar would be released in the form of CO₂ within 100 years. This means that half of the fixed carbon in biochar remains retained in the soil after its use and disposal of the material. In this context, the sensitivity of the LCIA results in variations in biochar carbon stability was tested for the cases of 100% (all carbon remains locked in the biochar) and 30% assuming 70% of the carbon is released as CO₂ within 100 years.

As shown in Table 6, the reduction of biochar's carbon decay to 0% resulted in the lowest scores for the GWP for the CW.BC10 and CW.BC20. An increment to 70%, had the contrary effect, increasing the GWP. Despite this potential increase, the two CWs with biochar have lower GWP and OEP scores than the CW. BC0. Changes in carbon decay rates do not seem to affect the ranking of the alternative and CWs continue to return better OEP scores than the ATUs.

The CWs in any of the configurations of the study have better overall environmental performance than the ATUs. The use of a reinforced concrete tank vs HDPE tank made little difference to the overall environmental performance of the ATU system. The operation phase was identified as the main contributor to the environmental impacts of the ATU systems; from which the electricity required for the aeration process was the principal contributor for all the impact categories, except for ADP, whose principal contributors was the production of electronic devices (blower and pumps), and the EP, depending on the most in the discharge of treated wastewater.

For the CWs, the construction stage has the highest adverse contribution to its environmental performance; in particular, the PVC pipes, coir peat, geomembrane, and the electronic devices were the inputs with the highest contribution to all impact categories with exception of the EP, which was mainly associated with the discharge of treated wastewater during the operation of the technology.

The data quality analysis revealed that the anticipated increase of renewable energy share in the Australian electric grid will improve the environmental performance of the OWTS and particularly the ATUs. Nevertheless, the ATU technology will continue to deliver lower environmental performance than the CWs.

The carbon sequestration rate of the biochar (varied between 0 and 70%) will have very negligible impact on the overall performance of the CW with its effect only felt on the GWP impact category.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

All relevant data are included in the paper or its Supplementary Information.