Alternative water sources in New Zealand's commercial buildings

Population growth, urbanisation and climate change are having an increasing impact on our cities and the buildings we design for them (Gross et al. 2015). Smart buildings of the future need to be efficient and sustainable. Water, and the way we use water in our buildings and cities, will increasingly be part of that equation. The application of systems, such as rainwater harvesting and greywater recycling, has implications on the wider water network. Reducing the volume of water that is supplied by the urban reticulated network reduces the cost of treatment to potable water standard and relieves the infrastructure required to transport this water from source (Gross et al. 2015). In addition, separation of greywater reduces the volume of outgoing water that requires treatment before eventual discharge to the environment.

Rainwater harvesting and greywater recycling are not new concepts, with evidence of such practices as far back as Minoan Crete – ca. 3200–1100 BC (Angelakis 2016). Global practices of rainwater harvesting and greywater recycling are varied, ranging from mandated or incentivised by government, to discouraged or banned (Campisano et al. 2017). However, they are not extensively practiced in developed countries if water shortages are not an ongoing issue. New Zealand, for example, receives on average 600–1,600 mm rain per year, but some areas may experience up to 10,000 mm per year, with large seasonal variations (NIWA 2018). The shift in rainfall patterns as a response to climate change may also have consequences for stormwater systems, which may be exceeded more frequently due to heavy rainfall events and could lead to surface flooding (Ministry for the Environment 2016). This has been an ongoing issue in Korea, which experiences continual cycles of flooding and drought. In that instance, rainwater harvesting and storage has been proven to be a sustainable solution, by their Star City Project in Seoul (Han & Mun 2011). Nonetheless, there is an undeniable need for greater water resilience in our cities. This was demonstrated clearly during the 2011 earthquakes in Christchurch, New Zealand, which left 80% of the city's water and sewage network severely damaged (Clifton 2011). This was compounded by the fact that almost 100% of New Zealand's commercial buildings are totally dependent on a water reticulation network, where treated potable water is used for hygiene, conditioning and other purposes – including irrigation and toilet flushing.

At present, there are no New Zealand-specific guidelines to assist and ensure effective delivery solutions for alternative water supplies. With increasing population and rates of urbanisation, compounded by climate change and New Zealand's unique geological landscape, these technologies could form part of the solution to the multi-faceted issue of future water supply and demand in New Zealand.

This research aims to provide the necessary information to increase industry knowledge on rainwater harvesting and greywater recycling systems. This includes the drivers and barriers, the operational and financial feasibility of systems in operation and the impacts on the reticulated water network.

To create a holistic overview of the rainwater and greywater system feasibilities, a multi-disciplinary team explored three research areas.

This study was undertaken to understand the perceived and actual drivers and barriers to the uptake of rainwater harvesting and greywater recycling systems in the New Zealand context. The following methods were used to collect relevant information:

• Two electronic surveys examined the perceptions from a wide range of individuals across New Zealand in 2014 (71 respondents) and then again in 2016 (265 respondents).

• A review of published literature and legislation was undertaken.

• Informal discussions with building-related and water-related industry professionals, including informal workshops, were held.

There are an estimated 41,154 commercial and industrial buildings in New Zealand (Amitrano et al. 2014). Approximately 370 of these buildings have a rainwater harvesting system, and at least one has a greywater recycling system in operation. Eight of these commercial buildings were assessed for their performance and feasibility (Table 1). Despite ranging in building use, size and location (Table 1), these buildings all used the collected rainwater and greywater for the flushing of toilets and urinals.

All eight sites were visited, and meetings were held with the building manager. As well as information gathered from these meetings, the building documentation, plans, costs and maintenance regimes were reviewed. From all of this information, a water audit was carried out to create a full water balance of the building.

Four of the case-study buildings had their rainwater quality tested monthly over 1 year. All samples were taken prior to any treatment and analysed for the pathogenic microorganisms Salmonella spp. and Campylobacter spp., and the pathogenic indicator organism Escherichia coli. Samples were also analysed for pH, total suspended solids (TSS), ammonia, nitrate, nitrite, total N, total P, copper, zinc, aluminium, boron, calcium, cadmium, iron, potassium, magnesium, manganese, sodium, nickel, lead and sulphur. All data were compared with the Drinking Water Standards for New Zealand (Ministry of Health 2008).

For one building, greywater was sampled monthly before and after treatment and tested for the same parameters as the rainwater samples. In addition, quarterly pre-treatment greywater samples were tested for Giardia spp., Cryptosporidium spp. and culturable Adenovirus.

Using the data collected above, the overall performance, feasibility and design lessons for each building were determined.

The volumetric impacts of the eight buildings on the water networks were calculated. This involved two series of analysis: current and future impacts to the water networks over 50 years. These are based on a range of potential uptake scenarios. The following data inputs were examined:

• Metered mains water.

• Harvested rainwater.

• Recycled greywater.

• Current and forecast regional water demand.

• Current and estimated non-residential building stock, based on building consents.

An industry advisory panel was established as a forum for advice on the direction of the research and to provide reality checks.

Internationally, there have been numerous studies examining the barriers and drivers for rainwater harvesting. A leading study summarised how the international perspective divides the barriers into four themes – institutional, economic, technological and educational (Ward 2010). Many of the specific subthemes from this international study reflected the New Zealand findings (shown in the grey cells in Table 2). The overarching issue was the knowledge gap of rainwater harvesting systems, especially for non-residential systems.

From the two electronic surveys (2014 and 2016; Figures 1 and 2 respectively), cost, education and storage were perceived as the biggest barriers for rainwater, while education and cost were the two biggest barriers to greywater recycling (Bint & Jaques 2017).

Anecdotally, informal feedback found an industry expectation for a maximum pay-back period of 3–5 years before management will approve inclusion in building design, which is a very tight timeframe.

In contrast, the biggest incentives or drivers for installing rainwater harvesting and/or greywater recycling systems were cost savings and environmental responsibility (Figures 3 and 4). A secondary (but important) reason for installing rainwater harvesting and/or greywater recycling systems was for resilience – i.e. to ensure that a building's function was maintained during and after a natural disaster in New Zealand.

Respondents' primary concerns with rainwater harvesting or greywater recycling systems were water quality, health and waterborne disease. For greywater quality, respondent concerns were health, general quality, cross-contamination with potable water, cleanliness of the system and society's perception of ‘dirtiness’.

Therefore, while cost savings and environmental reasons were the main drivers for installation, the underlying lack of knowledge and uncertainty in regard to health implications were perceived to outweigh the potential benefits. There was found to be an underlying resistance to the systems' implementation as a result.

All case-study buildings were better than average in terms of water efficiency – as measured by the total building water use intensities. Total water use in the buildings ranged between 0.13 and 1.13 m³/m².yr (average ranges, by region, 0.76–1.03 m³/m².yr; Table 3). This is consistently lower than the average range of New Zealand benchmarks (Bint 2012), indicating that the buildings were already designed with water efficiency in mind.

Further to Table 3, the monitoring of eight commercial buildings found rainwater use in the range 45–1,147 m³ in summer, 22–1,039 m³ in winter and a total annual water use of 307–23,525 m³. The average proportion of total water use that was comprised of non-potable, non-contact end-uses (i.e. toilets and urinals) was 23%. This indicates a potential saving of 23% of total water from the water network. This also equates to a financial saving for both the building owner and the water service provider (Bint 2017).

The feasibility of each rainwater harvesting and/or greywater recycling system to supply the required non-potable demand throughout the year is assessed in Figure 5. It is presented as a percentage of the annual demand that is either met in full, in part or not at all (none).

However, it is recognised that only two of the case-study buildings (C1, W1) were using the water systems to their full advantage. The others have significant underutilised potential.

Rainwater supplied between 9% and 62% of total water demand or an average of 89% of the case-study buildings' non-potable demand. A more consistent year-round average was found from the greywater system. However, the system was not being utilised to its full potential by only supplying one toilet block, which is 4% of the total water demand or 10% of the building's non-potable demand.

Much of the water sourced from the rainwater systems occurred between March and November (autumn, winter, spring), with lower supply during the drier, summer months. Figure 6 uses the annualised monthly average rainwater and/or greywater use for each case-study building, compared with actual monthly use. This shows, for example, building A2 was using less than the average during January, February and November. Overall, rainwater systems are less reliable during the drier, summer months in New Zealand, thus creating a greater reliance on the water supply network during this time.

The economic feasibility is almost entirely dependent on volumetric wastewater tariffs. In Auckland, potable water and wastewater are charged volumetrically. Elsewhere, a range of charging mechanisms apply, including volumetric or bulk allocation charging for potable water, to wastewater charged as a percentage of a building's capital value as part of the council rates.

Despite poor financial pay-back periods in their own regions (3.3–63.8 yr pay-back), applying Auckland-based tariffs to all case-study buildings meant the systems became more financially feasible (1.6–20.9 yr pay-back; Table 4). The charging mechanisms (i.e. volumetric wastewater tariffs) outside of the Auckland region are not providing the financial incentives to lower the use of water, or to become less reliant on the mains reticulated networks. Therefore, in addition to the drivers identified in this research, the observed uptake rate is also influenced by water service provider charging methods, but lack of education, guidance and standards are creating barriers at all levels.

Rainwater chemical results were generally lower than the current drinking water maximum acceptable values or guideline values for New Zealand. All samples were also well below the modified values, which were recalculated to reflect the lower volume of toilet flush water expected to be used compared with drinking water. With regard to microbiological results, of the 18 samples with E. coli detected, 11 were from the same building. This building did not have an enclosed tank, which is the only distinction between the other buildings sampled. These chemical and microbial results are reasonably consistent with previous New Zealand studies (Siggins & Cressey 2017) and show that, with correct design and maintenance, a high level of water quality can be maintained before any treatment or filtration.

Of the chemicals detected in the greywater samples, only aluminium exceeded the guideline value in a single sample. However, it should be noted that the guideline values are in New Zealand drinking water standards for aesthetic water qualities only. For microbial analysis, E. coli was found in three pre-treatment samples at low levels. No E. coli or other microbial detections occurred post-treatment. Overall, the quality of greywater in this single case-study building is better than expected.

The current impacts of the case-study buildings were projected forward 50 years to 2066 across the four New Zealand regions in which the buildings are located (Garnett & Bint 2017). To gain an indication of the potential volumetric savings to the networks, a range of building uptake and water availability scenarios were assessed.

Uptake scenarios were based on current and projected building consent figures against a recent building stock database. This applies to both new build and retrofit uptakes ranging low, medium and high. Furthermore, rainwater and greywater are used to supply non-potable, non-contact water demand (Figure 5), which is found to be 23% of total water demand across all case-study buildings. This is used as an optimistic supply scenario (Table 5) where non-potable demand acts as the volume achieved. The observed scenario applies the actual savings demonstrated in the case-study buildings.

Based on the limited number of case-study buildings, Figure 7 shows the highest and lowest projected volume of mains water that could be alleviated from the water network, assuming an average total water use (as per the case-study buildings) and an optimistic supply of 23% non-potable usage.

Using an optimistic scenario, the savings forecast across all four regions ranged as follows:

• Low uptake: 114,853 m³/yr in Bay of Plenty to 612,441 m³/yr in Auckland.

• Medium uptake: 246,164 m³/yr in Bay of Plenty to 1.6 million m³/yr in Canterbury.

• High uptake: 607,180 m³/yr in Bay of Plenty to 3.7 million m³/yr in Canterbury.

When the buildings were aggregated per region, the volume of non-potable demand that is able to be supplied varies (refer Table 5). Accordingly, a second supply scenario was used to project future savings to the water network, based on observed scenarios at a regional level.

Using an observed scenario per region, forecast savings across all four regions ranged as follows:

• Low uptake: 199,743 m³/yr in Bay of Plenty to 925,170 m³/yr in Canterbury.

• Medium uptake: 428,112 m³/yr in Bay of Plenty to 2.6 million m³/yr in Canterbury.

• High uptake: 1 million m³/yr in Bay of Plenty to 6.2 million m³/yr in Canterbury.

When compared with the regional water demand forecasts, the volumes of water that are projected to be saved are comparatively low (Figure 7). By increasing the water end-uses for non-potable water, an increase in the potential water savings to the networks could be achieved. The results of this research show that, whilst there is a reduction in network demand, it does not significantly reduce the amount of supply required to meet future demands.

Projecting the result from the 2015–16 case-study data with specific uptake scenarios shows the potential volumetric savings possible for the water network in the future based on estimated new builds and uptake scenarios alongside two supply scenarios. The addition of more commercial buildings data to this analysis would help to build a more comprehensive picture of potential future impacts of using rainwater harvesting and greywater recycling technologies across New Zealand.

Perhaps a large and somewhat overlooked advantage of using rainwater harvesting and/or greywater recycling systems is the capacity to maintain supply given conditions in which the reticulated network is constrained. Flood events can overwhelm the network and cause reduced water availability due to water treatment facilities becoming overwhelmed by increased flows and sediment content. Whilst rainwater and/or greywater would be used for non-potable uses only, reducing the demand for treated water during these events would be beneficial – as well as overflow storage.

Moreover, a common consequence of urban development is increased peak discharges and flood frequencies. Urban streams have been found to rise more rapidly than their rural counterparts (Konrad 2003). In part, this is due to the modification of the landscape required for urban settlement. Removing vegetation from the landscape and replacing it with buildings and roads reduces the imperviousness of the surface and its attenuation capability. These surfaces store little water and accelerate run-off (Konrad 2003). A solution to this increase in run-off is increasing the capacity to detain rainfall before it becomes surface run-off. The wide-scale adoption of rainwater tanks could help attenuate stormwater run-off contributing to large flood events.

The two self-selecting surveys found the following key areas of research, which have been explored through further building and water network assessments:

• Cost – one of the biggest incentives and also one of the barriers.

• Storage – one of the barriers.

• Water quality – the primary concern with the use of rainwater and greywater.

• Water quantity – resilience was one of the incentives as was cost.

• Education – one of the barriers.

Whilst cost savings were one of the largest drivers for the installation of rainwater harvesting and greywater recycling systems, the cost of installation was also found to be one of the greatest barriers to uptake. The financial incentive to install rainwater and greywater systems is almost entirely dependent on volumetric water and wastewater tariffs. Indeed, several studies have demonstrated that rainwater harvesting systems are not financially viable (Rahman et al. 2011; Roebuck et al. 2011; Campisano et al. 2017). Nonetheless, the eight case-study buildings showed that, despite having poor financial pay-back periods in their own regions, using Auckland's wastewater tariff meant the systems became financially feasible. The Auckland region of New Zealand is the only water service provider to volumetrically charge for wastewater. It was determined that, by including wastewater charges in general rates, the incentive to conserve water was reduced (Bint 2012). Thus, the tariff structures themselves act as a barrier to the uptake of alternative water sources. This has also been reflected in a study by Wang & Zimmerman (2015), which demonstrated a weak correlation between per capita precipitation (i.e. water demand) and water rates.

Furthermore, throughout the site visits and monitoring of the eight case-study buildings, only two were found to be using their rainwater and/or greywater system efficiently, thus potential cost savings were not being maximised. While the findings of this research can give an indication of the potential financial savings of installing alternative water sources in a range of commercial buildings, it should be noted that there is room to improve the design and operational performance of the systems.

In terms of greywater quality, the specific recurring issues were health, water quality, cross-contamination with potable water, cleanliness of the system and society's perception of ‘dirtiness’. In response to these concerns, the water quality of rainwater harvesting and greywater recycling systems was tested. The water quality study concluded that there is likely to be little or no potential human health risk surrounding the use of rainwater or greywater for toilet and urinal flushing. These results are in line with other studies that investigated rainwater (Cook et al. 2014) and greywater (Eriksson et al. 2002) quality. However, noting that only five buildings formed this part of the study means it cannot be considered representative.

Six of the case-study buildings underutilised their rainwater harvesting and/or greywater recycling potential, therefore falling short of their potential. An analysis of the eight case-study buildings found that, on average, 23% of a building's total water use was for non-potable, non-contact purposes. Therefore, under an optimistic supply scenario, rainwater and/or greywater sources could account for up to a 23% reduction in potable water from the reticulated network by using rainwater harvesting and/or greywater recycling. However, it was found that, in reality, not all regions were supplying 23% of the total water use with alternatively sourced water. Therefore, under an observed scenario, the supply rates were found to vary from 4% through to 38% of total water use. The greatest potential savings can be seen for the Canterbury region under all building uptake scenarios.

As the case-study buildings were limited in frequency and regional range, their volumetric impact on the potable water network is considered minor. However, with increased uptake and in combination with residential rainwater harvesting and/or greywater recycling systems, the capacity for these systems to reduce the network demand would increase.

The value of alternative water sources will only increase in future years as our population and rates of urbanisation increase. Expected increases in population create an expected supply deficit of, for example, 148 ML/day in Auckland alone (Klein et al. 2015). At present, most regions of New Zealand have at least one river and/or aquifer that is either fully or over-allocated or is likely to be so in the next 5 years (New Zealand Business Council for Sustainable Development 2008). The need for further research into alternative water sources will become more prevalent in future years.

The study found the following:

• There is likely to be little or no potential for risk to human health by using rainwater or greywater for toilet and urinal flushing.

• An appropriate tariff system must be in place for cost to be a motivating factor in the decision to implement rainwater harvesting or greywater reuse.

• Knowledge and training of building managers is essential for the best operation of sustainable water management systems.

• Increasing populations, urbanisation and climate change will push the rising demand for alternative water management options.