«Integrated urban water balance of an eco-building Etude du cycle de l’eau à l’échelle d’un bâtiment écologique Sam Trowsdale, Robyn ...»
Integrated urban water balance of an eco-building
Etude du cycle de l’eau à l’échelle d’un bâtiment écologique
Sam Trowsdale, Robyn Simcock, Jeremy Gabe and Robert Vale
Manaaki Whenua Landcare Research
Private Bag 92 170, Auckland 1142, New Zealand
Le bâtiment écologique de Landcare Research à Auckland, Nouvelle-Zélande a été
conçu de façon à minimiser l’utilisation d’eau du service publique, en favorisant l’utilisation des eaux de pluie et aussi de minimiser la production des eaux usées, afin de réduire l’impact sur les eaux naturelles. Cela fut un défi puisque les laboratoires et les serres de Landcare Research utilise une quantite importante d’eau. L’utilisation d’eau du service publique a été minimale grâce au recueillement et la réutilisation des eaux de pluie ainsi que la réduction de la consommation d’eau en utilisant des toilettes sèches. De plus, les toilettes sèches produisent un produit utilisable à partir des déchets et réduit ainsi la production de déchets non traités. La récupération des eaux de pluie à partir du toit réduit l’impacte de l’écoulement des eaux sur le bâtiment. Les autres pertes d’eaux sont les eaux de pluie qui s’écoulent sur le parking. Le parking fait parti du traitement des eaux de pluie par une zone de rétention et un jardin pluvial qui permettent la réduction du volume totale d’eaux et des crus qui entrent dans le système du cycle de l’eau. Les résultats de l’étude du cycle de l’eau bu bâtiment écologique sont présentés avec une discussion sur les données du système, ainsi que le besoin requit par le système et enfin les pertes d’eaux du au système.
ABSTRACTThe Landcare Research sustainable commercial building in Auckland, New Zealand, was designed to have a small demand for mains water and minimal discharge of stormwater and sewage, to reduce its impact on natural waters. This was a challenge given that the core business operations of Landcare Research, such as research laboratories and experimental glasshouses, require a large volume of water.
Minimising the use of mains water was achieved by harvesting and reusing stormwater as well as reducing demand, primarily by using composting toilets. In line with the principles of sustainability, the composting toilets produced a useable product from a waste and reduced untreated discharge from the building. Harvesting the stormwater from the roof reduced the impact of runoff from the building. The other key discharge was stormwater running off a carpark. The carpark forms part of a stormwater treatment train including a bioretention strip and raingarden which reduced the total volume and peak flow of water entering the stormwater network.
Monitoring results are presented as an integrated urban water balancewith discussion about the inputs to the system, followed by the demands placed on the system and then discharge from the system.
KEYWORDS Bioretention, Compost toilets, Raingarden, Urban water, Water consumption
1 INTRODUCTION Landcare Research is a New Zealand Crown Research Institute primarily concerned with the terrestrial environment and sustainable development. When Landcare Research chose to relocate to a brownfield site on the University of Auckland’s Tamaki Campus, staff wanted to demonstrate how it might be possible to build a new building that was more environmentally friendly than usual. This was made complex by the requirement for research laboratories, glasshouses and housing for the six million or so specimens in the national collections of insects and fungi.
The Tamaki building has been widely acknowledged as a leading example of a sustainable building. In May 2005 it won the Energy Efficiency and Conservation Authority’s Energy-Wise Commercial Building award and it also received an “environmental hero” Green Ribbon Award from the Minister for the Environment.
The design considered four waters: mains-, storm-, waste- and natural-. The Tamaki building aims to reduce the demand for mains water and discharges of stormwater and sewage compared to a conventional building, to minimise its impact on natural waters. The building was also designed to be energy efficient and make use of renewable materials and finishes that had a 100-year life.
The design and operation of the building’s integrated urban water management systems are presented here as a water balance. Discussion focuses on the input of water, demands for water, and output of water from the system. The building is being monitored to see if its environmental performance measures up to the design intentions. In this paper, the measured operational performance of the building is provided to show what has been achieved and allow comparison of its performance with national and international practice.
2 INPUTS OF WATER The inputs include rainfall, harvested rainfall and potable water supplied by the city’s water mains. Another input is human excrement, but in this building most excrement is composted and so is not combined with water.
2.1 Rainfall The location of the building in Auckland City, at 36°S in the south-west Pacific, means that summers are warm with high humidity and winters are mild and damp. Rainfall is measured onsite using several 0,2-mm tipping bucket gauges. Data records are complete from 19 January 2005 to date. These are supported by climatic data collected at a weather station just 500 m from the building since December 1992. The city experiences most rainfall in April to September (90–130 mm/month) and moderate moisture deficits (30–90 mm/month) from November through March. The long-term average rainfall is 1200 mm per year (ARC 1999).
2.2 Water supply The building is connected to the mains water supply. In addition, runoff from the 1526-m2 roof area drains through a syphonic drainage system to three 25 000-L rainwater storage tanks connected in series. The first two tanks are connected at their base and are in hydraulic equilibrium. The third tank collects overflow. Water is pumped from the two tanks in equilibrium to two smaller header tanks, using electricity supplied by a 400-W wind turbine. A mains backup is available for prolonged dry spells and for fire fighting.
4 NOVATECH 2007 SESSION 1.1 3 DEMAND FOR WATER The design goal was to improve upon the demand for potable mains water of a typical office building. This was a challenge given that the core business operations of Landcare Research, such as research laboratories and experimental glasshouses, require a large volume of water. Rainwater harvesting, low-flow fittings and composting toilets contribute towards the goal.
3.1 Mains water The buildings potable water is supplied by the city main supply. The daily consumption of mains water was manually measured, for working days between 19 July 2005 and 5 April 2006. Then in May 2006, a building management system became operational that automatically records the hourly rates of total mains water consumption and volume of mains water used to top up rainwater tanks.
3.1.1 Hand basins, showers and kitchen use As a potential source of potable water, hand basins, showers and kitchen appliances were required to be supplied by mains water but have low-volume, water-saving fittings to prevent unnecessary wastage.
3.1.2 Purified water for laboratory use The building also draws on the city water supply for water treated by reverse osmosis (RO), required for laboratory work such as washing glassware and making chemical solutions. The RO treatment process is a large consumer of water because only an estimated 30% (maximum) of the water used in the process is purified. The system has been configured so that the reject water can be sent to the rainwater tanks.
3.2 Rainwater harvesting and reuse Stored rainwater is delivered to the building from two header tanks. The separate header tanks distribute water to the glasshouses or flush toilets/ urinals. The building management system records the hourly demand for stored rainwater from each.
3.2.1 Toilet system The first and second floors have seven individual toilet pedestals connected to two Clivus Multrum composting bins. The bins are gravity-fed from the male and female toilets respectively and use no water for flushing. Composting toilets were not suitable for use on the ground floor because the depth of hole needed to house the compost bins would have been below the flood level and very expensive to dig as the building is situated on basalt rock. On the ground floor there are five dual-flush toilets. Manual flush urinals, using rainwater, were installed in the male facilities on all three floors.
3.3 Monitoring results 3.3.1 Ratio of mains water to harvested water The period of most complete data, 18 August 2006 to 6 November 2006, was used to calculate the ratio of mains to harvested water use. Between 6 p.m. on 7 September 2006 and 1 p.m. on 8 September 2006 a toilet malfunction resulted in leakage of 14,1 m3 of stored rainwater. This was omitted from the calculation.
A total of 227,7 m3 of water was consumed during the period analysed. This was made up of 125,1 m3 drawn from the city water main, 32,4 m3 of rainwater supplied to the toilet header tank, and 70,2 m3 of rainwater to the glasshouses. No mains water was used to top up the rain tanks. Thus, the total consumption was made up of 45% reused rainwater and 55% mains supply.
NOVATECH 2007 5 SESSION 1.1 3.3.2 Cost savings by reusing rainwater Landcare Research is charged NZ$4 per cubic metre of water it draws from the water main, which includes the cost of subsequent wastewater treatment. Landcare Research paid NZ$500 for mains water supplied between 18 August and 6 November
2006. The reuse of rain water saved NZ$410 during the same period. Cost savings would have been greater if the building had been designed with more conventional appliances.
3.3.3 Mains water consumption The total annual consumption of mains water equalled 879 m3, for the year ending 1 November 2006. The number of full-time-equivalent (FTE) employees during this period was 80,25, making the total mains water used 11,0 m3 per FTE per year. The total floor area is 4766 m2, resulting in a mains water use of 0,18 m3/m2 per year.
Many design goals are reported for water use in commercial buildings, but few actual measurements. Measurements have been used by The Department of Sustainability and Environment in Victoria, Australia, to determine a benchmark figure of 50 L per FTE per day (Kyle Garland, pers. comm. 20 March 2003). Based on a 250-day operational year (5 days a week, 50 weeks a year) this equates to 12,5 m3 per FTE per year. The organisation manages its water consumption, so the figure contains attempts at conservation. The consolidated environmental data for 2001 presented in the 2002 Corporate Social Responsibility Report of the worldwide operations of the CGNU Insurance Group (CGNU 2002) provides consumption measurements of 9,2 m3 of water per FTE per operational year throughout their buildings. Twort et al.
(1994) suggest a figure of 65 L per day per employee for water consumption in commercial and institutional establishments. Normalising Twort et al.'s figure for the operational year gives a total consumption of 16,3 m3 per FTE per year.
Despite the large water demand in laboratories and glasshouses, the consumption of mains water in the Landcare Research building (11,0 m3 per FTE per year) is similar to or better than measured targets for commercial buildings, including those that have adopted some environmentally sustainable practices. The composting toilets and rainwater harvesting have clearly reduced the demand for mains water.
4 OUTPUT OF WATER The stormwater and waste-water discharges are the most important in terms of the urban water balance and the potential impact of the building.
4.1 Stormwater 4.1.1 Runoff from the roof Much of the roof runoff is harvested, so stormwater discharge into the municipal stormwater network is limited. The overflow tank is used to irrigate the site’s gardens via a gravity-fed drip irrigator. The water is presumably evaporated, transpired by plants, or recharges groundwater. To provide detention volume during the rainy months, when there is a surplus of runoff water and the gardens do not require much irrigation, the tank is slowly drained to a raingarden.
4.1.2 Stormwater treatment train A building’s carpark is often a major source of stormwater and so the carpark was designed as the first step in a stormwater treatment train. The carpark drains to a bioretention strip, which in turn drains to a raingarden. The water that passes through the raingarden enters the local network of stormwater pipes, which delivers it offsite to a constructed stormwater pond and then to a local stream. The onsite devices are described below.
6 NOVATECH 2007 SESSION 1.1 4.1.3 Design of the carpark The 761-m2 carpark was designed as a permeable area and constructed of compacted gravels. The gravel provided some interconnected pore space but had a “slow” (McQueen 1983) infiltration rate. Mean saturated hydraulic conductivity, measured over 48 h using the twin-ring method, and following 48 h of pre-wetting, was 1,5 mm/h, with one-third of the test sites having nil infiltration and no site more than 5 mm/h infiltration. The durability of the carpark surface and turbidity of runoff were below expectations. This type of surface is not recommended.
In an effort to control pollution at source the 38-space carpark was designed to accommodate about half the number of cars as staff. Encouraging staff not to drive to work was supported by locating the building c. 900 m from a railway station and bus depot and setting up an internal website to assist with carpooling.
4.1.4 Design of the bioretention strip The carpark drains to an adjoining bioretention strip that is 1,5 m wide and runs the length (30 m) of the main building. It exceeds sizing guidelines (USEPA 1999), being about 8% of the catchment’s area. The bioretention strip is gently graded (1–3 %) and runoff is designed to pond in an extended detention depth of c. 50 mm.