20 December 2011–This is an edited version of their paper, presented at the World Sustainable Building Conference 2011 in Helsinki.
Significant effort, capital and time are being invested by western countries to address climate change impacts by constructing more sustainable buildings.
Building a sustainably designed dwelling may assist in reducing water and energy consumption through passive measures, but consumption efficiencies may also be enhanced or reduced by the behaviour of the occupants within the dwelling and to date this has not been quantified.
Here we describe the method and interim results of research aimed at improving our understanding of whether sustainably designed homes truly achieve lower levels of impact on the environment and also to what extent occupant attitudes play a role in any improvement.
Key utility consumption data have been measured using real time monitoring systems embedded with the SD homes at The Ecovillage at Currumbin, Queensland.
Environmental attitude questionnaires have been used to better understand residents’ behaviours.
For this study the primary household resource utilities of energy and water consumption were studied at two comparable estates, the Ecovillage and The Observatory Estate, Queensland.
The impact of major variables such as (sustainable) house design, size of house, hours occupied and the number of occupants were analysed from consumption data and environmental attitude questionnaire results.
The analysis of the data suggests the sustainable design of a home contributes significantly to reducing its energy use.
In the case of the Ecovillage (which contains homes that must meet comprehensive sustainable design criteria) energy use (on average) was typically 58 per cent less that of the comparable Observatory estate.
When considering the power generated by homes (most extensively at the Ecovillage), the ecovillage consumed just 5.7kWh of grid power per day on average (76per cent lower than the contemporary estate which consumed energy at rates (23 kWh/hh/pd).
In terms of water, both estates used a similar amount of water and this was not found to correlate with other data collected. The number of people in each home did correlate positively with the amount of energy consumed, as expected.
The size of a home did correlate with energy use but when included in a model with other variables more strongly associated with energy use, its size did not have a statistically significant independent association with energy use.
The results from the use of a new ecological paradigm environmental attitude questionnaire were not able to demonstrate that environment based attitudes and behaviours contributed significantly to lower energy use, when other house design factors had already been taken into account.
However people interested in preserving the environment tended to use less energy when considering just these two variables.
The sustainable design of a home was the largest independent contributor to reduced energy use among the variables included in the study.
Sustainable design predicted 34 per cent of the variance in energy consumption per person whereas the two environmental attitude variables predicted only 14 per cent and 20 per cent of the variability.
This suggests that the (sustainable) design of a house is twice as likely to reduce a home’s energy consumption compared to the influence of environmental attitudes.
An equivalent sample of contemporary homes from the same area provides high quality comparative data. Results reveal the sustainably designed homes use only 25 per cent (5.7 kilowatt hours per day) of the net energy that the contemporary homes use on average.
Higher levels of attitudes favourable to environmental conservation also correlate closely with (lower) energy use but were not found to be significant when analysed with other stronger variables.
Building a sustainable residential home may assist in reducing water and energy consumption through passive measures, however consumption efficiencies may also be impacted by the behaviour of the occupants within the dwelling and this requires quantifying to focus policy and funding on the areas that will make the most difference.
First, it was hypothesised that sustainably designed residential buildings would demonstrate significantly lower utility usage compared to contemporary homes.
Second, it was predicted that residents with stronger environmental consciousness (pro-environmental attitudes) would demonstrate lower utility use.
In combination, it was hypothesised that people with stronger environmental attitudes and living in sustainably designed homes would achieve the greatest relative reduction in utility usage of all groups.
To assess the technical (sustainably designed) and non-technical (behaviour) aspects of influences on utility consumption, two types of data were collected.
First, key utility consumption data (including water, gas and electricity) were measured using real time monitoring systems embedded in 35 sustainably designed homes within the sustainably designed (Ecovillage) estate.
Second, data on occupants’ environmental attitudes towards environmental conservation and resource usage were collected. It was postulated that strong values favouring environmental conservation would impact on behaviour and resource usage.
The Ecovillage was selected as the primary case study due to the high probability that occupants would have environmental attitudes aligned to conservation.
A control group of 36 equivalent sample contemporary homes from the same area was also studied as a well matched comparison group in the same climate zone.
The sustainable homes were developed using strict architectural and landscaping design codes which assisted the Ecovillage to win the FIABCI Prix d’Excellence Award in 2008 and 30 other national awards.
Airconditioning not permitted in Ecovillage development codes
A variety of tools can be used to assess the performance of a building. In the case of the Ecovillage, owners were required to meet comprehensive architectural and landscaping codes which banned air conditioning (despite it being located in a sub-tropical climate region) and promoted solar passive design techniques making them theoretically much more sustainable compared to contemporary home designs.
It is suggested that strong conservation values in residents toward environmental conservation will impact on behaviour and in turn resource usage (Bonaiuto M and Bonnes M 2010).
The behaviour of the occupants can be quantified using relevant measures such as actions of the residents to reduce power and water usage.
Examples include turning appliances off when not in use or putting on more clothes before turning heaters on.
It is highly probable that the environmental attitude of a person influences their behaviour in their home environment.
In 2006 our selected EA questionnaire demonstrated that preservation predicted self-reported ecological behaviour, whereas utilisation predicted attitudes toward economic liberalism (Milfont T. L. and Duckitt J. 2006); and meta-analysis has previously demonstrated a significant correlation between environmental attitudes and behaviour (Hines J, Hungerforda H et al. 1987).
This study and its partners have access to several factors that make it unique:
- The Ecovillage has ensured a hot spot of environmentally friendly homes to be built in the same area. This allows climate and other potential differences to be held constant.
- There are contemporary estates of similar house age in the same area and climate zone, enabling various direct comparisons. In this case the comparison homes are in an estate in the same local climate region, 7km north, called the Observatory by the Stockland Development Company.
- Real time data from Ecovillage homes supplements the metre data from the homes and allows extraneous data to be identified and potentially removed from the study to enhance its accuracy. For example if a gas metre had a leak, it can be removed from the study until the day it was fixed, whereas billing data from utility providers does not allow for such interrogation.
- Collection of a range of demographic data allows conversion of the home data to square metres or per person or make allowances for house occupancy (or lack thereof) over a period.
- The dual study permitted utility data to be correlated to surveys of attitude, from the same population unlike other studies that have typically focused only on consumption data or attitude/environmental happiness or well-being.
For homes that have a real time monitoring system (EcoVision), the more precise consumption data was used. For homes without an EcoVision system, energy and water data extracted from paper utility bills were used.
As part of the wider project, occupants also completed optional questionnaires to evaluate their environmental consciousness, beliefs and behaviours and perceived satisfaction with their neighbourhood. More than 15 pieces of demographic data were collected from each participant, with over 185 other responses gathered from each EA questionnaire.
Variables were inspected to check whether they met the assumptions for parametric analyses. One home within each group had an extremely high value for energy use
In terms of energy consumption the Ecovillage homes used 58 per cent less energy than the comparison contemporary estate known as the Observatory.
Ecovillage homes use reticulated gas for cooking and hot water boosting and this was included in the total energy calculations along with standard grid electricity usage data.
The Ecovillage used an average of 10 kilowatt hours per household per day compared to 23 kWh/hh/pd for the Observatory homes (and state average of 20 kWh/hh/pd (Queensland Government 2011). The scatter plot in Figure 2 displays the correlation between SD and energy.
On average the number of occupants per dwelling was 24 per cent higher in the Observatory (3.4 people per night) compared to the Ecovillage (2.6).
In terms of size (internal rated floor area, which excludes garages), the Observatory dwellings were significantly larger (329 sq m by 42 per cent than the Ecovillage homes (194 sq m).
The total lot (block) sizes of the Ecovillage were larger (987 sq m on average compared to the Observatory blocks which had an average of 770 sq m, indicating the Observatory homes built on more of their lot and the Ecovillage homes had more open space available for gardening and landscaping.
Average water usage was similar in both Ecovillage (426 litres per household per day) and Observatory homes studied (489 L/hh/pd; 164L/pp/pd) and no correlation was found between water consumption and the sustainable design of the home.
The amount of time the house was used (per cent occupied) did not correlate significantly with either energy use or water use so this was excluded from further analyses. The amount of time homes were occupied was very similar (79 per cent-80 per cent) from each community.
Analysis showed there was there was a significant beta weight for sustainable design, indicating that sustainably designed homes used significantly less energy.
The semi-partial correlation indicated that sustainable design uniquely accounted for 11 per cent of the variance in energy consumption.
The number of house occupants was positively related to energy use, accounting for 10 per cent of variance in energy. The size of the home accounted for less than one per cent of the variance in energy use and this was not statistically significant.
Environmental attitude variables (preservation and utilisation) did not add significantly to predicting energy use once dwelling characteristics were already accounted for.
The overall analysis including both sets of predictor variables accounted for 57 per cent of variance in energy use and this was statistically significant. Both sustainable design and number of occupants remained as significant independent predictors of energy used.
This suggests that the sustainable design of a home was the largest independent contributor to reduced energy use, among the variables included in the present study. We note however that the EA (preservation) scores are significantly correlated with energy use and further analysis will investigate this relationship.
The results highlighted the synergies between passive design and built environment operations.
The dwellings within the Ecovillage were found to consume significantly less energy. These results suggest that the difference was primarily due to the sustainable design of the homes. Further analysis may be able to reveal what impact people’s environmental attitudes have (if any) on energy use in contemporary or sustainably designed homes.
The reduced energy use is likely to be attributable to key features in each of the Ecovillage homes, including their use of only solar based hot water systems (often one third of a home’s energy use), insulation exceeding minimum building code of Australia standards.
Higher specification “low-e” glazing, eaves the optimal length to keep the sun out in summer and allow it inside in winter, limited west-facing thermal mass, protected internal thermal mass, efficient lighting and a dark sky policy to minimise lighting consumption and pollution are all feature of the Ecovillage homes
This result was achieved even though each home uses a water pump for potable water movement (transfer of rain water to each bathroom/kitchen outlet within the homes).
Note also that 23 Observatory homes had pools and these often use pool pumps which are a known high energy demand appliance. The Ecovillage shares one pool and this was not included in the analysis due to a lack of data and its likely negligible (average) effect.
Each Ecovillage home is required to install a minimum of 1kW of solar photovoltaic panels on their building and on average these generated 4.1kWh per day.
Some residents with 1.5kW or 2kW solar energy systems were able to ensure more energy was created than they used, triggering their electricity companies to pay them for the power (rather than having to pay any electricity bills themselves).
The size of homes and number of people per home were greater in the Observatory estate; however, when multivariate regression testing included these variables, the sustainable design of the home was still found to be a greater independent contributor to the variance in the model.
In terms of water, both estates used a similar amount of water and this was not found to differ significantly in relation to the design of homes or other factors.
We note however that the Ecovillage has its own internal water cycle and management system which takes no water from the town supply where as the observatory would use primarily town water (489 Litres/hh/pd). To evaluate the full lifecycle cost, the price of supplying water should be considered as previously studied in 2008 (Kenway S.J., Priestley A.J et al. 2008).
Those who reported they were in favour of environmental preservation tended to consume less energy, implying their attitudes and values impacted their behaviour and specifically household operations affecting energy (for example turning lights off, closing blinds and curtains on hot days, eliminating standby appliance power and activity using natural ventilation and fans instead of air conditioning).
This research was funded by the Australian Research Council, Landmatters Ltd, Department of Natural Resources, and Gold Coast City Council. The assistance of Dr Heather Green is also gratefully acknowledged.
The methods section of this report is similar to that described in a paper for Sustainable Buildings 2011 as the same method was used and this paper explains the results of the most recent analysis.
Ben O’Callaghan is Principal of sustainability consultants, Cundall and a Masters student at the faculty of architecture, design and planning, The University of Sydney.
Richard Hyde is Professor of Architectural Science in the faculty of architecture, design and planning, The University of Sydney.
For full details of this paper contact Mr O’Callaghan, firstname.lastname@example.org