Comparing Measured Consumption with Modelled Consumption Using the Code for Sustainable Homes (2009)

This investigation was undertaken by Portsmouth Water to assess how water-efficient fittings can reduce water demand in new homes. However, no water-saving devices were installed by the developer as planned, which is why only water consumption patterns were assessed. The sample of 60 homes with a mean occupancy of 2.17 people consumed on average 125.9 litres/household/day. Consumption was lower in the shared than in private ownership properties, despite having very similar water fittings. A 6/4 L dual flush toilet was seen to be relatively ineffective at driving down demand compared to a 6 L single flush option.

Summary

This investigation was initially undertaken in order to assess the extent to which water-efficient fittings reduced water demand in the home.  The Dunsbury Way development was awarded “good” status under Ecohomes and it was expected to have flats, maisonettes and houses that were essentially identical to each other apart from their water fittings, being delineated along a private and rented/shared ownership divide.  The dwellings built to a rented/shared ownership spec were to have 6L/min (max. flow) showers, aerated basin taps throughout and water butts (where the dwellings had gardens).  The private dwellings were to have unrestricted showers and taps, and no water butts.  This would have been a very good opportunity to look in detail at the PCCs of the occupants either side of this divide, and infer the effectiveness of water efficient fittings at reducing demand.  However, disappointingly, it was found during a site visit that the developer had installed similar fittings throughout.  This was contrary to the items that were listed in the specification and the subsequent rating that was awarded under Ecohomes, and a product of the fact that there was no post-build inspection required.  

Despite this, some useful information resulted.  Mean consumption in the development was 125.9 L/h/d (some with gardens, some without), and the development was built to a lower environmental standard than future new housing (which will be assessed against the Code for Sustainable Homes, the successor to Ecohomes, and fall under the Building Regs Part G revision), and therefore this supports Portsmouth Water’s assumption to treat all future new housing as 120L/h/d (+10 for misc./external, +9 for dry year).  Mean occupancy in the development is 2.17 (compared to company average of 2.3), supporting the trend towards smaller households.

The Environment Agency WRMP Guidance states that water companies must clearly outline how they have taken account of the Code for Sustainable Homes in their demand forecast, and notably the condition that all future new social housing must meet “Code Level 3” of the environmental standard.  This includes achieving a theoretical domestic (internal) water consumption of no more than 105 L/h/d, which is calculated on a per capita basis from a very simple micro-components model.  This micro-components model is currently under review by Communities and Local Government, to which this study has made a contribution.  It is widely accepted that modelling at a per capita, rather than per household level is of little use to the water industry, that the error in the calculation is far greater than the difference between Code Levels, and that in its present form the calculator essentially forces expensive technologies (such as greywater and rainwater) at Code Level 3, and may lead to what are known as “perverse outcomes”.  As found in all other known examples from the industry and BRE, the Code Calculator significantly over-estimates PCC in all cases in the Dunsbury Way development.  A sample of 60 homes (flats, maisonettes and houses) was used in this study.  On average, these homes combined used 16,423 L/d, whereas the aggregated Code for Sustainable Homes Calculator puts the consumption at 24,556 L/d.  This was found using the real occupancies of the households (known from a questionnaire); if the company mean occupancy of 2.3 had been used this predicted consumption rises to 28,608 L/d.  Interestingly, the Ecohomes predicted consumption is much more accurate, but the assessment method is no longer accepted in England and Wales.

All the properties in the sample are new and from a high density development, and many are in fact approaching a Code Level 3 water consumption existence. On average, the occupants are falling just short of Code Level 1 consumptions levels (which is 120 L/h/d).

In addition, we have collected data on micro-components, consumption and occupancy, and have established contact with the occupants on site such that there is scope to use them in future water efficiency trials.  We have also collected anecdotal evidence on the understanding of the dual-flush function (not all understood, and felt that this was down to the housing association providing a poor introduction to the home in general), and also found examples of the dual flush buttons sticking, breaking and leaking, despite being only 8 months to 2 years old.

The report concludes that:

• Consumption is lower in the shared ownership properties than in the private properties, despite having very similar water fittings.  This may be because many of the shared ownership occupants had young children and babies who may not use as much water
• Future new housing can be treated as 120L/h/d (internal consumption, average year) on average in the area of supply
• Occupancy in future new housing may be as low as 2.17
• A 6/4 L dual flush toilet seems to be relatively ineffective at driving down demand compared to a 6 L single flush option
• The contacts in this development are possible participants in any water efficiency trials that the company conducts