Cooling the urban heat island with more reflective roofs

Research says Melbourne could benefit from less black. mugley/Flickr

Can a whiter roof make your home cooler? What about your whole city?

The existing literature and theory suggests that increasing the albedo - or reflectiveness - of a building will reflect incoming sun light and associated heat, reducing the building’s cooling requirements and also reducing the “Urban Heat Island” effect. The City of Melbourne recently commissioned our team to look into whether these claims hold true for Melbourne’s climate.

The urban heat island (UHI) effect refers to the phenomenon whereby a metropolitan or built up area is significantly warmer than its surrounding areas. In some cases, the UHI effect makes average urban daytime air temperatures around 5-6°C higher than the surrounding rural areas in summer.

The urban heat island effect can be detected throughout the year, but it is of particular public policy concern during the summer. This is because higher surface air temperature is associated with air pollution, heat stress-related mortality and illness, and increases in electricity demand for air conditioning. For example, in Melbourne during the heat wave of February 2009 there were 374 “excess deaths” reported.

Many civilizations rely on white roofs to keep cool: will Australia join them? mhobl/Flickr

Increased vegetation, higher albedo surfaces and higher albedo pavements are cited as the main opportunities to reduce this urban heat island effect. During a typical sunny day, there is approximately 1 kW/m² of solar radiation hitting a roof’s surface. Between 20% and 95% of this radiation is absorbed - the difference is based on the different roof colours. This massive heat load affects the microclimate around our cities. The thermal or long wave radiation reradiated from building surfaces affects air temperature, relative humidity and wind speed.

Akbari and Konopacki have calculated the how much energy – used for cooling – would be saved by heat island mitigation strategies. They looked at the application of cool materials and increases in vegetation cover for 240 regions in the United States. They found that for residential buildings, the cooling energy savings vary between 12% and 25%; for office buildings between 5% and 18%; and for commercial (retail stores) buildings between 7% and 17%.

Our Melbourne-based research tested four buildings in a suite of six full-scale buildings of approximately 12m². We also constructed a smaller half-scale building to help us look at things that are hard to test on a real building: what happens when the roof gets dirty; where best to put sensors; and taking thermal images. The building treatments included three different types of paint, a control and a “green roof” (which we’ll be reporting on later). Our results are available in the full report on the City of Melbourne website.

We found that for older buildings (those that do not meet current Building Code of Australia insulation requirements), the high albedo paints will always provide a significant reduction in cooling requirements and increased comfort. Those that benefit the most are industrial buildings such as warehouses, airports, shopping centres, factories and commercial buildings.

Residential buildings don’t benefit as much from white roofs. Mikenan/Flickr

Residential buildings show the least benefit. This is because homes need to be heated for a significant part of the year; painting the roof white loses the passive benefit of solar gain on the roof in winter (even if it’s minor). Further, when buildings have cooling systems on the roof, then the roof surface can be up to 40° cooler, leading to an efficiency benefit. That is, the system uses less energy to cool the air and the building requires less cooling because of the decreased solar gain in summer.

We looked at heat transfer, reflected energy off the roofs, internal ambient temperature, roof and ceiling temperatures and background weather data. We took into account variables such as insulation levels, paint tint and colour, roof pitch and overshadowing. All of these showed that there was still a benefit to using the white paint.

The only exception was a residence with R3.5 or greater insulation in the ceiling, because in Melbourne we spend 60% of the time heating and only 5-8% cooling. Under this scenario there was a benefit in summer but a cost in winter (though temperatures of twenty degrees or more can still build up in the roof cavity adding to the UHI).

This leads to the really interesting discussion which underpins this research. How does all this affect the larger context: urban heat islands, community benefits and the overall reduction of cooling energy use (and resulting greenhouse gas emission) across an entire city?

Our research, and that of the authors mentioned above, shows that there is a benefit to using the higher albedo treatments, green roofs and lighter roads and paths. The easiest and cheapest of these is paint. It will reduce cooling energy use, especially peak energy use. It will reduce the UHI and its associated health impacts. And it will lead to increased comfort.

To further enrich these findings we plan to continue our research for the next three years to look at different roofing materials, green roofs and photo-voltaic panel performance. We will also develop an urban microclimate model.