Climate Change, White Roofs, and Common Sense

by Bill Chameides | November 1st, 2011
posted by Erica Rowell (Editor)

Permalink | 7 comments

This roof on a Las Vegas Walmart is white (and studded with skylights) to reduce energy use. But is it really a cool roof? (Photo courtesy of Walmart/Flickr)

Crossposted with National Geographic’s Great Energy Challenge blog.

White roofs may not be so green.

Sometimes It Gets Complicated

Recently in these posts, I’ve argued that in some cases, as in discerning the long-term global temperature trend (or employing smart regulation when it comes to deepwater drilling), a little common sense will suffice.

But sometimes the climate system can surprise you — what seems like common sense may prove to be not so commonsensical. In a paper published in the Journal of Climate authors Mark Z. Jacobson and John Ten Hoeve of Stanford University argue that the climate impact of white roofs is a case in point.

Be a Painter, Paint It White

When faced with a tough problem, it’s often a good idea to start with the low-hanging fruit — the simple, easy stuff that gets the ball rolling.

In the case of global warming, white roofs looked to be the low-hanging no-brainer. Things are getting too hot? Simple, paint the roofs white so that they reflect more sunlight, and presto chango, things cool down a bit.

Cool idea, right?

I thought so, and so did our very own Nobel prize-winning secretary of energy, Steve Chu, who has publicly urged a grassroots paint-roofs-white movement. (See video.)

Chu cites estimates that painting roofs (specifically flat roofs) white would be the equivalent of reducing carbon emissions from all cars on the road for 11 years. Imagine a world with no cars for 11 years. Hard to do. But imagine a world with white roofs. Easy as pie — and that, I guess, was the idea.

Easy, But Not Such a Good Idea, Say Authors Jacobson and Ten Hoeve

When you paint a roof white, there’s more happening than meets the eye. Sure, white roofs reflect sunlight and that cools the Earth. But white roofs, by reducing the amount of solar heating at the surface, also decrease the temperature difference between the surface and the air about a half mile to a mile up from the surface. It turns out that that temperature difference is what drives cloud formation. So, by decreasing the temperature difference, white roofs can suppress cloud formation, which allows more sunlight to penetrate, and that, lo and behold, heats things up — exactly what we don’t want.

The net effect of making roofs white depends on which of these two competing processes dominates — cooling by reflection or heating by cloud suppression? To find out which is more important, Jacobson and Ten Hoeve carried out a series of climate simulations using a model called GATOR-GCMOM, a global-regional “Gas, Aerosol, Transport, Radiation, General Circulation, Mesoscale, and Ocean Model that attempts to simulate climate, weather, and air pollution on all scales.” Impressive. If I had my druthers I wouldn’t call it GATOR-GCMOM, I’d call it OMOE — Our Model of Everything.

Whatever the model’s name, Jacobson and Ten Hoeve ran it for 20 years of simulation time for a world with no urban areas, a world with today’s distribution of urban areas, and one with all white roofs in all the world’s urban areas. Here are their results:

  • The urban heat island effect is negligible. The phenomenon that causes urban areas to be a few degrees warmer than nearby rural areas (because of buildings and other impervious surfaces) has only a small contribution to gross global warming — on the order of two to four percent. This result is in line with other investigations, including data released last week by the Berkeley Earth Surface Temperature (BEST) project, but counter to claims from the climate denier camp (see also here).
  • When roofs were painted white, local temperatures in cities did cool as one might expect. On average (weighted for population) the drop in temperature was about 0.04 degrees Fahrenheit. That’s not a whole lot, but if you hang out in a city during the summer, you’d be thinking right about now, Bring on those white roofs! And I wouldn’t blame you.
  • However, globally, those roofs were estimated to cause warming. Again, the temperature is not much, we’re talking only about 0.13 degrees Fahrenheit, but still, the authors argue, it’s an amount larger than the statistical noise in their model.

Another Complication

One thing the authors didn’t assess is the impact of white roofs on energy use and the concomitant fossil fuel emissions. By cooling temperatures in the summer, white roofs should lessen the need for air conditioning, which in turn reduces electricity use, greenhouse gas emissions, and warming. However, during the winter, white roofs increase the demand for heating, which in turn leads to more emissions and more warming.

While Jacobson and Ten Hoeve did not look at that aspect of the problem, Oleson et al did in a paper published in the journal Geophysical Research Letters in February last year. They found that on average over the globe, wintertime energy demand trumped warm weather cooling gains, implying that this aspect would only add to the net warming effect from white roofs. When I covered this paper last year, I noted that there was a way around this problem — temperature-sensitive, color-changing roofing material. And wouldn’t you know it, a team of students at MIT was already developing such a material. One could also only paint roofs white in warmer areas where heating needs are small.

A Cautionary Tale

Models of everything can be impressive things, and kudos to the scientists who have developed them, but they are in the end just models and models that are attempting to simulate a very complex array of processes. For that reason, I wouldn’t take Jacobson and Ten Hoeve’s result that white roofs cause warming to the bank just yet. And the authors agree and in fact recommend further work.

But their calculations do illustrate very nicely that using white roofs to cool the climate might actually turn out to be a “no-brainer” in a somewhat different sense. And there’s another point. There are folks out there who are suggesting that we handle the glo
bal warming problem through geoengineering — consciously altering the environment to counter warming rather than, or in addition to, lowering greenhouse gas emissions. The argument against this path is that one can never be sure what the actual effect of such actions might be — because the climate is so complicated, geoengineering could have counterintuitive, even harmful effects. (See here, here, here, and here.)

Of all the geoengineering ideas, the one often cited as the simplest, most obvious one, the one we should get started on right away is painting roofs white — after all how can that not help? How indeed.

filed under: climate change, faculty, global warming, temperatures
and: , , , , , , , , ,


All comments are moderated and limited to 275 words. Your e-mail address is never displayed. Read our Comment Guidelines for more details.

  1. Mikaela
    Nov 22, 2011

    Have you ever thought of spraying a large piece of fabric white to reflect the sun and then remove it in the winter? If it’s a possibility, then check it out at

  2. MMM
    Nov 21, 2011

    Two issues in the studies linked: First, it seems like these models look at the reduction of the UHI effect due to white roofs, but not the actual effect inside the house: I think the way the models are set up, a city with all black roofs will be warmer than a city with alternating white and black roofs, but in the latter case, the buildings will cool almost identically regardless of whether a specific house is white or black, and I think in reality that wouldn’t be true. Second, in a world with global warming air conditioning will increase and heating will decrease. Add that to the spread in air conditioning beyond just the US, and I think this “counterintuitive” result will disappear… -MMM ps. A little late now, but your post on the Himalayas states that “an IPCC official initially dismissed the criticism as “voodoo science,””… unfortunately, this is perpetrating a misreading pushed by the skeptic community. If you go and dig up the original Pachauri “voodoo” statement, it was related not to the 2035 date, but rather to the last chapter of an Indian report which got climate science pretty badly wrong. See, which tries to claim that glaciers are responding to medieval warm period heating with a 1000 year delay, and says that it would “tax the mind” to link glacier retreat to the “by and large” warming of the globe…

  3. Chris Winter
    Nov 15, 2011

    “However, during the winter, white roofs increase the demand for heating, which in turn leads to more emissions and more warming.” This, like the clouds result, seems counter-intuitive. I’d expect that on a well-insulated building the color of the roof would make negligible difference in the heating bill.

  4. Hank Roberts
    Nov 5, 2011

    I’d like to know if they got anywhere with the idea. I haven’t found new news.

  5. Luise
    Nov 2, 2011

    Would I use the air conditioner less if the temperature were 90.00 degrees as opposed to 90.04 degrees? Or am I misunderstanding the temperature measurement listed in the article?

    • Bill Chameides
      Nov 21, 2011

      Luise: The amount of air conditioning you use is less a function of the temperature of the outside air than the amount of energy your roof absorbs from sunlight. A white roof reflects much of that heat; a black roof absorbs it. The amount of the absorbed heat that actually gets into your attic/house/apartment depends on the effectiveness of the insulation you use. But in general you can expect a white roof to give you some AC relief.

©2015 Nicholas School of the Environment at Duke University | Box 90328 | Durham, NC 27708
how to contact us > | login to the site > | site disclaimers >

footer nav stuff