Everyone talks about global warming, but it's not easy to get one's mind around just how much heat we're talking about. Even more difficult is putting that heat energy in terms that the average layperson can grasp. Fortunately, some scientists are making an effort to do just that.

In a recent paper in Geophysical Research Letters, "Observed changes in surface atmospheric energy over land," Thomas Peterson, of NOAA's National Climatic Data Center in Asheville, NC, Katharine M. Willett of the Met Office Hadley Centre in Exeter, UK, and and Peter W. Thorne, who works alongside Peterson at the Cooperative Institute for Climate and Satellites, try to separate the various elements of all that energy being trapped by the greenhouse effect. There's surface temperature, kinetic energy (wind) and latent heat (energy associated with water changes from one state to another, such as during evaporation).

All that is useful stuff from people who make their living studying climate. But what's really interesting for our purposes is the team's effort to express the energy being absorbed by the atmosphere. As part of the paper's concluding section, they convert that energy into a gravitational equivalent: the energy required to lift an object:

The density of the atmosphere in lowâlying land areas is approximately 1.2 kg m^{ [Committee on Extension to the Standard Atmosphere, 1976]. So a cylinder of air 100 meters in diameter and two-meter-high holds approximately 18,800 kg of air. Our analysis indicates that on average, this amount of air is gaining energy at a rate of 1.1 Ã 107 J decadeâ1. The Gravitational Potential Energy (in joules) of an object held above the earth equals the mass of the object, times gravity, times the distance it is above the earth. The heaviest car we own, Dr. Thorne's SUV, weighs 1,535 kg and our lightest vehicle, Dr. Willett's bicycle, weighs 9.5 kg. For these objects to gain the equivalent amount of gravitational potential energy as this two-m-tall by 100-m-diameter cylinder of air gained in heat content, the car would have to rise 700 meters decadeâ1 while after 10 years the bicycle would be just above the mesosphere at an elevation of 110 km.}

Doing a little bit more math, and we can apply all that the planet as a whole:

... a two-meter-high layer of the atmosphere covering the global land surface would contain 3.37 Ã 10

^{14}kg of air and be gaining heat content at a rate of 1.9 Ã 10^{17}J decade^{â1}. This seems like a tremendous amount of energy and it is. Yetit is a drop in the bucket, three orders of magnitude less than the concurrent increase in heat content of the top two meters of the ocean and five orders of magnitude less than the concurrent increases in ocean heat content from 0 to 700 m depth.

That's a lot of heavy lifting.

--

Peterson, T., Willett, K., & Thorne, P. (2011). Observed changes in surface atmospheric energy over land Geophysical Research Letters, 38 (16) DOI: 10.1029/2011GL048442

- Log in to post comments

J decade^-1 is a horrible unit.

It's no more horrible than Joule per second, Dunc.

Yes, it is, because "seconds" are SI units, whereas "decades" are not, and "joules per second" (or, as we normally call them, "Watts") are well-established units that I use all the freakin' time. I know what a Watt is, I have a feel for it, I can think of lots of things (light bulbs, kettles, power plants and so forth) which give me a sense of scale for different orders of magnitude. Not so for J decadeâ1 - I have absolutely no idea what 1.9 Ã 1017 J decadeâ1 looks like, without converting it into a nice sensible SI unit first. (Which, TBH, I can't be bothered to do right now...)

It's not like the number of seconds in a decade is the sort of handy conversion factor anybody already has in their heads.

Dunc:

I calculate approximately 315,576,000 seconds per decade -

1 decade = 10 years

= ~3,652.50 days (at 365.25 days per year approx.)

= 87,660 hours (at 24 hours per day)

= 315,576,000 seconds (at 3,600 seconds per hour)

Since this is a blog comment, let's round that to 3.2 x108.

So 1.9 x1017 J decade-1 should come out to approximately 5.9 x108 J second-1 (or as you say, Watts).

Wikipedia claims the world energy consumption in 2008 (http://en.wikipedia.org/wiki/World_energy_consumption) was 142.3TWh. 365days/year*24hours/day=8760hr/yr. 142.3TWh/8760hr=16.2*10^9 W. So the top 700m of the ocean are heating up with 1000x the energy used by everyone on earth.

Just out of interest - are these facts socialist facts, undetectable by James Delingpole, or are they actual facts that even global warming deniers can accept?

As far as Delingpole is concerned, all facts are socialist facts.

Composer99: looks right to me. I was just being lazy... (Well, lazy and at work. ;))

> Yes, it is, because "seconds" are SI units, whereas "decades"

Nope, feet are not SI units. They're not silly.

All you're doing is saying "Joules per decade are not SI units".

So? That doesn't make them silly.

What WOULD be silly is using a time so short that the variation swamps the effect.

You know, using a second in climate budgets when it takes decades for climate to appear.

What if, Dunc, we moved to a second that wasn't based on weird divisions of the orbit of earth like 60 and 24?

Isn't it "silly" to define a second using such divisors?

So isn't the current SI unit of a second silly, and therefore using it in other measures likewise silly?

Parsec isn't an SI unit.

Meter is.

But wouldn't using the distance to the stars in meters be silly?

The distance km isn't an SI unit. But you won't find many people using "meters per second" when driving a car. Neither would they use the number of meters to the shops or the next town.

Isn't the kilometer silly?

How about the gramme. Who uses it? Nobody. Everyone uses the KILO gramme. But that just means "the thousand gramme". Isn't that silly? Not to mention "Tonne". Which also comes in "short" versions, "imperial" versions and "metric" tons.

How about knots. Silly, aren't they. But then again if you're measuring distances where the curvature of the earth makes a difference, when your maps all have to be flat and your ruler or dividor likewise measuring a flat (shortest) distance if you've got a more accurate spherical map, wouldn't using flat miles be a silly thing? Wouldn't using a nautical mile per hour that corrects for the non-flat distance be somewhat less silly, since you'd then be able to measure your flat distance and know that your knots will cover that flat distance measured mile in a more correct time without having to use a long decimal very close to 1 to make up the difference?

Please. Convert it into really standard units, like fully-loaded 747's per football field. Otherwise no one will get it. /snark

:-) George.

The point is that you use units that are appropriate unless you're working as an engineer or similar, where accuracy is more important than understanding.

Since the variation by second of energy use of the population of earth varies notably by the second, a single figure for joules per second wouldn't be a good number to use, either.

This is why you'll see energy use by country in joules in one year. Not a single figure of Watts.

The distance between here and the CBD is better measured in kilometers or miles, not meters or yards, therefore the number of yards per mile makes not one jot of difference, and the "ease" of having a thousand meters per kilometer is no advantage since nobody CARES what the distance is in meters.

But when it comes to doing maths testing, the number of square yards per acre makes a good mathematics test for schoolkids.

"hogsheads per furlong and that's the way I likes it!" is the same as "3.2 Libraries of Congress", a joke, not valid criticism.

The units always "cancel each other out". What counts are the orders of magnitute. Everything is on the log scale. power of two or power of three, four, five ?. That is the question. http://www.worldometers.info/