Hemispheres

Throughout the 20th century, global average temperature shows an interesting evolution. From about 1915 to about 1944, temperature rose. Then from about 1944 to 1951, global temperature declined. It remained roughly constant from about 1951 to about 1975, and since then it has risen sharply.

People often wonder why the planet didn’t warm from 1944 to 1975. Denialists often say that the planet actually cooled for 30 years or more, but this is simply not so; the cooling was confined to a brief period (about 1944 to 1951), followed by relative stability for several decades. But the question remains, with man-made CO₂ (and other greenhouse gases) in the atmosphere, why did the planet not warm for several decades mid-century?

The answer is that during that time, the warming from man-made greenhouse gases was offset by the cooling from man-made aerosols. Aerosols are tiny particles which remain suspended in the atmosphere for several days to weeks if they’re in the troposphere (the lower part of earth’s atmosphere), or several years if they’re in the stratosphere (the upper part of earth’s atmosphere). They tend to block incoming sunlight from reaching earth, thereby cooling earth’s climate.

Much of man-made aerosols are sulfates emitted by industrial activity; the boom in industrial production late in world war II, and during the post-war era, caused them to become a potent climate-forcing agent. Man-made aerosols are generally confined to the troposphere, so they settle out in a matter of days to weeks, but if they’re being continually replenished their climate forcing effect can be persistent. In addition to acting as a climate forcing agent, sulfate aerosols are also responsible for acid rain; that’s why in the 1960s-1970s environmental regulations were enacted to limit their emission, and man-made aerosols declined. Since tropospheric aerosols settle out of the atmosphere before they can disperse worldwide, their environmental impact is concentrated in the geographic region in which they’re produced.

Aerosols can also be injected into the atmosphere by volcanic explosions; for massive explosions the aerosols can be blasted into the upper atmosphere, the stratosphere, where they can disperse worldwide, and it takes years for them to settle out of the air. Hence large volcanic explosions have a distinct cooling effect on the climate; the most recent such event was the explosion of the Mt. Pinatubo volcano, which caused global temperature to take a nose-dive for a couple of years afterward.

CO₂ stays in the atmosphere a long time; it’s “residence time” is better measured in centuries than years. Hence it tends to mix throughout the planetary atmosphere (with a “mixing time” of about 7 or 8 months), and its warming impact is truly global. Also, because they’re so persistent, even after emissions stop it takes a long time for the atmosphere to return to its undisturbed state; the warming effect of CO₂ is both global and persistent.

If man-made aerosols were responsible for the cessation of global warming mid-century (which they were), and the climate effect of man-made aerosols tends to be concentrated in the region of emission (which it does), and the vast majority of industrial activity is in the northern hemisphere (which it is), then the mid-century cooling impact of aerosols should be concentrated in the northern hemisphere. Because of this, the question arose recently on the RealClimate blog concerning the different warming/cooling between the northern and southern hemispheres throughout the 20th century. Let’s take a look at what the data tell us about the differences in 20th-century temperature evolution between the northern and southern hemispheres.

NASA GISS, in addition to estimating global average temperature, also estimates average temperature separately for the two hemispheres. In an earlier post I showed the difference between the hemispheres, based on data for land-only stations in the two hemispheres. Here’s the temperature evolution of the two hemispheres using both land and ocean data (northern hemisphere in red, southern in blue):

The “smooth” versions show that while the northern hemisphere did indeed cool for several decades mid-century, the southern hemisphere did not — exactly what we would expect from sulfate-aerosol forcing. In fact, the smoothed version tends to smooth the very brief southern-hemisphere cooling a little too much; the annual numbers show that the southern hemisphere mid-century cooling is pretty much confined to a single year, from 1945 to 1946. If we smooth the southern-hemisphere temperature data in two separate pieces, up to 1945, and from 1946 onward, we get this:

It’s evident that the northern hemisphere shows decades of cooling mid-century, the effect of man-made aerosols. But the southern hemisphere shows only a single year of notable cooling; besides the 1945-1946 cooling event, the southern hemisphere shows consistent warming throughout the century.