Greenhouse effect
Electric thermometer
Losing ozone over Canada
Losing ozone down south
Rainfall trends
Mysterious monthly cycle
Pinatubo's winter warmer
Back to John Gribbin's Home Page
Living with the greenhouse effect
John & Mary Gribbin
A great deal of what you read and hear about the so-called "greenhouse effect" is either exaggerated, or misrepresented, or both. But the basis for concern about uncomfortably rapid global warming occurring within our own lifetimes and those of our children rests on just three facts, and a reasonable inference.
The first fact is that there is an atmospheric greenhouse effect, which keeps the Earth warmer than it would otherwise be. The simplest way to get an idea of how important this is is to compare temperatures at the surface of the Earth with those on the airless Moon. There is no significant difference between the distance of the Moon from the Sun and the distance of the Earth from the Sun, so both receive the same amount of heat on each square metre of the surface that faces the Sun, and, other things being equal (which they are not) should reach the same equilbrium temperature. The average temperature at the surface of the Moon (averaging over the whole surface, including day and night sides) is actually -18 oC, while the average temperature on the surface of the Earth is 15 oC. The blanket of our atmosphere keeps the surface of the planet a full 33 oC warmer than it would otherwise be, and crucially (as far as life forms like us are concerned) raises the temperature above the freezing point of water.
There is no mystery about how it does this. Some gases in the air (chiefly carbon dioxide and water vapour) absorb infrared radiation (the same kind of heat radiation you can feel if you hold your hand near a warm radiator). Sunlight passes through the atmosphere essentially unaffected, and warms the surface of the Earth. The warm surface radiates infrared, not light (because it is cooler than the surface of the Sun), and some of this outgoing infrared radiation is absorbed in the atmosphere and re- radiated in all directions. Some of the re-radiated infrared radiation goes back down and increases the temperature at the surface. This is the atmospheric greenhouse effect -- and nothing to do, incidentally, with the way a greenhouse keeps warm, which is by letting sunlight in and stopping convection, trapping hot air that tries to rise under a roof of glass.
There are complications. Any increase in surface temperature increases the amount of evaporation from the oceans, which puts more water vapour in the air and increases the greenhouse effect, in a feedback process. As a result, adding a relatively small amount of carbon dioxide can produce a disproportionate warming, once the feedback is allowed for. On the other hand, there is a limit to the amount of infrared heat trying to escape from the surface in the wavebands where the greenhouse gases are absorbing energy. Heat that has already been trapped can't be trapped twice, so doubling the amount of greenhouse gases in the air will not double the strength of the greenhouse effect, but will have a lesser effect. A law of diminishing returns applies. All of these complications (and others) are taken account of in computer models of the greenhouse effect. These models accurately describe the differences in temperature between the airless Moon and the Earth, and the pattern of temperatures seen on Mars (which is further from the Sun and has a thin carbon dioxide atmosphere) and Venus (closer to the Sun with a thick carbon dioxide atmosphere). Astronomers and climatologists have a good understanding of how planets in the Solar System keep warm.
The second fact is that the amount of greenhouse gases in the atmosphere has increased over the past hundred years or so. Since the early 19th century, the amount of carbon dioxide in the air has increased from below 280 parts per million to above 350 parts per million. The absolute quantities are small, which only goes to show just how powerful the greenhouse effect is -- that 280 ppm plays a considerable part in keeping us 33 degrees warmer than the Moon. But the proportional increase is more than 25 per cent, clearly a dramatic change in any natural system. There is no doubt that this increase comes mainly from burning fossil fuel -- coal, oil and gas. First, not only the increase itself but the rate of increase matches the rate at which such fuel has been burnt, including "blips" caused by two World Wars and the Arab oil crisis. Secondly, analysis of air bubbles trapped in the icecaps of Greenland and Antarctica show that the natural concentration of carbon dioxide in the air has been below 280 ppm for hundreds of thousands of years. Other greenhouse gases, including methane (which occurs naturally and is a byproduct of agricultural activities) and CFCs (which are entirely synthetic, and are also implicated in the destruction of the ozone layer) are also building up in the air.
According to the computer models (and basic physics!), all of this should have increased the average temperature at the surface of our planet. Taking all the known greenhouse gases into account, and allowing for the feedbacks and diminishing returns that we have already mentioned, the Earth should have warmed by about two-thirds of a degree, Celsius, over the past hundred years.
The third fact is that the Earth has got warmer over the past century, by about half a degree, Celsius. The evidence comes from meteorological stations scattered about the surface of the planet, and chiefly located on land masses. But the accuracy of the measurements was dramatically confirmed by satellite observations in the 1980s and 1990s. Since 1979, some weather satellites have carried instruments which can indicate the average temperature of the globe. They provide complete cover, including the large amount of the planet covered by oceans, and the average temperatures revealed by these instruments from month to month closely match the averages obtained from traditional ground-based observing stations. This makes meteorologists confident that the ground-based record is a good guide to past temperature trends. Other satellite data also show, from even longer observations, that the climate change we are living through is not caused by a change in the output of energy of the Sun itself.
Unfortunately, the satellite instruments are not sensitive enough to detect any warming that may have happened since 1979, which would be less than the natural variability from year to year. In order to see the long-term warming trend, you have to average over different decades (some opponents of the greenhouse theory do not understand this, and mistakenly think that the "failure" of the satellites to see warming taking place from year to year "proves" that the Earth is not getting warmer). The last complete decade, the 1980s, was the warmest on record, with seven of the eight warmest years on record up to then. To put this in perspective, the coldest years of the 1980s were all hotter than the warmest years of the 1880s.
The reasonable inference is that the global warming (which is real) is related to the buildup of greenhouse gases in the atmosphere (which is real), but that the computer models may slightly overestimate the strength of the additional greenhouse effect. There is even a "smoking gun", in the form of the effect of the eruption of the volcano Mount Pinatubo on the global temperature.
Pinatubo erupted in June 1991, just after the record breaking warmth of the 1980s. It threw an enormous amount of debris into the stratosphere, spreading around the globe and partially shielding the surface of the Earth from incoming solar energy. The computer models (the same computer models used in calculations of the greenhouse effect) predicted that in the short term this would produce a dramatic cooling of the globe, temporarily returning the average temperature to the level typical of a century ago. Exactly the predicted cooling was seen over the next two years. But the models also predicted that as the volcanic debris cleared from the atmosphere in the 1992 and 1993, average temperatures would swiftly return first to the level of the 1980s and then, by the middle of the 1990s, to the slightly higher levels appropriate for the extra enhancement of the greenhouse effect as carbon dioxide and other gases have continued to pour into the atmosphere. The data for 1994 have now been analysed, and show that it was one of the warmest years on record, bringing global temperature levels back to the heights reached in 1990 and 1991, just before Pinatubo blew its top. The prediction is that, with greenhouse gases being emitted in ever increasing quantities, the warming will continue to new record levels in the rest of this decade, and beyond -- perhaps with brief setbacks if other major volcanic eruptions occur.
Should we care? That's really another story, but if nothing is done to curb the increasing buildup of greenhouse gases, the models suggest that temperatures will rise by a further 1.5 oC by the year 2030, bringing flooding of coastal regions around the world as sea levels rise, diseases normally associated with lower latitudes spreading out from the tropics, drought in the US Midwest (still the most important grain producing region in the world) and other climate changes. You may or may not feel that this is a price worth paying (by our children) for our own reliance on fossil fuels; but there is enough evidence to persuade an unbiased observer that it really is going to happen.
Electric guide to temperature
CONFIRMATION that the average temperature of the Earth can be determined from a single measurement at one location has come from a study carried out by Colin Price, of Columbia University, New York. This builds on the suggestion, made last year by Earle Williams, of the Massachusetts Institute of Technology, that the amount of lightning activity going on around the world is dependant on the average global temperature (see Science, 13 June 1992).
The link between lightning and surface temperature is through convection -- the higher the surface temperature, the stronger the convection, producing more storms and more lightning. The atmosphere of the Earth behaves like a spherical capacitor, with the conducting surface of the Earth below, conducting ionosphere above, and what Price describes as a "leaky dielectric" in between. Thunderstorms are continuously charging the capacitor, so in principle one measurement of the ionospheric potential should reveal the overall level of thunderstorm activity, and therefore the average temperature of the globe.
Price has confirmed this using satellite data which give genuine global average surface temperatures, and, unlike conventional weather records, are not biassed towards the continental regions of the northern hemisphere. Seasonal variations show that a 1 per cent increase in temperature produces an increase of about 20 per cent in global lightning frequency and a 20 per cent increase in the ionospheric potential (Geophysical Research Letters, vol 20 p 1363). This sensitivity of the potential to temperature changes means that a 1 oC rise in temperature would increase the potential by 10 kV, and makes the technique ideal for monitoring small changes in global mean temperature. But Price does sound one note of caution -- if the global climate changers so much that it significantly alters the amount of fine particles, ions and water vapour in the air, then the ionospheric potential may be affected by these changes as well as directly by the temperature changes.
Canadian ozone in decline
GROUND-BASED measurements of the amount of stratospheric ozone over Canada show that stratospheric ozone was depleted by about 14 per cent in the first four months of 1993, compared with the average before the 1980s. The measurements confirm and extend the evidence from satellite studies of ozone depletion at that time, and pinpoint the altitude of maximum depletion (about 30 per cent) as 16 km. The timing and altitude of the ozone loss point to debris from the Pinatubo eruption as the likely main cause.
The measurements were made as part of a continuous monitoring programme which has been carried out for more than thirty years. Instruments at five Canadian sites have been monitoring the ozone concentration of the stratosphere since the 1950s and 1960s, and seven new observing stations have been established since the middle of the 1980s. There are also eight sites where balloon-borne instruments are launched into the stratosphere once a week.
Although the monitoring network was upgraded in the 1980s, researchers from the Canadian Atmospheric Environment Service (based in Downsview, Ontario) have used the data from before the 1980s to provide their baseline against which to compare ozone fluctuations, because this represents the state of the stratosphere before any depletion caused by CFCs. The ground based spectroscopic observations indicate that the amount of ozone depletion in the lower atmosphere between January and April 1993 was between 11 per cent and 17 per cent; the balloon data, compared with a 1980-1982 baseline, indicate a decline of 13.5 per cent, and pinpoint the height of maximum ozone loss (Geophysical Research Letters, vol 20 p 1979).
The losses were occurring below the altitudes where ozone is thought to be at risk from chlorine released by the breakdown of CFCs produced by human activities. They are, the team point out, "consistent with the Pinatubo volcanic debris being the main contributor to the effect", but the data "do not rule out other depleting processes".
Ozone hits new low down south
FIGURES just released by the US National Oceanic and Atmospheric Administration show that the concentration of ozone above Antarctica reached the lowest figure ever recorded on 12 October 1993. The data come from a balloon flight over the South Pole, and show that a region 5 km thick, from an altitude of 14 km to an altitude of 19 km, was completely devoid of ozone.
The total amount of ozone in a column from the ground up through the atmosphere at that time was 91 Dobson Units (DU); the previous record low was 105 DU, recorded at the South Pole in October 1992. On that occasion, there was no ozone at all in a layer about 3 km thick, from 14 km upwards. Before ozone began to be depleted as a result of human activities, the normal concentration above Antarctica in southern spring exceeded 300 DU.
The NOAA researchers suggest that the record low recorded last year was a result of the continuing depletion of ozone caused by chlorofluorocarbons (CFCs) released through human activity, exacerbated by the effects of acidic material released by the eruption of Mount Pinatubo in the Philippines (Geophysical Research Letters, vol 21 p 421). This caused the region of total ozone depletion to grow upwards, with considerable partial depletion of ozone at heights up to 23 km.
As the volcanic material is removed from the atmosphere by natural processes, "It may," in the words of the NOAA team, "be a rather long time before Antarctic ozone levels reach the record lows of 1993". But since major volcanic eruptions tend to occur every ten years or so, another event of this kind is likely in the near future, when the burden of CFCs in the atmosphere will be even greater than it was in the early 1990s. That "would probably result in the further [upward] extension ofthe ozone hole".
Rainfall trends confirm that the climate is changing
The 1980s was not only the warmest decade on record, it was also the wettest, at least for regions north of latitude 50o (about the latitude of the south coast of England). This is one of the principal conclusions to emerge from a study of global rainfall trends carried out by Mike Hulme, of the University of East Anglia. The study provides clear evidence that the climate is changing, and backs up the evidence for climate change from the more familiar temperature record.
Over the past 140 years, the average temperature of the globe has increased by just under half a degree, Celsius. But although this is an important indication of climatic change, climatologists need more information about how other meteorological parameters, such as wind speed and direction, and precipitation, have changed in order to understand what is going on. Rainfall changes are particularly hard to gauge, because rainfall is not uniform and varies considerably from place to place and time to time, even on small scales. So you need a much higher density of observing stations to get an accurate guide to rainfall trends than you do for temperature trends.
Writing in Weather (volume 50, number 2, page 34), Hulme describes an analysis of rainfall trends since 1900, carried out using a data base which contains the records from 8300 rain gauges around the world. Many regions of the Earth's surface (including the Middle East and the interior of South America) are very poorly sampled, and even in the well-sampled regions there are fewer recording stations for earlier decades. Because at first measurements were not being made at enough sites around the world, the recording stations that did exist in the early part of the 20th century gave a high value for the average precipitation. By the 1930s, however, a more global network was available to give an accurate average of annual rainfall, and the earlier figures have now been corrected, using this information, for their "wet bias".
The pattern that emerges from the study is one in which there has been a marked decline in rainfall in the northern hemisphere tropics (south of 23.4 oN), little change at mid-latitudes (between 23.4 oN and 50 oN), and a marked increase in rainfall north of 50 on. Unfortunately, data for the southern hemisphere are too sparse to break down into latitude bands in this way. Hulme is too cautious to offer any explanation for why the rainfall patterns have changed in this way -- but this pattern of change is, as it happens, exactly the one predicted by the standard computer models of rainfall changes in a world warmed by an increased greenhouse effect.
Met men mystified by monthly cycle
What could possibly be the cause of a regular monthly variation in average temperatures, affecting the whole world but with the northern and southern hemispheres precisely out of step with one another, and amounting to a full fifth of a degree, Celsius? If you've got any idea, the meteorologists would like to hear from you, because they are as baffled by the discovery as everyone else.
The evidence comes from data gathered by the European Centre for Medium Range Weather Forecasting, and covers the five-year period from 1986 to 1991. It essentially gives the temperature everywhere around the world at 1200 GMT each day, and it has been analysed by Clive Best, of the Institute for Systems Engineering and Informatics, Ispra, Italy.
The temperature oscillation is stronger nearer the poles, and has a period averaging 30 days (plus or minus 3 days) long. Each day's data is averaged from 51200 temperature measurements. The variation is almost perfectly sinusoidal, and is bigger in the northern hemisphere, presumably because there is more land there (Geophysical Research Letters, vol 21 p 2369).
It might be natural to guess that the Moon itself is somehow responsible for the variation -- but there is no obvious way in which it could produce such a large effect. Lunar tides raised in the atmosphere of the Earth have been linked with rainfall variations in some parts of the world over a cycle 18 years long; but such a regular and relatively large, rapid variation in temperature is hard to explain. One thing is for sure, though -- anybody using these data to study long-term climate change had better take care to subtract out the effect before jumping to any conclusions about global warming.
Pinatubo's winter warmer
ALTHOUGH large volcanic eruptions cool the globe overall, researchers at the University of Maryland have now found that they also provide winter warming over Eurasia and North America in the first winter after the eruption. The effect is independent of which hemisphere the eruption occurs in, and it was very pronounced in the winter of 1991/92, following the eruption of Mt Pinatubo in the Philippines in 1991.
Volcanoes cool the globe by injecting large quantities of material, especially sulphuric acid droplets, into the stratosphere, where they absorb heat coming in from the Sun and prevent it reaching the ground. The influence can last for several years, and the Pinatubo eruption has cooled the Earth's surface, on average, by about half a degree, Celsius. But as well as cooling the suface of the Earth, this absorption of solar energy at high altitudes leads to a warming of the stratosphere. Both effects are weak at high latitudes in winter, when the Sun is low in the sky. So there is an uneven pattern of warming in the stratosphere, which increases the contrast in temperature between high and low latitudes and changes the pattern of winds around the globe.
Alan Robock and Jianping Mao have shown that the pattern of surface temperature changes in the months following 12 major eruptions, from Krakatau in 1883 to Pinatubo in 1991, closely matches the pattern produced in a computer model subjected to this kind of regional warming of the stratosphere (Geophysical Research Letters, vol 12 p 2405). It explains why both Eurasia and North America experienced warm winter weather in 1991/92, while the Middle East was unusually cold.
The key factor is a strengthening of zonal winds in high and mid latitudes in winter, which brings warm maritime air over the continents. Because the sea is generally warmer than nearby land in winter (and colder, of course, in summer), this warms the land. The same effect does not occur in the southern hemisphere, because there are no cold mid latitude continents there to be warmed by an influx of maritime air.