Covering some 71% of the Earth’s surface, the oceans are a fundamental component of the climate system. Interactions between the rapidly mixing atmosphere and the slowly changing ocean basins are largely responsible for the climatic variations. The high heat capacity of the oceans dampens the much higher temperature changes that would otherwise occur each day, each season and each year — both in coastal areas and often farther inland. The oceans are the birthplace of all tropical cyclones and most mid-latitude storms. Half the heat transported to the poles is carried by ocean currents, which is why Western Europe, for example, is such a hospitable place. The oceans are also the single most important sink for carbon dioxide produced by human action. The oceans can hold 50 times more carbon than the atmosphere and, when there is equilibrium between these reservoirs, the oceans can absorb up to 85 per cent of any additional carbon released into the atmosphere. Present high emission rates, however, prevent this equilibrium and only about 30 per cent of anthropogenic (human made) emissions now seem to enter the oceans.
Land surface-atmosphere interactions
Within the atmospheric boundary layer (the first few tens of metres above the ground) there are many complex physical processes at work. Understanding these processes is an essential part of improving our knowledge of climate, developing better climate forecasting models, estimating the impact of human activities on climate, and understanding how a changing climate might affect us. On a hot day, it is cooler within a canopy of leafy trees than where the soil or grass is exposed to direct sunlight. In winter, ground frost develops first on exposed grass rather than under trees.
Until recently, the representation of the land surface in computer models of weather and climate was quite inadequate. However, most coupled models now employ some representation of how vegetation controls evaporation and most can estimate river runoff for the ocean component of the model. Freshwater runoff and local rainfall affect the salinity distribution of the oceans and together are an important part of the development of the latest climate models.
The feedback process whereby climate-induced changes in vegetation affect the climate system, which further affects vegetation, potentially has large climatic implications. So far, however, it has proven difficult to incorporate this feedback process adequately in the coupled-model experiments used to estimate climate sensitivity. Also, the amount of carbon that is either extracted from the soil or stored in it by decaying vegetation is another source of considerable uncertainty. Snow, with its high reflectivity, is an important component of the land surface. Current climate models have some capability in simulating the seasonal cycle of snow extent but tend to underestimate interannual variability. These weaknesses limit confidence in the details of changes, particularly at middle and high latitudes, simulated by current climate models and goes to show how much the land surface influences climate.
Volcanoes can inject vast amounts of dust and, more significantly, the aerosol particle sulphur dioxide, into the upper atmosphere where aerosol particles remain suspended for up to several years and are spread around the entire globe forming a veil. The particles absorb sunlight and locally heat the stratosphere but at lower levels cause compensating cooling as less solar radiation reaches the Earth’s surface.
After large explosive tropical eruptions, the Southern Hemisphere shows a cooling (somewhat smaller than the Northern Hemisphere) in the three years following the eruptions, but the spatial patterns of the responses have been less well studied than in the Northern Hemisphere. The fact that climatically significant eruptions have, in recent centuries, occurred roughly every decade means that they are a significant factor in understanding climatic variability and climate change. Two recent eruptions, El Chicon (Mexico) in 1982, and Mount Pinatubo (Philippines) in 1991, provided the opportunity to make more detailed measurements. Mount Pinatubo appears to have injected the greatest amount of sulphur compounds into the stratosphere in the 20th century. This eruption also produced an extensive dust veil and generated significant cooling for several years. Somewhat surprisingly, however, a warming was observed over the continents of the Northern Hemisphere at higher latitudes in the first winter after the Mount Pinatubo eruption. Overall, the eruption of Mount Pinatubo caused quite a strong cooling of the global surface temperature (about 0.2°C) and in the troposphere (perhaps 0.4°C) from late 1991 to 1994.
The output of the Sun varies on all timescales. The best-known variation is the regular fluctuation in the number of sunspots, which show up as small dark regions on the solar disk, and affect the energy output of the Sun. Other aspects of solar activity include changes in the solar magnetic field, which influence the number of cosmic rays entering the Earth’s atmosphere from deep space, and variations in the amount of ultraviolet radiation from the Sun that may lead to photochemical changes in the upper atmosphere. All these variations have the potential to induce fluctuations in the climate.
Ground-based efforts during the first half of the century, designed to show that there were appreciable changes in the output of the Sun, were plagued by problems in correcting for the effects of atmospheric absorption. It was only in 1980, with the launching of specialized satellite instruments, that it was possible to measure accurately the changes in energy radiated by the Sun. Observations now show a modulation of about 1.5 W/m² in the solar output received by the Earth over the 11-year solar cycle. This is equivalent to about 0.1 per cent of the average incoming solar radiation (1370 W/m²). These changes cannot, however, be explained in terms of sunspots alone. Sunspots are areas of lower temperature and an increase in their number might be expected to coincide with reduced solar output. On the contrary, the energy output from the Sun peaks when the sunspot number is high.
It is now known that solar output is a balance between increases due to the development of bright areas, known as faculae, at times of high solar activity and the decrease resulting from increased sunspots. Overall the heating effect of the faculae outweighs the cooling effect of the sunspots. Estimates have also been made of the longer-term fluctuations in solar energy output over the past two or three centuries. The possibility that the Sun’s energy output may have varied more appreciably in the past could explain the marked parallel between these changes and estimates of the Earth’s surface temperature over much of the past four centuries.
Land-use changes have led to changes in the amount of sunlight reflected from the ground (the surface albedo). The scale of these changes is estimated to be about one-fifth of the forcing on the global climate due to changes in emissions of greenhouse gases. About half of the land use changes are estimated to have occurred during the industrial era, much of it due to replacement of forests by agricultural cropping and grazing lands over Eurasia and North America. The largest effect of deforestation is estimated to be at high latitudes where the albedo of snow-covered land, previously forested, has increased. This is because snow on trees reflects only about half of the sunlight falling on it, whereas snow-covered open ground reflects about two-thirds.
Overall, the increased albedo over Eurasian and North American agricultural regions has had a cooling effect.
Other significant changes in the land surface resulting from human activities include tropical deforestation which changes evapotranspiration rates (the amount of water vapour put into the atmosphere through evaporation and transpiration from trees), desertification which increases surface albedo, and the general effects of agriculture on soil moisture characteristics. All of these processes need to be included in climate models.
Except for climate change studies there are few reliable records of past changes in land use. One way to build up a better picture of the effects of past changes is to combine surface records of changing land use with satellite measurements of the properties of vegetation cover. Such analyses show that forest clearing for agriculture and irrigated farming in arid and semi-arid lands are two major sources of climatically important land cover changes. The two effects tend, however, to cancel out, because irrigated agriculture increases solar energy absorption and the amount of moisture evaporated into the atmosphere, whereas forest clearing decreases these two processes.