Causes Of Climate Change
There is not space here to go into all the proposed explanations of the changes that have occurred in the last 100 kyr: there are more than enough textbooks on the subject (see Bibliography). All that is needed here is a brief summary of how virtually all the observed forms of change set out in Section 2.1 can be explained in terms of known physical processes. Starting with the broad cycle of the ice ages, which occur every 100 kyr or so, and have lesser cycles of around 20 and 40 kyr, received wisdom is that they are explained remarkably well in terms of the Earth's orbital parameters (Imbrie et al., 1992, 1993). These modify the amount of sunlight falling at different latitudes at different times of the year. It is found that the key to explaining how variations in the orbital parameters can trigger ice ages is the amount of solar radiation received at high latitudes during the summer. This is critical to the growth and decay of ice sheets. At 65° N this quantity has varied by more than 9% during the last 800 kyr. Computer modelling studies show that, when fluctuations of this order are combined with realistic assumptions about the time taken for ice sheets to build up, it is possible to reproduce the long-term behaviour of the ice ages with surprising accuracy. These calculations can reproduce the broad form of the climate changes measured in
ocean sediments and the EPICA ice core (see Fig. 2.5), with its sawtooth pattern of short interglacials every 100 kyr or so, followed by a slow descent into full glacial conditions and then a sudden warming into the next interglacial. Superimposed on this pattern are cycles of around 20 kyr in duration.
The fact that the orbital theory of ice ages is measured in tens of millennia means that these changes form the background to what follows. For the most part this is all we need to say, but for one particular feature. This is the nature of the Holocene variability in the Sahara. In the current interglacial the peak in summer insolation at 65° N occurred around 10 kya. In particular, rainfall over the Sahara during the past 20 kyr has exhibited a remarkable non-linear sensitivity to these gradual changes in insolation (see Section 2.7). What is particularly interesting is that the sudden changes in the Sahara appear to reflect a response to the lengthy cycles associated with the variations in the Earth's orbit.
Understanding changes in the Sahara requires us to look at tropical circulation around the end of the ice age. During the LGM much of northern Africa was, as now, arid desert. If anything this desert was even more extensive as the tropical rain forests were greatly reduced in extent. The reason for this shift has to do with the strength of the Intertropical Convergence Zone (see Section 2.7). This rising air produces heavy rainfall and the now dry air then spreads north and south before descending at around 20° N and 20° S. The very dry descending air then flows back towards the Equator. The entire circulation pattern is known as the Hadley cell. During the LGM the tropical oceans were cooler and the Hadley cell was less vigorous.
The more arid conditions during the LGM show up as a high level of wind-blown mineral dust in the sediment cores from the northern tropical Atlantic. Around 14.5 kya the level of dust dropped suddenly, suggesting the onset of much moister conditions (Fig. 2.11; deMenocal et al., 2000). The amount of dust blown westwards into the Atlantic remained low apart from an increase during the Younger
Depth of core (cm)
figure 2.11. Results from an ocean sediment core from the tropical Atlantic west of the Sahara Desert. This data shows a sharp decline in the amount of mineral dust being transported from the Sahara between around 15 kya and 5 kya (from deMenocal et ah, 2000, using data available on Peter deMenocal's website).
Dryas and then a return to moister conditions. These moister conditions lasted until around 5.5 kya when the sediment records show a sudden increase in the amount of dust being transported in the winds from the Sahara. Abruptly the climate shifted to a drier form and the desert began to take over.
What appears to have happened is that the cycles in solar insolation influenced the position of the ITCZ. Any shifts in its position or intensity can have a crucial impact on how wet or dry different areas of the tropics are. Around 14.5 kya the level of summer insolation rose to a critical level at 65° N: sufficient to trigger a sudden shift northwards of the ITCZ into the Sahara during the summer. This brought relatively heavy rainfall to much of the Sahara. The level of summer insolation remained sufficient to maintain this pattern well into the Holocene, except during the Younger Dryas when the disruption of North Atlantic circulation was sufficient to override it. Then around 5.5 kya the summer insolation at 65° N fell back below the critical level, the ITCZ moved south, and the Sahara rapidly became desert. This change was most rapid in the central and eastern Sahara, plus Arabia, where extreme aridity took over within a century on so. The fact that this sudden shift coincided with changes in the tropical Pacific may well have reinforced the scale of climatic change over Africa and Arabia as El Nino conditions tend to be associated with southward movement of the ITCZ.
Almost all the other features of the climate can be explained in terms of the natural variability of the global weather system. On timescales of millennia and shorter, various components of the climate system can interact to produce major fluctuations. In particular, the sudden changes identified as DO and Heinrich events in Section 2.5 appear to be linked to switches in the large-scale movement of water in the North Atlantic. These shifts are an example of the process driving the deep waters of the oceans, known as thermohaline circulation (Broecker, 1995). This results from changes in seawater density arising from variations in temperature and salinity. Where the water becomes denser than the deeper layers it can sink to great depths. The temperature depends on where the surface waters come from and how much heat the oceans either pick up or release to the atmosphere in both sensible heat and evaporative loss. The salinity of a given body of water depends on the balance between losses through evaporation and gains from either rainfall, or freshwater run-off from rivers and melting of the ice sheets of Antarctica and Greenland plus the pack ice of the polar oceans. In practice there are few regions where sinking waters have a major impact. Deep waters, defined as water that sinks to middle levels of the major oceans, are formed only around the northern fringes of the Atlantic (North Atlantic Deep Water). Bottom waters, which constitute a colder denser layer below the deep waters, are formed only in limited regions near the coast of Antarctica in the Weddell and Ross seas.
The changes that matter here are those associated with the North Atlantic. During and around the end of the last ice age the circulation here underwent sudden and substantial shifts. These led to the global climate being able to exist in distinctly different regimes even though the overall energy balance of the system had not changed appreciably. Heinrich events are most easily linked to changes of this type. During the last ice age the partial collapse of the ice sheet over North America, which caused a surge of icebergs to flood out into the North Atlantic, would have had a radical impact on ocean circulation. The combination of the huge influx of freshwater and the ice cover in winter would have effectively capped the northern part of the Atlantic and switched off any thermohaline circulation north of about 40° N.
The link between Heinrich events and changes in the northern hemisphere ice sheets appears to be well established. The climatic explanation of the first warming after the LGM, the Bolling (see Section 2.6) seems to have involved events in the southern hemisphere (Clark et al., 1996; Weaver et al., 2003). The influx of a large amount of meltwater into the Southern Ocean altered the formation of deep water around Antarctica and meant more tropical water moved northwards into the North Atlantic. This effect extended to high latitudes and led to a profound warming of not only the North Atlantic region but the northern hemisphere as a whole.
Although the most extreme events can be explained in terms of major adjustments in global ocean circulation, DO events rely on more complicated mechanisms. The most intriguing one is based on a stability analysis of the Atlantic Ocean during the ice age (Rahmstorf & Alley, 2002). In this approach there appear to be two circulation modes: one stable, and the other a weakly unstable one that lasted several centuries before spontaneously ending. In addition, there was a second unstable mode in which the formation of North Atlantic deep water is shut down completely, corresponding to a Heinrich event. This model suggests that the glacial Atlantic was an excitable system, in which a suitable perturbation could trigger a temporary transition of the state to an unstable circulation mode. In contrast a warm climate, like the present, is bistable and appears to be much less susceptible to disturbance. It is postulated that this process exhibits a threshold in respect of the background noisiness of the climate. Even today, this stochastic response to changes in the freshwater flux into the North Atlantic could have major climatic implications (see Section 8.5).
The importance of this proposal is that the available evidence suggests that the sensitivity of the climate to noise could have changed appreciably during and since the ice age. So it is possible that in the early warmer stages of the ice age, when the warm mode of the North Atlantic circulation was more stable, there were fewer and longer-lasting DO events. Conversely, at its nadir during the extreme cold around 20 kya, when the cold mode of circulation was more stable, there were also fewer DO events. In between, when the climate existed in a metastable state, random flickers between the two states might easily occur. If these flickers were in response to a weak 1500-year cycle, then they could take the form of a stochastic resonance between the two modes of North Atlantic circulation.5
5 This phenomenon is described in Burroughs (2003), pp. 234-7.
There is some evidence of the existence of the 1500-year oscillation in ocean-sediment records for the Holocene (Bond et al., 1997). These are, however, only very faint echoes of the DO events during the last ice age. Nevertheless, measurements of ice-rafted debris in ocean-sediment cores taken from various locations in the North Atlantic, notably between Greenland and Iceland, and to the west of Ireland, show evidence of changes in the circulation of the ocean during the Holocene. This debris originates from the east coast of Greenland and the Arctic archipelago Svalbard, and provides a measure of the strength of the Irminger current and the North Atlantic Drift (the extension of the Gulf Stream across the North Atlantic and into the Nordic Seas). The sediment records show periods of marked cooling at 2.8, 4.1, 5.9 and 8.2 kya. These changes in circulation correlate closely with measurements of solar variability, and suggest they could be the result of solar influences (Bond et al., 2001). This has led to the proposal that the observed changes in the circulation of the North Atlantic are the result of stochastic resonance driven by solar variability (Rahmstorf, 2003).
Although these changes do not show up strongly in the Greenland ice-core records, apart from the 8.2 kya event, their implications for the climate in Europe and the Middle East must have been considerable. Significant shifts in sea ice extent would have had an impact on the North Atlantic Oscillation (NAO). In turn, this would have altered rainfall patterns across Europe (see Section 5.1). The cooler episodes would have reduced the amount of rainfall, and it is interesting that these periods do coincide with periods of drought and social breakdown in Middle East (see Chapter 6).
These observations raise one further important question about the stability of the Holocene: in what circumstances does the climate of the North Atlantic become metastable again? There is lively debate in the climatological community as to whether current global warming resulting from human activities could push the climate into a more erratic mode. Some computer models of the global climate suggest that this is a distinct possibility. It is a frightening thought.
The evidence of the last ice age shows that a less stable climate would pose daunting challenges for the human race. So both understanding the nature of current climate change and appreciating fully how modern society emerged from the chaos of the last ice age assume even greater importance.
Average user rating: 5 stars out of 1 votes
Post a comment