Ice age

The time span of the last 130,000 years has seen the global climate system switch from warm interglacial to cold glacial conditions, and back again. This broad interglacial-glacial-interglacial climate oscillation has been recurring on a similar periodicity for about the last 900,000 years, though each individual cycle has had its own idiosyncrasies in terms of the timing and magnitude of changes. As is usually the case with the study of the past, data are in short supply, and only a few sketchy outlines are known for the earliest cycles (Winograd et al. 1997). Even for the most recent oscillation beginning around 130,000 years ago there is still too much ambiguity in terms of the errors in geological dating techniques, in the gaps in the record, and in the slowness of responses by indicator species, to know precisely when certain events occurred and whether the climate changes were truly synchronous between different regions. The general picture summarized here (and in the separate map sections below) roughly reflects the present consensus gained from ice cores, deep ocean cores, and terrestrial and lake sediments around the world.

Warmth. Around 130,000-110,000 years ago (the Eemian interglacial), the Earth's climates were generally much like those of today, though somewhat warmer and moister in many regions. The climate record derived from long ice cores taken through the Greenland ice cap suggested that the warm climate of the Eemian might have been punctuated by many sudden and fairly short-lived cold phases, but these results are now thought of as inaccurate because the lower layers of the ice sheet have become buckled and jumbled up. However, at least one major cold and dry event during the Eemian seems to be corroborated by the terrestrial pollen record from Europe and China (Zhisheng & Porter 1997). The issue remains controversial, as this review article explains.

Cooling. Though the time at which the Eemian interglacial ended is subject to some uncertainty (it was probably around 110,000 years ago), what does seem evident from the sediment records that cross this boundary is that it was a relatively sudden event and not a gradual slide into colder conditions taking many thousands of years. The recent high-resolution Atlantic sediment record of Adkins et al (1997) suggests that the move from interglacial to much colder-than-present glacial conditions occurred over a period of less than 400 years (with the limitations on the resolution of the sediment record leaving open the possibility that the change was in fact very much more rapid than this).

Following this initial cooling event, conditions often changed in sudden leaps and bounds followed by several thousand years of relatively stable climate or even a temporary reversal to warmth, but overall there was a decline. Northern forest zones retreated and fragmented as the summers and winters grew colder. Large ice sheets began to grow in the northern latitudes when the snow that fell in winter failed to melt, and instead piled up from one year to the next until it reached thousands of metres in thickness.

As the cold grew more severe, the Earth's climate also became drier because the global 'weather machine' that evaporates water from the oceans and drops it on the land operates less effectively at colder temperatures and when the polar sea ice is extensive. Even in areas that were not directly affected by the ice sheets, aridity began to cause forests to die and to give way to dry grassland, which requires less water to survive. Eventually, much of the grassland retreated to give way to deserts and semi-deserts, as global conditions reached a cold, dry low point around 70,000 years ago (this is called the Lower Pleniglacial). By this time, most of northern Europe and Canada were covered by thick ice sheets.

In-between. By around 60,000-55,000 years ago, conditions around the world had become warmer, though still generally colder than today. The ice melted back partially, and there followed a long 'middling' phase in which the climate oscillated between warmer and colder conditions, often in sudden jumps. During some parts of this phase, conditions in the tropics may have been moister than they are at present, and at other times they were drier. Generally, the mid-latitude zones seem to have been drier than present, with cold steppe and wooded steppe instead of forests.

Cooling again. After about 30,000 years ago, the Earth's climate system entered another big freeze-up; temperatures fell, deserts expanded and ice sheets spread across the northern latitudes much as they had done 70,000 years ago. This cold and arid phase which reached its most extreme point sometime around 21,000-17,000 years ago (18,000-15,000 radiocarbon years ago) is known as the Late Glacial Cold Stage (and is also sometimes called the Upper Pleniglacial).

The point at which the global ice extent was at its greatest, about 21,000 years ago (18,000 14C years ago) is known as the Last Glacial Maximum. The Last Glacial Maximum was much more arid than present almost everywhere, with desert and semi-desert occupying huge areas of the continents and forests shrunk back into refugia. But in fact, the greatest global aridity (rather than ice extent) may have been reached slightly after the Last Glacial Maximum, somewhere during the interval 19,000-17,000 years ago (17,000-15,000 14C years ago).

Interstadials. Sudden warm and moist phases occurred at various times during the timespan of the last glacial phase, often taking Greenland and Europe from a full-glacial climate to conditions about as warm as at present. For the time period between 115,000 and 14,000 years ago, 24 of these short lived warm events have so far been recognized from the Greenland ice core data (where they are called 'Dansgaard-Oeschger events'), although many lesser warming events also occurred (Dansgaard et al. 1993). From the speed of the climate changes recorded in the Greenland ice cap (Dansgaard et al. 1989), and by observation of the speed of change in sedimentation conditions on land, it is widely believed that the complete 'jump' in climate occurred over only a few decades. The interstadials lasted for varying spans of time, usually a few centuries to about 2,000 years, before an equally rapid cooling returned conditions to their previous state. Recent study of high-resolution deep sea cores (Bond et al. 1997) suggests that for at least the last 30,000 years, interstadials tended to occur at the warmer points of a background north Atlantic (and global?) temperature cycle which had a periodicity of around 1500 years. Not every warm peak was marked by an interstadial, but when each interstadial did occur it tended to begin at around the peak of this background temperature cycle. The same pattern seems to have dominated the occurrence of Heinrich events (below), which tended to begin at the coldest point of the temperature cycle, and the same basic 1500-year climate cycle has apparently continued into the very different world of the Holocene (below).

Apart from the north Atlantic region, interstadials may well have affected climate in other parts of the world; some of them show up as strong temperature changes in the Antarctic ice cores at the other end of the world. They are also associated with brief peaks in atmospheric methane concentration, suggesting that the biological activity of swamps and herbivores around the world increased as a result of moisture and warmth.

Heinrich events. Opposite in sign to the interstadials were sudden intense cold and dry phases which occasionally affected Europe and the north Atlantic region, and possibly many other parts of the world. The Heinrich events were first recognized as the traces of 'ice surges' into the north Atlantic, but they show up in the Greenland ice cores and at least some are also detectable in the European pollen records and distant Antarctic ice cores. They may also show up as pine pollen peaks in Florida, and environmental changes in the Middle East, China, New Zealand and South America, though without better dating control it is difficult to say with confidence that these really are part of a global-scale pattern that fits the timing of northern Atlantic Heinrich events. In fact, Heinrich events 'sensu stricto' are merely the most extreme of a spectrum of sudden, brief cold events which seem to have occurred very frequently over the last 115,000 years. The Greenland and North Atlantic record (Bond & Lotti 1995) suggests that during the last 50,000 years, Heinrich events occurred around 41,000, 35,000, 23,000, 21,000 and 17,000-15,000 'real' years ago, apparently tending to start at the 'low point' of a 1500-year temperature cycle. Each interstadial lasted between several hundred and several thousand years, with the 21,000 y.a. event and the 17,000-15,000 y.a. event both perhaps representing the 'extreme' Last Glacial Maximum conditions mapped below. If this is the case, slightly milder (though still much more cold and arid than present) conditions may have prevailed during some parts of this period.

Warming, then a cold snap. Around 14,000 years ago (about 13,000 radiocarbon years ago), there was a rapid global warming and moistening of climates, perhaps occurring within the space of only a few years or decades. In many respects, this phase seems to have resembled some of the earlier interstadials that had occurred so many times before during the glacial period. Conditions in many mid-latitude areas appear to have been about as warm as they are today, although many other areas - whilst warmer than during the Late Glacial Cold Stage - seem to have remained slightly cooler than at present. Forests began to spread back, and the ice sheets began to retreat. However, after a few thousand years of recovery, the Earth was suddenly plunged back into a new and very short-lived ice age known as the Younger Dryas. Although the Younger Dryas did not affect everywhere in the world, it destroyed the returning forests in the north and led to a brief resurgence of the ice sheets. This map by D. Peteet shows the possible distribution of Younger Dryas cooling around the world. The main cooling event that marks the beginning of the Younger Dryas seems have occurred within less than 100 years, according to Greenland ice core data (Alley et al. 1993). After about 1,300 years of cold and aridity, the Younger Dryas seems to have ended in the space of only a few decades (various estimates from ice core climate indicators range from 20 - 70 years for this sudden transition) when conditions became as warm as they are today. Around half of the warming seems to have occurred in the space of a single span of 15 years, according to the latest detailed analyses of the Greenland ice core record (Taylor et al. 1997).

The start of the present warm phase, the Holocene. Following the sudden ending of the Younger Dryas, about 11,500 years ago (or 10,000 14C years ago), forests quickly regained the ground that they had lost to cold and aridity. Ice sheets again began melting, though because of their size they took about two thousand more years to disappear completely. The Earth entered several thousand years of conditions warmer and moister than today; the Saharan and Arabian deserts almost completely disappeared under a vegetation cover, and in the northern latitudes forests grew slightly closer to the poles than they do at present. This phase, known as the 'Holocene optimum' occurred between about 9,000 and 5,000 years ago (8,000-4,000 14C years ago), though the timing of the warmest and moistest conditions probably varied somewhat between different regions. Some of the events and regional climatic trends of the last 10,000 years are summarized in this time line by N.C. Heywood. The 'optimum' may have been punctuated by a severe cold and dry phase that affected climates across north Africa, southern Asia, Europe, the Americas and Antarctica about 8,200 years ago (7,500 14 y.a.), perhaps lasting for a century or two before a return to warmer and wetter conditions (Stager & Mayewski 1997). In Africa at least, the climate does not seem to have returned to the moist warm 'optimum' state that prevailed before this sudden drought, but it was significantly moister than at present. After about 5,000 years ago, there was a further cooling and drying in many areas (again, often sudden and stepwise), and conditions became more similar to the present-day. A particularly widespread cool event associated with relatively wet conditions seems to have occurred in many parts of the world around 2600 years ago (van Geel et al. 1996). A general pattern in climate during the Holocene has been detected from high-resolution cores in the north Atlantic. It seems that at least in the North Atlantic region, and possibly globally, there was a warm-cold cycle with a periodicity of around 1500 years (Bond et al. 1997). In the north Atlantic region, and probably adjacent oceanic areas of Europe, the change from peak to trough of each period was about 2 deg.C , a very substantial change in mean annual temperature (though only a small fraction of the change between glacial and interglacial conditions). The cold phases seem to have been relatively abrupt, and each lasted several centuries before an apparently rapid switch back to warmer conditions. on this approximate periodicity are dated at 11,100 10,300 9,400 8,100 5,900 4,200 2,800 and 1400 years ago; they include the 8,600 y.a. and 2,600 y.a. events which seem to have been the most extreme in terms of showing up in terrestrial records around the world.

The unstable nature of the Earth's climate history suggests that it may be liable to change suddenly in the future. By putting large quantities of greenhouse gases into the atmosphere, humans are exerting pressure on the climate system which might produce a drastic change without much prior warning. As the geologist W.S. Broecker has said, "Climate is an angry beast, and we are poking it with sticks".

If you can remember back to the bitter winters of the late 1970s and early 80s you might also recall that there was much discussion in scientific circles at the time about whether or not the freezing winter conditions were a portent of a new ice age.

Over the past couple of decades such warnings have been drowned out by the great global warming debate and by consideration of how society might cope in future with a sweltering planet rather than an icebound one. Seemingly, the fact that we are still within an interglacial period, during which the ice has largely retreated to its polar fastnesses, has been forgotten - and replaced with the commonly-held view that one good thing you can say about global warming is that it will at least stave off the return of the glaciers.

Is this really true, or could the rapidly accelerating warming that we are experiencing actually hasten the onset of a new ice age? A growing body of evidence suggests that, at least for the UK and western Europe, there is a serious risk of this happening - and soon.

The problem lies with the ocean current known as the Gulf Stream, which bathes the UK and north-west Europe in warm water carried northwards from the Caribbean. It is the Gulf Stream, and associated currents, that allow strawberries to thrive along the Norwegian coast, while at comparable latitudes in Greenland glaciers wind their way right down to sea level. The same currents permit palms to flourish in Cornwall and the Hebrides, whereas across the ocean in Labrador, even temperate vegetation struggles to survive. Without the Gulf Stream, temperatures in the UK and north-west Europe would be five degrees centigrade or so cooler, with bitter winters at least as fierce as those of the so-called Little Ice Age in the 17th to 19th centuries.

The Gulf Stream is part of a more complex system of currents known by a number of different names, of which the rather cumbersome North Atlantic Meridional Overturning Circulation (Namoc) is probably the most apt. This incorporates not only the Gulf Stream but also the cold return currents that convey water southwards again. As it approaches the Arctic, the Gulf Stream loses heat and part of it heads back to warmer climes along the coast of Greenland and eastern Canada in the form of the cold, iceberg-laden current responsible for the loss of the Titanic. Much, however, overturns - cooling and sinking beneath the Nordic seas between Norway and Greenland, before heading south again deep below the surface.

In the past, the slowing of the Gulf Stream has been intimately linked with dramatic regional cooling. Just 10,000 years ago, during a climatic cold snap known as the Younger Dryas, the current was severely weakened, causing northern European temperatures to fall by as much as 10 degrees. Ten thousand years before that, at the height of the last ice age, when most of the UK was reduced to a frozen wasteland, the Gulf Stream had just two-thirds of the strength it has now.

What's worrying is that for some years now, global climate models have been predicting a future weakening of the Gulf Stream as a consequence of global warming. Such models visualise the disruption of the Namoc, including the Gulf Stream, as a result of large-scale melting of Arctic ice and the consequent pouring of huge volumes of fresh water into the North Atlantic, in a century or two. New data suggest, however, that we may not have to wait centuries, and in fact the whole process may be happening already.

So that the warm, saline surface waters of the Gulf Stream can continue to push northwards, there must be a comparable, deep return current of cold, dense water from the Nordic seas. Disturbingly, this return current seems to have been slowing since the middle of the last century. Bogi Hansen at the Faroese fisheries laboratory, and colleagues in Scotland and Norway, have been monitoring the deep outflow of cold water from the Nordic seas as it passes over the submarine Greenland-Scotland ridge that straddles the North Atlantic at this point. Their results show that the outflow has fallen by 20% since 1950, which suggests a comparable reduced inflow from the Gulf Stream.

Although there is as yet no direct substantiation of this, and his colleagues point to reports of the cooling and freshening of the Norwegian Sea and to temperatures that are already falling in parts of the region as possible evidence of contemporary Gulf Stream weakening.

It also seems that it is not only the intensity of the outflow of cold water that is changing. Bob Dickson of the Centre for Environment, Fisheries, and Aquaculture Science at Lowestoft, and colleagues, have reported a sustained and widespread freshening of returning deep waters south of the Greenland-Scotland ridge, which appears to have been going on for the past three or four decades.

Already the freshening is extending along the North American eastern seaboard towards the equator, in the so-called Deep Western Boundary current.

One of the scariest aspects of the current dramatic changes occurring in the system of North Atlantic currents is that the deep, southward-flowing limb of the Namoc can be thought of as representing the headwaters of the worldwide system of ocean currents known as the Global Thermohaline Circulation. The possibility exists, therefore, that a disruption of the Atlantic currents might have implications far beyond a colder UK and north-west Europe, perhaps bringing dramatic climatic changes to the entire planet.

Yet again, this highlights the fact that global warming, for which we have only ourselves to thank, is nothing more nor less than a great planetary experiment, many of the outcomes of which we cannot predict. Wallace Broecker, an ocean circulation researcher at New York's Lamont-Doherty Earth observatory, described the situation perfectly when he pointed out that "climate is an angry beast and we are poking at it with sticks". Let's hope that when it truly turns on us, its teeth don't match its outrage.