Figure 1 A snapshot of the travels of some polar bears around Svalbard
Figure 1 shows a colour snapshot of the WWF (World Wide Fund for Nature) Species Tracker online dynamic map showing the travels of some polar bears around Svalbard. The map shows Svalbard in white and ocean in blue and five different coloured polar bear icons and their associated tracks. One bear has travelled extensively around the western Svalbard coastline (over the land and ocean areas), one a short distance in the north western inland region, and one around the northern coastline moving from the land to the ocean. For the other two bears the tracks are very short. The text to the right of the map states that the Norwegian Polar Institute studies and tracks these polar bears with the support of WWF and gives tag numbers for four of the bears. Figure 1 shows the movements of several satellite-tracked females around Svalbard, which is a group of islands about halfway between mainland Norway and the North Pole. A polar bear typically travels several thousand kilometres per year in search of its main prey species – the seal. The state of the seas and ice of the region will therefore directly affect the bearsIt turns, however, that polar bears are also impacted by effects from much further afield as you will look at next. 1.1 Polar bears and pollutantsAttaching a satellite-tracking device to polar bears is not easy, and they have to be drugged (Figure 2). This gives an opportunity for them to be weighed, measured and tagged, and have various samples such as hair, fat and teeth removed for later chemical analysis.
Figure 2 Scientists examine a drugged polar bear
Figure 2 shows a colour photograph of a drugged polar bear lying in the snow with its head closest to the viewer. A silhouetted scientist kneels to the right of the bear, reaching towards it with a notebook on their lap, while another stands behind it. The amount of body fat on a bear indicates whether it has been eating well or is starving. But a chemical analysis of this body fat gives a surprise: polar bears have measurable amounts of a family of chemicals called polybrominated diphenyl ethers (PBDEs) in their fat. The same family has also been measured in Arctic ringed seals and other Arctic wildlife (Figure 3). PBDEs are a group of synthetic chemicals developed over the 20th century as fire retardants. Fabrics and furniture are impregnated with them, with the sole aim of slowing the rate at which they burn, and for which they have been very successful. However, once created, PBDEs are very difficult to destroy and will not break down into their elements over time. For this reason they are considered a persistent organic pollutant (POP).
Figure 3 Measurements of PBDE concentration in wildlife at different Arctic sites (Arctic Monitoring and Assessment Programme (AMAP), 2009)
A map showing a stereographic plot of the Arctic regions. The North Pole is at the centre and the lowest latitude is 50° N. There are concentric lines of latitude at 60° N, 70° N and 80° N. The Arctic Circle (66.6° N) is shown as a dashed line. The ocean regions are coloured blue and land is coloured pale green. Superimposed on this plot are six coloured bar charts in different locations showing the concentration of PBDE in the fat of various wildlife. The bar charts are for beluga (2 bar charts in different locations), burbot liver, ringed seals (2 bar charts) and thick-billed murre seabird eggs. All are in northern Canada, except one bar chart for ringed seals which is in western Greenland. Each bar chart shows PBDE concentration measurements during the period 1975 to 2005. The measurements are sparse through this period but broadly show an increase through time, typically from 1-5 nanograms of PBDE per gram of lipid weight in the 1970s-80s increasing to 25-35 ng/g in the 2000s. The exceptions to this pattern are ringed seals in northern Canada, where the increase is from around 1 up to around 5 ng/g over this period, and beluga in northern Canada, where there are no data before 1990 so it is not possible to know whether the concentrations increased. In the late 1970s and early 1980s, scientists began to detect POPs in the tissues of fish and shellfish close to populated areas. Concentrations were then detected in human breast milk, and the levels were shown to be increasing with time – perhaps through direct exposure to PBDEs or through bioaccumulation (see Section 1.2). The scale in Figure 3 is given in nanograms per gram. So in every gram of the sample of beluga fat from Pangnirtung in 2004 there are about 30 nanograms of PBDE. This is 0.000 000 03 grams of PBDE in every gram of sample, or 0.03 parts per million (ppm). This may seem an extremely small amount, but PBDEs are potentially very toxic to liver and thyroid function, and have been shown to hinder development of nerve tissue in mammals. For this reason, the European Union banned several of them in 2004 and then more in 2008. The migration of PBDEs into humans and shellfish can be explained by proximity to where they were used. While it is relatively simple to see how PBDEs can get into subjects close to their source, the PBDEs that end up in some of the wildlife in the Arctic have to be physically transported there. You will look at how pollutants are transported to the Arctic by flows around the Earth later in this course, but before you do, the following section looks at how pollutants can accumulate in the environment. 1.2 Pollutants and bioaccumulationThe term ‘pollutant’ is a very wide-ranging term. When the introduction or action of something into any environment causes harm, it is considered a pollutant. This could be a harmful chemical such as smoke from a chimney, or it could be a more subtle and transient effect such as floodlights at an evening football match preventing stargazing. There are many examples of how society has responded to pollution, such as the removal of lead in petrol, which affected human health, or the banning of chlorofluorocarbons (CFCs), which damaged the ozone layer. In both of these cases (i.e. lead and CFCs), when the pollution source was removed, the levels of them in the environment reduced and consequently so have the effects – albeit with a time delay. By definition, persistent pollutants such as PBDEs do not break down, so continued introduction of even minute levels into an environment leads to accumulation and perhaps magnification of potential harm. For example, at a landfill site the PBDE level is likely to increase with time. Animals around that landfill may ingest PBDEs directly, but this bioaccumulation (intake and concentration of the chemical in their tissues) may be so small that it does not cause problems to any particular animal. However, a predator such as a cat might eat dozens of rats that live around the landfill, so it would receive the combined dose that each of these rats had within it. If this dose were subsequently absorbed by the cat, then the resulting accumulated level could be significantly more harmful. This concentration of pollutants at higher levels in the food chain is called biomagnification, and the result is that higher predators can be poisoned and suffer harm while animals at lower levels in the food chain are apparently unaffected. 1.3 Summary of Section 1In examining the European name of the polar bear there is an apparent contradiction between its common and scientific names. This is because the polar bear is at home in the natural environments of land, sea and ice. In their search for food, bears can travel huge distances. Chemical analysis of the fat in the bears and their main prey species, the ringed seal, shows that they contain PBDEs – manufactured persistent organic pollutants that do not occur naturally. In Section 2 you will look at how different flows around the Earth can transport pollutants.2 The atmospheric and ocean flowsPBDEs end up in the Arctic through their physical transport by the winds, the ocean and the rivers of the world. All three mechanisms are important, but the most rapid carrier is the wind. The basic principle of global atmospheric circulation is simple: warm air rises and cold air sinks. The warming effect of the Sun is much greater at the equator than at higher latitudes, so the air is much warmer and rises. At high latitudes the air cools and sinks. This drives a horizontal wind. To help picture this, imagine a room with a radiator on one wall, and at the other end of the room an open fridge (Figure 4).
Figure 4 (a) A room with a radiator on one wall and an open fridge on the other will cause air to rise and sink at opposite ends; (b) horizontal winds are set up to replace this ascending and descending air
A schematic picture showing how horizontal flows are initiated by warm air rising and cold air sinking. There are two parts to the figure. Part (a) shows a cross section through a room. On the left wall of the room is a radiator above which warm air is rising. At the opposite, right, end of the room is an open fridge, adjacent to which cold air is sinking to the floor. Part (b) shows the same cross section some time later. The rising and sinking air has set up a clockwise circulation pattern around the room.The radiator heats up the air around it, and the air rises in what is called a convection current all the way to the ceiling and starts to spread. At the other end, the fridge is doing the opposite and cooling the air, which sinks and spreads across the floor. To replace the air that has risen, the air beneath the radiator is pulled upwards and then heated and rises, while the opposite is happening at the other end of the room. At the most basic level, on Earth the same process is happening, with warm air rising from lower latitudes and sinking at higher, colder latitudes. High-level winds therefore tend to blow from the hotter regions to the colder ones. This general pattern is modified by the rotation of the Earth, which deflects the wind flow away from the apparently direct path. These wind flows are further complicated by the distribution of continents and their mountain ranges across the globe. Winds are modified as they move around and over mountain ranges. They are also affected as they travel over land and sea surfaces, where the air is warmed to different extents. This is because of two additional processes: land and sea surfaces reflect different amounts of solar energy falling on them, and materials such as rocks and water need different amounts of heat to warm them up. You will look at the impact of these processes next.2.1 AlbedoWhen solar energy reaches the Earth’s surface, a proportion of it is reflected straight back out into space, and only the fraction which is not reflected heats the terrain. Different materials have a different albedo and so reflect a different amount of solar energy. If you put your hand on a black car on a warm sunny day, and then on a white car, you will notice that the black car feels warmer. This is because it reflects less energy so it heats up more. The black car has a lower albedo than the white car. Table 1 shows the albedos of some typical surfaces. For example, the surface of the ocean has an albedo of 3%, which means that 100% – 3% = 97%, or almost all of the incoming energy from the Sun, actually heats the water. Fresh snow, on the other hand, reflects away most solar energy, a property that has important consequences for the climate of the Arctic.
Table 1 The albedos of typical features on Earth
Surface
Albedo
Ocean surface
3%
Conifer forest in summer
9%
Grassy fields
25%
Sea ice
40%
Desert sand
40%
Fresh snow
80–90%
Activity 1 The importance of albedoAllow about 5 minutesIf the Sun’s energy falls on a desert and also sea ice on a frozen sea, what proportion of the energy is available to heat up each material? If snow then falls to cover the sea ice, what will be the amount of energy available to heat up the ice?If the Sun’s energy falls on a desert and a frozen sea, the amount of energy available to heat up the material will be the same, because Table 1 shows that the two substances have the same albedo: 40%. In both cases, the amount of energy available to heat up the material is So 60% of the incoming energy will be available to heat up the material. If snow falls on the sea ice, then its albedo will increase from 40% to 80–90%, so the amount of energy available to heat up the ice is Only 10% of the incoming energy is now available to heat up the ice, and almost all of the incident energy is reflected away. Clearly, albedo is extremely important for the polar regions. Next you will look at what is meant by specific heat capacity and its effects.2.2 Specific heat capacityWhen energy reaches the surface of an object, the amount the object heats up is determined by its specific heat capacity. This is a measure of how much energy it takes to raise the temperature of 1 kg of a particular substance by 1 °C. A lower specific heat capacity means that it takes less energy to heat up something, and vice versa. Although the term may be unfamiliar, the concept most likely is not. Activity 2 The effect of specific heat capacityAllow about 5 minutesOn a very hot sunny day on a table outside in the sunshine there is a glass containing 1 kg of water (i.e. 1 litre), a 1 kg piece of cork and a 1 kg piece of iron. Ignore the effects of albedo and assume that all three items absorb the same amount of energy from the Sun. Which will be the hottest after 1 hour, and which the coolest? (Ignore all sources of heat except that received directly from the Sun.) You probably recognised that the 1 kg of iron would be the hottest. It does not take very much heat energy to change the temperature of the iron because it has a low specific heat capacity. The other two items are harder to place, but the cork will be cooler than the iron, and the water, which has the highest specific heat capacity, will be the coolest item on the table. Water has an extremely high specific heat capacity and it takes a vast amount of energy to heat it. This is why virtually all car engines use water in their cooling systems. Taking into account the combined effects of albedo and specific heat capacity, even two adjacent areas such as a beach and the sea lapping on it will heat up by different amounts on a sunny day. Areas with lower heat capacities and lower albedos heat up more. This heat is transferred to the air above, so in these areas it will rise at a faster rate, whilst in cooler areas the air sinks. The rising and sinking air drives horizontal winds much as in Figure 4, although on a planetary scale. Sea ice cover is also constantly moving. It is pushed by the winds and ocean currents, and drifts in the pattern shown in Figure 5.
Figure 5 The mean ice drift across the Arctic Ocean. The ice is trapped in two major circulation features, the Beaufort Gyre and the Transpolar Drift Stream. White arrows show the general movement of the ocean currents; blue arrows show the general drift of the sea ice.
A cropped stereographic map of the Arctic Ocean surrounded by landmasses. The North Pole is offset to the right of the centre. In the Arctic Ocean are two kinds of arrows. Large white arrows show the general circulation of the ocean. To the left of the North Pole, north of Canada, is a large clockwise circulation labelled Beaufort Gyre. Above this are two white arrows from the coast of Siberia directed across the Arctic Ocean and out of Fram Strait which are labelled Transpolar Drift Stream. The second kind of arrow is light blue and represents the movement of sea ice in the Arctic Ocean. Above the Beaufort Gyre the arrows show a clockwise circulation and above the Transpolar Drift the arrows show a flow of ice in the same direction and out of the Arctic Ocean. 2.3 Russian tree trunks in the ArcticIn Northern latitudes, the treeline is often used as a means to define the Arctic region. The ‘treeline’ is a physical boundary of altitude or latitude beyond which trees cannot thrive because of a combination of light availability and temperature that would prevent tree growth. Although the Arctic is north of the treeline, it is not unusual to find tree trunks in the Arctic regions. Figure 6 shows a Svalbard beach strewn with tree trunks.
Figure 6 A photograph of a typical scene on a Svalbard beach (Norway)
A colour photograph of a typical scene on a Svalbard beach. A mixture of sand and pebbles make up the beach with tree trunks scattered around the foreground, and some patches of ice and snow. The shoreline runs across the middle of the picture from left to right. There is ice floating in the sea and the mountains in the background at the top of the picture are rounded and smooth.How do the tree trunks get there? They are mostly Siberian fir trees (Abies sibirica), natives of the great forests of northern Russia. Tree trunks are carried out to sea in summer by rivers such as the Lena, Ob and Yenisei. Then they are frozen into sea ice and travel in two ocean currents called the Transpolar Drift Stream and the Beaufort Gyre. Eventually they reach the shores of Svalbard and Greenland. Dating of these tree trunks using carbon dating shows that some are several thousand years old. 2.3.1 Nansen and the voyage of the <i>Fram</i>Wood on the shores of Svalbard and East Greenland caused confusion to the first explorers. But when wreckage from a ship called the Jeanette was found on the coast of East Greenland in the late 19th century, the best environmental scientist of the age, the Norwegian Fridtjof Nansen (Figure 7(a)), had a eureka moment. Nansen knew that the Jeanette had sunk off Alaska on the other side of the Arctic Ocean and deduced that the wreckage must have been carried across the frozen sea by the sea ice. He decided to try to use the ice drift to reach the North Pole and study the Arctic environment on the journey. He had the ship Fram (Norwegian for ‘forward’) built (Figure 7(b)). The ship had a round hull so that it would not get crushed like the Jeanette, and Nansen left Norway in 1893 for the Arctic and the North Pole. It was over three years before he and his colleagues returned.
Figure 7 (a) Fridtjof Nansen (1861–1930); (b) his ship the Fram frozen into the Arctic Ocean and being carried along with the moving ice in the Transpolar Drift Stream
Figure 7(a) shows a black and white photograph of the polar explorer Fridtjof Nansen wearing wolf skins. The image is cropped to just the face and upper shoulders. Nansen is in his early thirties with close-cropped blonde hair, and a heavily styled blonde moustache. He looks very lean and is giving a serious look, straight at the camera. Figure 7(b) shows a black and white photograph of Nansen’s ship Fram, frozen in the Arctic Ice and drifting in the Transpolar Drift Stream.Nansen and his crew followed the Russian coast (i.e. against the ocean currents), and the Fram froze into the sea ice off Siberia. As they drifted northwards, Nansen realised that the Fram was going to miss the pole so he and Hjalmar Johansen left the ship to make for the pole on foot. This was incredible. They knew the ship was drifting and they must have been certain that they would never find her again. The Fram survived the Arctic drift and reached Svalbard in the summer of 1896. Nansen and Johansen turned back just north of 86° N, having reached the highest latitude then attained. After an epic journey across the sea ice they endured the winter of 1895 on the island of Franz Josef Land and then caught a ship back to Norway, arriving only a few days before the Fram in August 1896 (see Figure 8). The sea channel between the Svalbard archipelago and Greenland was named the Fram Strait in honour of the famous polar research ship.
Figure 8 The voyage of the Fram (solid line) and route of Nansen and Johansen (dashed line) during their expedition of 1893–6
An old map of the Arctic regions showing the route taken by the Fram as it drifted through the Arctic. The map is a cropped stereographic projection of the Arctic with the North Pole in the upper left quadrant. The Arctic Ocean is coloured white which represents the sea ice. On the right of the map is the coast of Norway and Russia. From the coast of Norway there is a solid red line showing the route of the Fram on its voyage through the Arctic. From Norway, dated 21 July 1893, the ship sailed close to the coast of Russia to about 135° E where Fram was frozen into the ice on 22 September 1893 and started its drift across the Arctic Basin and towards the North Pole. On 14 March 1895 there is a dashed line which leaves the ship and represents the path of Nansen and Johansen. Fram continues across the Arctic in the Transpolar Drift Stream until it reaches Svalbard on 13 August 1896, and finally returns to Norway on 20 August 1896. Nansen and Johansen attempted to reach the pole before turning back from their pole attempt on 9 April 1895 and then headed south to spend the winter of 1895-96 at Franz Joseph Land, before returning to Norway on 13 August 1896.2.4 Pollutant pathways to the ArcticWinds, ocean currents and flow from rivers can all carry pollutants from their source to the Arctic. On a stereographic plot, the routes of wind-borne contaminants from the warmer, populated areas of Earth to the cooler Arctic are clear (Figure 9). These winds can transport contaminants to the poles, where they are removed from the atmosphere most likely through snowfall and are then absorbed by animals, perhaps through direct contact. The North Atlantic Current shown in Figure 9 flows directly past the waters off Western Europe, likely to be a major source of PBDEs. For top predators such as polar bears, there is also likely to be biomagnification from the high levels of PBDEs in their prey, the seals.
Figure 9 Transportation pathways for persistent organic pollutants (POPs) to the Arctic. Note the curving path of the wind currents caused by the rotation of the Earth. (adapted from Macdonald et al., 2005)
A map showing a stereographic plot of the Arctic regions. The North Pole is at the centre and the lowest latitude is 30° N. There are concentric lines of latitude at 40° N, 50° N, 60° N, 70° N and 80° N. The ocean regions are coloured blue and land is coloured green. Superimposed on this plot are large white arrows from low latitudes into the Arctic Basin in an anticlockwise spiral north towards the North Pole. These arrows represent the mean wind flow. Mean ocean currents are shown as smaller arrows. With the exception of through the Canadian Archipelago and along the East Greenland coast, virtually all ocean currents are into the Arctic Ocean. From the Russian and Alaskan land are purple arrows marking the flow of rivers into the Arctic Ocean. The size of these arrows represents the strength of the river and the largest are at the Yenisey, the Lena and the Ob in Russia.Overall, the toxicity of POPs to the polar wildlife is not completely clear, but the fact that they are manufactured only in populated regions and yet can be detected in Arctic wildlife is striking. POPs give a graphic demonstration that a region once thought of as remote is clearly physically connected to the rest of the planet. The poet Nick Drake responded to his experience of the Arctic by writing a series of poems. His ‘one poem in many voices’ The Farewell Glacier sought to give a voice to people, places and other animals and things related to the region.Listen to Nick reading two extracts from The Farewell Glacier, related to themes of the first two sections of this course. The first is about Wally Herbert (1934–2007), the British polar explorer, writer and artist. In 1968–9, Herbert led the British Trans-Arctic Expedition to walk 4000 miles from Alaska to Svalbard, making him the first man confirmed to have walked to the North Pole.
Video 1 Nick Drake’s Wally Herbert video
NICK DRAKE: When I was 12, to win a bet, I walked across the thin ice of the frozen Severn and never looked back. Later, I resolved to walk from Alaska to Svalbard across the sea ice. My Inuit friends left a map pinned to the door, marked with the places they thought I would die. It was 3,800 miles. We left in February. 4 men and 40 dogs. And in July, we made camp because the sea ice was not drifting in our favour. When the sun returned, we continued through the next summer to reach 90 degrees north. Trying to stand on the North Pole was like trying to step on the shadow of a bird circling overhead. I telegraphed the Queen. Two weeks later, a man took the first step on the moon, and by the time we got home, we were forgotten. You couldn't walk it now even if you wanted to. Why not? Because the sea ice is melting, and no one can walk on water.Now listen to Nick read his poem on pollutants and how they make their way to the Arctic.
Video 2 Nick Drake’s Mercury video
NICK DRAKE:We were born in your dream of the future. Released by fire, we ascended the winding stairs of the smokestacks until we reached the orange sunrise and the blue sky. No one waved goodbye. One saw us go. We were uncountable and invisible. One way or another, we were carried north in the hands of the winds on the wheels of the rivers by the generosity of the ocean. And when we arrived at the cold top of the world, it felt like home sweet home. And we waited in the long darkness until at last the first light of the year transmuted us out of thin air and we came to rest in ice and snow and black water. Now we accumulate and magnify in the cells of fish, in the eggs of birds, in the warm coats of seals and bears. And in the wombs of mothers, we concentrate so the faces of the future take on our features. And we sing our names into the ears of the unborn-- PCB, POP, DDT, magnesium, technetium, mercury. 2.5 Summary of Section 2Differences between the albedo and specific heat capacity of terrains mean that they heat up at different rates. Air in contact with the warm terrain rises in convection currents, and horizontal winds are set up across the whole planet. The winds can transport pollutants such as PBDEs to the Arctic, where they are deposited in snowfall and as a result can be detected in Arctic wildlife. Ocean currents and rivers can also transport pollution into the Arctic. 3 The ice time machineSnowfall differs depending on whether it falls in summer (when snow is comparatively warm and moist) or winter (when snow is cold and dry). These differences mean that when snow turns into ice, on the surfaces of glaciers and ice sheets, it is possible to see distinct annual layers. The layers are in a sense similar to tree rings: thick annual layers mean high snowfall, and thin annual layers mean low snowfall. The accumulation of snowfall on the Greenland and Antarctic ice sheets – and most importantly what is trapped within the crystals as it turns to ice – can provide a record of the past. Digging down into the ice cap is equivalent to going back in time through the snowfall of previous years and you have to dig down a long way (equivalent perhaps to 300 years of snowfall) before reaching the ice. To make the digging back in time easier, a drilling rig that extracts ice cores about 13 cm in diameter is used to get to very deep levels (Figure 10(a)). Once extracted, the annual layers in the cores are clear (Figure 10(b)).
Figure 10 (a) The NEEM (North Greenland Eemian Ice Drilling, where Eemian is the name of the last interglacial period) ice camp on the summit of the Greenland ice cap being dragged nearly 500 km to a new location to become EastGRIP (East Greenland Ice-core Project).
Figure 10(b) is a colour photograph of a section of a model of a Greenland ice core. There are obvious dark layers visible in the ice which mark successive annual layers of snowfall.
Figure 10(b) Annual layers in a model of a Greenland ice core. Light bands represent summer and dark bands represent winter.
Figure 10(b) is a colour photograph of a section of a model of a Greenland ice core. There are obvious dark layers visible in the ice which mark successive annual layers of snowfall.The British Antarctic Survey (BAS) is world renowned for its polar research, including analysis of ice cores. Video 3 visits the BAS research laboratories in Cambridge, UK where Liz Thomas, head of ice core research at BAS, explains how ice cores can provide a time capsule of past snow falls that record what past atmosphere and climates were like.
Video 3 British Antarctic Survey (BAS) and its polar research
TAMSIN EDWARDS: Climate science is a global endeavour involving scientists of every kind from every nation. The British have a long history of polar exploration and science, and much of it is carried out at the British Antarctic Survey. Hello, I'm Dr. Tamsin Edwards. I'm a lecturer in environmental sciences at the Open University. And I'm a modeller. I use computer models to study environmental change in the past and the future. Today I'd like to explore how it is we know what we know about climate change. How do we do scientific research? How do we use the data that we collect? So I've come here to the British Antarctic Survey, which is one of the institutes that has really key research into climate change to find that more. LIZ THOMAS: Here at the British Antarctic Survey, we conduct a range of research centred around climate change. So we investigate the atmosphere-- looking at the air-- and we investigate oceans-- oceanography. And we also look at the ice. In this country, when it snows, if you're lucky, you get to make a snowman. And then it melts. But in Antarctica you don't get that. So each year, the snow will build up. Year on year, you'll get it. And so you build up with these huge, great ice sheets. And what we can then do is we drill the ice core through this. And it's like going back in time. So we get a time capsule which can actually record what the Earth's atmosphere and what the climate was like at the time when that snow fell. And we can do this over years. We can drill down just shallow cores. Or we can actually drill back hundreds, thousands, and even close to a million years. So one of the things that we particularly focus on here is actually looking at the chemistry of the ice cores. The chemistry can actually give us indications of what's happening in the sea ice, can tell us about the atmosphere, so we can see how the atmosphere gets dustier during the winter, as there's more storms, and less dusty during the summer periods. And we can also look at the bubbles trapped in the ice. And this is particularly interesting, because that catches records of the Earth's atmosphere, particularly things like greenhouse gases-- carbon dioxide and methane. And we can not only see what the atmosphere was like at the time the snow fell, we can then take that backwards in time nearly a million years. So we're now in the Ice Core Labs, kept at minus 20. And this is where we see the majority of our work. When we bring the cores back from Antarctica or from the Arctic, we actually cut and subsample the ice here. So what I've got over here is actually one of the cores that we've drilled. And this is from a particularly deep core that we retrieved. And this is actually the bottom core. And we estimate that the age down at the bottom is 140,000 years. TAMSIN EDWARDS: So what have you over here then? LIZ THOMAS: So one of the really interesting things and really valuable things in terms of the ice core research is that we can actually get a record of volcanic eruption. So what we can see here, if you look up, we've got a sort of dark band, a grey mark. That's actually the volcanic eruption. So that's actually physically the ash from the volcano. And the really useful thing about that is that it allows us to date the core, because some of these big volcanic eruptions, we actually have historical records from the time that they erupted. And we can use this to not only provide the date of the core that we're drilling, potentially in Antarctica, but also some of these very big volcanic eruptions, the same volcanic signal will show in Antartica core as it would all the way up in Greenland. TAMSIN EDWARDS: And I see you've got some spare bits. LIZ THOMAS: These are just some offcuts. So as we come back and we process the core and we divide it up for samples, these are some of the chippings and some of the offcuts. And what I really want to see here is that you can actually look. And these tiny little white dots are actually the bubbles. So this you can see, visually, how the ice has been trapped into these bubbles. So when we talk about being able to look at how the atmosphere of the earth has changed, particularly these big important gases-- the greenhouse gases-- what we actually mean is we take this section of ice, you melt it, the air comes out of those bubbles. And then we have a record of what the atmosphere was like. TAMSIN EDWARDS: So you literally just melt the ice, and the bubbles come out, almost just like a fizzy drink. LIZ THOMAS: Exactly. Just like a fizzy drink. And actually we were standing outside, you'd start to hear them now, very much fizzing and crackling as air that's potentially been trapped in here for thousands of years becomes reintroduced to the Earth's atmosphere again. TAMSIN EDWARDS: We've heard about how ice cores provide a really amazing record of the past of our planet. I'd like to talk now to my colleague Mark Brandon about the present. Mark, you're a polar oceanographer. I know you go out into the field to difficult environments. Tell me the kinds of things that you're measuring. MARK BRANDON: So as a polar oceanographer, I'm interested in how the ocean is interacting with what I call the cryosphere. That's the frozen parts of our planet. So I've been out on ships and working on the frozen oceans in the Arctic and the Antarctic and making measurements around the continents. TAMSIN EDWARDS: So I'm a computer modeller. I use data like use in my work. I study Antarctica. This is a model of the bedrock underneath Antarctica. And I know that one of the areas you've worked in is down here in the Amundsen Sea area. MARK BRANDON: So I was on a ship that was working down at the edge of the ice front in the Amundsen Sea. And we were working out on decking conditions of about minus 20. And we were deploying this equipment, which measured the temperature and salinity of the ocean from the surface right the way down to the seabed. And what about enabled us to do is to work out how much heat is in the ocean, and how much of the heat is flowing towards Antarctica. And what we found was the heat from the ocean is responsible for melting about 10 metres a year. That's one of parts of Antarctica that's melting the most rapidly at the moment. TAMSIN EDWARDS: Now 10 metres a year might not sound like that much, but of course it all adds up. And I think the key to thinking about how science works is we put all of this information together. We have different kinds of data from around the world and different kinds of scientists. And we feed that into computer models-- like I use-- to try and make predictions. And what I'm particularly interested in is the uncertainty in those predictions. It's a huge scientific effort to put all this information together and try and work out the range of possible futures that we face. The next section shows you how ice cores are extracted and illustrates how data from ice core analyses can be used to help develop our understanding of past atmospheric conditions.3.1 Ice cores and the atmosphereAnalysis of ice cores collected in the Polar Regions can also tell us about how the climate has changed. Watch Video 4 which shows how scientists extract cores from the ice sheets and then saw them up for analysis in a laboratory.
Video 4 Ice core drilling
SUBJECT 1:We're at the corner of [INAUDIBLE].SUBJECT 2: Might stick a radio about him if he's got a laptop.[INDISTINCT CONVERSATIONS][WHIRRING]Up it goes![WHIRRING]SUBJECT 3:Well, that's a nice piece I cut, about two metres long. And around about here is 500-metres depth from the surface. And that's ice that fell as snow about 5,800 years ago.[WHIRRING]In addition to looking at snowfall, the use of different chemical and physical techniques to analyse ice cores can tell you about dust and pollen in the atmosphere, past volcanic activity, and even the industrial production of civilisations long past. For example, Figure 11 shows the concentrations of lead in the ice of different ages, and compares it with the recorded production of lead starting with the discovery of ‘cupellation’ (separating precious metals like silver from base metals like lead). Notice that the vertical axis of the lead production graph in Figure 11(a) is a logarithmic scale. Each successive tick mark up the axis has a value ten times bigger than the previous one. For example, 100 is equal to 1, 102 is equal to 100, and the tick mark between them is 101 (i.e. 10). A logarithmic axis enables changes over a large range to be compressed onto a small scale.
Figure 11 (a) Global lead production; (b) the concentration of lead in a Greenland ice core (years before present or ‘BP’) (adapted from Hong et al., 1994)
Figure 11(a) shows a graph showing total global lead production. The horizontal axis is time and it ranges from a minimum of 5000 years before present on the left to the present day on the right; the scale is non-linear. The vertical axis is lead production in tonnes per year in a logarithmic scale and ranges from 0 at its minimum to 106 tonnes per year at its maximum. At 5000 years before present the production of lead was first discovered and production is at the minimum. This is labelled as ‘discovery of cupellation’ on the graph. A second point on the graph, around 2700 years before present, is labelled ‘use of coinage’ for which the lean production is between 102 and 103 tonnes per year. This point coincides with the start of the period labelled ‘rise and fall of Athens’ which lasts until about 2300 years before present. There is a peak of lead production around 2000 years before present at around 105 tonnes per year. After this peak, at about 1800 years before present, the graph is labelled ‘exhaustion of Roman lead mines’. Lead production then falls relatively slowly to 1000 years before present at around 104 tonnes per year, before starting to rise again. At about 1000 years before present, the graph is labelled ‘silver production in Germany’. Lead production rises slowly at first until about 300 years before present, when about 6x104 tonnes per year were made. This point is labelled ‘Spanish production of silver in the New World’. After this, the production of lead increases rapidly to more than 106 tonnes per year at present time.
Figure 11(b) shows a graph showing the lead concentration in a Greenland ice core. The horizontal axis is time and it ranges from a minimum of 7760 years before present on the left to the present day on the right. The horizontal scale is non-linear although the times from 3000 years BP to present line up with the graph in (a). The vertical axis is lead concentration in grams per gram of ice core sample and ranges from less than 1x10−12 at its minimum to 4x10−12 grams per gram of sample at its maximum. There is a very low lead level of less than 1x10−12 grams per gram of sample up until 3000 years ago. The concentration then rises rapidly to a maximum of more than 3x10−12 grams per gram of sample around 2000 years before present which coincides with the exhaustion of the Roman lead mines. The lead concentration then falls back to very low values about 1800 years before present. The lead concentration then rises to reach a maximum of about 4x10−12 grams per gram of sample about 500 years before present.3.1.1 Powers of ten and scientific notationIn this study note you will look at how to write small and large numbers using scientific notation.Study note: Powers of ten and scientific notationFigure 11(a) shows the production of lead in tonnes (also known as metric tons) on a scale using different powers of ten (100, 102, etc.). When you see numbers written down, it is quite easy to read and understand them when they have few digits; for example, 0.01, 0.5, 4, 15 or 132. But when numbers have a lot of digits, for example, a small number such as 0.0000067, or a very large number such as 1 700 000 000, they are less easy to read, and consequently it is harder to understand what they are telling you. For example, if you are asked to say ‘75 kg’ you would probably respond immediately with ‘seventy-five kilograms’. But if you were asked to say the mass 330000000 tonnes, you would probably have to start counting the zeros. To make large and small numbers easier to comprehend, there are two options. One is to use the prefixes for words illustrated in Table 2 below. The other is to use numbers as in the final column of Table 2 which is labelled ‘Power of ten’, where the power is the number of tens that are multiplied together. For example, 102, which you would say as ‘ten to the power of 2’, means that two tens are multiplied together (i.e. 10 × 10). So 102 = 100.Similarly, ten to the power of three (i.e. 10 × 10 × 10) is103 = 1000.And so on. Clearly, 107 is easier to understand than 10 000 000. Note that 101 implies just one ten, that is, 101 = 10, so you do not add the power 1 in this case. When dealing with powers of 10 you could also just say that the power is the number of zeros after the 1, so 100 is just the number 1. That covers numbers greater than 1, but what about numbers less than 1 such as 0.1? In powers of ten this would be written as 1 divided by 10, so and this is written as 10–1. Similarly, 10–4 is 1 divided by 10 four times: So how would you write the number 150 using powers of 10? The number 150 is 1.5 × 10 × 10, so would be written 1.5 × 102. This form of writing numbers is known as scientific notation. A number written in scientific notation always looks like this: (number between 1 and 10) × 10some power.This superscript notation can also be used to show powers of units. For example:Square kilometres (for area):Metres per second (for speed): Square kilometres per year (e.g. for a change in area through time):
Table 2 Examples of different ways of writing large and small numbers
Prefix
Prefix name
Meaning
Number or fraction
Decimal
Power of ten
G
giga
billion or thousand million
1 000 000 000
1 000 000 000
109
M
mega
million
1 000 000
1 000 000
106
k
kilo
thousand
1000
1000
103
one
1
1
100
m
milli
thousandth
1/1000
0.001
10–3
µ
micro
millionth
1/1 000 000
0.000 001
10–6
n
nano
billionth
1/1 000 000 000
0.000 000 001
10–9
Sounds, seismic waves and starlight all have something in common: they are measured in powers of ten. Each can vary by so much that logarithmic scales are needed to describe the whole range. For example, a sound level of 110 decibels (dB) is 10 times louder than one of 100 dB. An earthquake of magnitude 8.0 has seismic waves that are 10 times larger than in an earthquake of magnitude 7.0. The brightness (‘apparent magnitude’) of stars is also measured on a kind of logarithmic scale. Activity 3 Powers of tenAllow about 10 minutesAround 66 million years ago an asteroid or comet around 10 km wide hit the Earth, creating the 180 km wide Chicxulub crater in Mexico and causing a mass extinction including that of the dinosaurs. The impact has been estimated as causing a magnitude 13 earthquake. In recent times, the fifth largest earthquake ever measured (at the time of writing) was the 2011 Japanese Tōhoku earthquake, which had a magnitude of 9. How many times larger would the seismic waves have been for the impact earthquake than the Tōhoku earthquake?3.1.2 Interpreting a graphGraphs can both reveal and conceal information. Read the study note below on how to interpret a graph, then complete Activity 4.Study note: Interpreting a graphIt is important that you look closely at the axes of a graph to make sure that you understand what is being plotted, and on what scale. Some graphs just show the general trend; others may show individual points, with or without connecting lines. Where connecting lines are drawn, as in Figure 11(b), the effect may be to lead your eyes to think that an isolated point is more important than it really is. The visual impact of a graph is both a strength and a weakness! Activity 4 Taking readings from a graph Allow about 10 minutesFrom the graph in Figure 11(a), what was the maximum global lead production in tonnes per year before the Industrial Revolution? When did this occur, and what was the lead concentration in the Greenland ice core at this time?
Figure 11 (repeated) (a) Global lead production; (b) the concentration of lead in a Greenland ice core (years before present or ‘BP’) (adapted from Hong et al., 1994)
Figure 11(a) shows a graph showing total global lead production. The horizontal axis is time and it ranges from a minimum of 5000 years before present on the left to the present day on the right; the scale is non-linear. The vertical axis is lead production in tonnes per year in a logarithmic scale and ranges from 0 at its minimum to 106 tonnes per year at its maximum. At 5000 years before present the production of lead was first discovered and production is at the minimum. This is labelled as ‘discovery of cupellation’ on the graph. A second point on the graph, around 2700 years before present, is labelled ‘use of coinage’ for which the lean production is between 102 and 103 tonnes per year. This point coincides with the start of the period labelled ‘rise and fall of Athens’ which lasts until about 2300 years before present. There is a peak of lead production around 2000 years before present at around 105 tonnes per year. After this peak, at about 1800 years before present, the graph is labelled ‘exhaustion of Roman lead mines’. Lead production then falls relatively slowly to 1000 years before present at around 104 tonnes per year, before starting to rise again. At about 1000 years before present, the graph is labelled ‘silver production in Germany’. Lead production rises slowly at first until about 300 years before present, when about 6x104 tonnes per year were made. This point is labelled ‘Spanish production of silver in the New World’. After this, the production of lead increases rapidly to more than 106 tonnes per year at present time.
Figure 11(b) shows a graph showing the lead concentration in a Greenland ice core. The horizontal axis is time and it ranges from a minimum of 7760 years before present on the left to the present day on the right. The horizontal scale is non-linear although the times from 3000 years BP to present line up with the graph in (a). The vertical axis is lead concentration in grams per gram of ice core sample and ranges from less than 1x10−12 at its minimum to 4x10−12 grams per gram of sample at its maximum. There is a very low lead level of less than 1x10−12 grams per gram of sample up until 3000 years ago. The concentration then rises rapidly to a maximum of more than 3x10−12 grams per gram of sample around 2000 years before present which coincides with the exhaustion of the Roman lead mines. The lead concentration then falls back to very low values about 1800 years before present. The lead concentration then rises to reach a maximum of about 4x10−12 grams per gram of sample about 500 years before present.The peak in global lead production before the Industrial Revolution was approximately 2000 years before the present (BP). At this point, the global lead production was about 105 tonnes per year. The concentration of lead in the Greenland ice core at this time was approximately 3 × 10–12 grams of lead per gram of ice. Extracting lead from its ores, and to a lesser extent working the lead into pipes etc. (the word ‘plumbing’ derives directly from the Latin for lead, plumbum, as does its chemical symbol, Pb) results in a discharge of lead-rich dust to the atmosphere. Given the pattern of wind movements shown in Figure 9, it is therefore not surprising that lead should appear in the precipitation over the Arctic for the corresponding period. 3.2 The past temperature of the planetMeasuring the concentration of lead in the ice is called a direct measurement: the ice sample is melted and the water produced contains a very small but readily measured quantity of lead dust. A very accurate set of scales is needed to measure it, but it is a directly measured quantity. There are also many indirect measurements that can be made using proxies. The concept for using proxies is both simple and brilliant: one measured property allows inference about other states of the system (Box 1). Box 1 Proxies and correlationThe word proxy is used in various settings to mean a stand-in: representing someone or something else. One example is a proxy vote, where one person agrees to represent the voting intention of another person in the voting booth. In science, the word ‘proxy’ is used when scientists measure one, two or even several direct quantities and use these values to infer some other quantity they wish to know. This is an indirect method of measurement. It is possible for measurements of one quantity to represent another quantity when there is a relationship between the two. You can say that the quantity is a proxy, and that the measurements of the quantity are proxy data. Take the following as an example:I measure my waistline, my weight and my height every week for a year, there will be a data set consisting of three variables measured 52 times over the course of a year. They are called variables because they are varying quantities; in this case, they vary with time. Typical results might be like those shown in Figure 12. Because I have stopped growing, my height does not change throughout the year so, as in the top panel of Figure 12, the graph is a flat line. However, both my waistline and weight do vary. With my body shape, when my weight goes up it all goes onto my waistline, so the graph of my waistline and the graph of my weight vary in the same way. As my waistline gets bigger, I get heavier. The opposite also applies – when my weight goes down, my waistline reduces. Because my waistline and weight seem to vary together, you say the two variables are correlated. In this case, they are positively correlated because when my waistline gets bigger, so does my weight. If, for some strange reason, as my waistline got bigger my weight decreased (not a likely scenario!), then the two variables would be said to be negatively correlated.
Figure 12 Schematic measurements of height, waistline and weight for the author throughout a year
A schematic of three graphs showing characteristics of the author over a year. In the top graph the horizontal axis is time measured in weeks and the vertical axis is height. In the middle graph the horizontal axis is time measured in weeks and the vertical axis is waistline, and in the bottom graph the horizontal axis is time measured in weeks and the vertical axis is weight. There are no units on any axis. The height of the author does not change over the year and so the data on the top graph is a flat line parallel to the horizontal axis, i.e. the value for height does not change. The data on the middle graph (the author’s waistline) and bottom graph (the author’s weight) do vary throughout the year and they vary together. For the first quarter of the timescale both the middle (waistline) and bottom (weight) graphs show similarly increases. For the second and third quarters of the timescale both plots show a decreases. Finally, for the last quarter, both plots show increases again. The three graphs show that although the author’s height does not change over the timescale, the author’s waistline and weight vary proportionally.Because my waistline is correlated to my weight, there is a mathematical relationship between the two variables. So, for example, it might be that when my waistline increased by 2 cm, I was 1 kg heavier. If I just gave you the data for my waistline over a year, and my starting weight, you could derive values for my weight over the whole year. This makes my waistline a proxy for my weight. If I then told you that I tended to eat more over Christmas and exercised a lot in the summer, then you could think it reasonable to add dates to the graphs in Figure 12. My weight and waistline would then be a proxy for the time of year as well. It is important to understand that correlated variables do not tell you anything about the cause of the observation – they only tell you that the items vary in a particular way. In the example above, clearly the expansion of waistline is not the cause of weight changing – it is the result of it. A more extreme example of this is that the number of people in the British armed forces has decreased since the First World War, and at the same time global atmospheric temperatures have risen. While these two variables are negatively correlated, there is no physical mechanism for one influencing or controlling the other. So, just because two things are correlated it does not necessarily mean that one causes the other, although in the case of the lead data there is an obvious causal link. What is perhaps not so obvious is that you cannot be sure just by looking at a graph whether two variables are correlated. To be sure that the observations do show correlation, scientists use formal statistical tests. The details of these are beyond the scope of this course, but they are essential in scientific investigation. In principle, statistical tests use mathematics to tell the likelihood that the results you see occur just by chance. If the mathematics suggest that the results are indeed just chance, you cannot draw any conclusions from them. If, however, the likelihood of it being just a chance relationship is very small, then you can assume that there really is some robust relationship between the two. To use one item as a proxy for others, you therefore need first to be sure that there really is a correlation, according to accepted scientific standards. Observing a correlation should also lead you to look for a plausible mechanism whereby one item affects the other. In the example of temperature and service personnel given above, such a mechanism is almost totally implausible. Even if the correlation were statistically acceptable, its implausibility would lead a scientist to reject it as being due to chance. Activity 5 Proxy variablesAllow about 10 minutesDo the data in Figure 11 suggest that lead production and the concentration of lead in ice cores are correlated, so that one could be used as a proxy for the other?
Figure 11 (repeated) (a) Global lead production; (b) the concentration of lead in a Greenland ice core (years before present or ‘BP’) (adapted from Hong et al., 1994)
Figure 11(a) shows a graph showing total global lead production. The horizontal axis is time and it ranges from a minimum of 5000 years before present on the left to the present day on the right; the scale is non-linear. The vertical axis is lead production in tonnes per year in a logarithmic scale and ranges from 0 at its minimum to 106 tonnes per year at its maximum. At 5000 years before present the production of lead was first discovered and production is at the minimum. This is labelled as ‘discovery of cupellation’ on the graph. A second point on the graph, around 2700 years before present, is labelled ‘use of coinage’ for which the lean production is between 102 and 103 tonnes per year. This point coincides with the start of the period labelled ‘rise and fall of Athens’ which lasts until about 2300 years before present. There is a peak of lead production around 2000 years before present at around 105 tonnes per year. After this peak, at about 1800 years before present, the graph is labelled ‘exhaustion of Roman lead mines’. Lead production then falls relatively slowly to 1000 years before present at around 104 tonnes per year, before starting to rise again. At about 1000 years before present, the graph is labelled ‘silver production in Germany’. Lead production rises slowly at first until about 300 years before present, when about 6x104 tonnes per year were made. This point is labelled ‘Spanish production of silver in the New World’. After this, the production of lead increases rapidly to more than 106 tonnes per year at present time.
Figure 11(b) shows a graph showing the lead concentration in a Greenland ice core. The horizontal axis is time and it ranges from a minimum of 7760 years before present on the left to the present day on the right. The horizontal scale is non-linear although the times from 3000 years BP to present line up with the graph in (a). The vertical axis is lead concentration in grams per gram of ice core sample and ranges from less than 1x10−12 at its minimum to 4x10−12 grams per gram of sample at its maximum. There is a very low lead level of less than 1x10−12 grams per gram of sample up until 3000 years ago. The concentration then rises rapidly to a maximum of more than 3x10−12 grams per gram of sample around 2000 years before present which coincides with the exhaustion of the Roman lead mines. The lead concentration then falls back to very low values about 1800 years before present. The lead concentration then rises to reach a maximum of about 4x10−12 grams per gram of sample about 500 years before present.Yes, they do appear to be correlated as the values rise and fall together. There is also a direct physical link between the two items, so it might be acceptable to use one as a proxy for the other. The example in Activity 5 shows that some measurements can be direct or a proxy, depending on the question of interest. If you wish to know about lead dust concentrations in Greenland ice in the past, you can measure them directly from the lead dust trapped in ice cores. If you wish to know about global lead production in the past, you can try to use measurements of lead concentrations in Greenland ice cores as proxy data, as long as you can estimate the relationship between the two. 3.3 Proxy data and past climatesThe process of analysis and checking for plausible mechanisms using proxy data has revolutionised the study of past climates. This is because many parts of the environment respond to climate: they change if the climate becomes warmer, or wetter, and so on. Wherever these changes are preserved, they serve as a record of the past climates. For example, the thickness of annual layers in an ice core is a simple proxy for moisture in the atmosphere at the time snow fell. This is because more snow forms and falls when the air is more moist. A thicker layer means more snow fell, so the atmosphere must have been wetter to form and hold the increased snow before it fell. A thinner annual snow layer would imply the opposite. Another type of proxy data from ice cores is the chemical composition of the water itself. Past Antarctic temperatures can be deduced from ice cores. Past temperature records have been constructed entirely from the relative amounts of oxygen-16 and oxygen-18 isotopes (see Study note: the central part of an atom). Water molecules in the ice have a proportion of all three isotopes of oxygen in them, and it has been shown that the relative amounts of the different isotopes vary depending on the temperature of the oceans at the time the snow fell. So scientists can measure the amount of oxygen-16 compared with the amount of oxygen-18 in an annual layer of an ice core to derive the temperature at that time. The ratio of the oxygen isotopes is a proxy for the temperature of the planet.You now know that proxy data measured in ice cores include:the concentration of lead, as a proxy for global lead productionthe thickness of annual layers, as a proxy for atmospheric moisturethe ratio of oxygen isotopes, as a proxy for temperature.There are also many other types of proxy than those found in ice cores. For example, the types of pollen found in ancient lake and ocean sediments are a proxy for the temperature and rainfall in the area at the time the plants grew.The great advantage of proxies is that they form a historical record of the planet, surviving from the past and giving information about things that cannot be observed directly. An important disadvantage is that proxy data are less accurate than direct measurements. This is because as well as measuring the proxy variable, scientists need to know the relationship between this and the variable of interest, which is an extra source of error.The following video sees poet Nick Drake read another of his poems from The Farewell Glacier. This one is about ice cores and how they can give a picture of the past. Nick is filmed reading in the ice core laboratory at the British Antarctic Survey, which is kept at a temperature of around minus 20°C.
Video 5 The ice core
NICK DRAKE:This is the library of ice. A high security auditorium of silence far below zero. An archive of cold that keeps me as I am and reminds me of home now that it is going, going. I am a long story, 10,000 feet long, 500,000 years old. A chronicle of lost time back to the first dark, too dark for telling. I am every winter's fall. I am the keeper of the air, of every vanished summer. I distill lost atmospheres pressed into ghosts kept close to my cold, cold heart. And as for you, what story would you like to hear on your two feet tracking the snow, two by two, two by two, two by two? Here is the dust and music of your brief cities. Here is the ash and smoke. Here are your traffic jams and vapour trails. Here are your holidays in the sun and your masterpieces and your pop songs. Here are your first cries and last whispers. Here is where it went right and where it went wrong. Easy come, easy go. So I know why you slice moon after moon from me, holding each fragile face up to your search lights while you measure and record the tiny cracks and snaps of my melting mysteries. Because you know you are the people who have changed nature and now you are on your own. I have no more to tell. No questions, please, about the future, for now the great narrator, silence, takes over. Listen carefully to her story for you are in it. 3.3.1 The central part of an atomIn this study note you will look at the central part of the atom, the nucleus, and isotopes which were discussed in the previous section. Study note: The central part of an atomThe central part of an atom, which makes up most of its mass, is called the nucleus; this is surrounded by an ‘electron cloud’, which largely determines how the atom reacts with other atoms or molecules. The nucleus of an atom is made up of building blocks called protons and neutrons. The number of protons determines what element the atom actually is. An atom with one proton is hydrogen, and an atom with eight protons is oxygen. However, the number of neutrons in the nucleus of an atom can vary. Oxygen exists in its natural state with eight protons and either eight, nine or ten neutrons. Atoms with the same number of protons but different numbers of neutrons are called isotopes. The most abundant oxygen isotope, with eight protons and eight neutrons, is called oxygen-16 (8 protons + 8 neutrons), the oxygen isotope which has eight protons and nine neutrons is oxygen-17 (8 protons + 9 neutrons), and the oxygen isotope which has eight protons and ten neutrons is called oxygen-18 (8 protons + 10 neutrons). 3.4 Ice core going back 800 000 yearsThroughout this course so far the focus has been on the Arctic, but because some data from the ice cores tell us about conditions over the entire planet (such as Figure 11), you will now look at data from another core, this time from Antarctica. The Antarctica ice cores go back much further in time than any Greenland ones. The particular core you will look at now is called the EPICA (European Project for Ice Coring in Antarctica) – Dome C core. Dome C is currently the longest ice core and has snow layers going back almost 800 000 years throughout the Quaternary, and includes the period when Homo sapiens evolved. In fact, the EPICA core can be used to reconstruct Antarctic temperatures more than half a million years before Homo sapiens ever walked the Earth (Figure 13).
Figure 13 (a) Antarctic temperature changes from the EPICA ice core from 800 000 years before present (BP) up to 1911. The vertical temperature scale has 0 °C for the mean temperature over the past 1000 years, and goes from −10 °C to +5 °C relative to this (Jouzel et al., 2007). (b) Map showing location of Dome C in Antarctica.
Figure 13(a) is a graph of temperature data inferred from the EPICA ice core going back 800 000 years. The horizontal axis is time measured in years before present (BP) and the minimum is 800 000 BP. The scale is linear and there are ticks every 100 000 years. The maximum on this axis is 0 which is the present day. The vertical axis is temperature change from the present day with a minimum of −10 °C and a maximum of 5 °C. A temperature change of 0 °C refers to the temperature of the present day which is defined as the mean temperature over the past 1000 years. The data on the graph is an irregular orange line which seems to oscillate between two states: one with a temperature close to the present day, and one approximately 8–10 °C colder. At the present day the temperature change is 3 °C, and at around 14 000 years BP the temperature change is about −10 °C. Around 130 000 years BP the temperature change is about 5 °C, but at around 140 000 years BP the temperature change is about −9 °C. At around 250 000 years BP the temperature change is about 3 °C, but at around 275 000 years BP the temperature change is about −9 °C. Around 340 000 years BP the temperature change is about 4 °C, but at around 345 000 years BP the temperature change is about −9 °C. Around 405 000 years BP the temperature change is about 3–4 °C but at around 450 000 years BP the temperature change is about −9 °C . Further back in time the temperature oscillations decrease in magnitude and the warm periods are colder than present by about −1 °C to −2 °C, whilst the cold periods are mostly around −8 °C. The final four cold periods are at times of around 550 000 years BP, 640 000 years BP, 720 000 years BP and 800 000 years BP. Figure 13(b) is an outline map of Antarctica, the landmass is white and a blue dot marks the location of the South Pole. To the South-east of the South Pole is a green dot marking the location of Dome C, the location of the EPICA ice core. The area surrounding Antarctica is shown in blue and represents the oceans: to the top left is the Atlantic Ocean, to the top right the Indian Ocean and the bottom left is the Pacific Ocean. In the bottom right is a scale indicating 1000 kilometres. Dome C is around 1700 km from the South Pole.Figure 13(a) shows that Antarctic temperatures have varied considerably, but there also appear to be regular cyclical patterns. At the low points, the temperature shown by the core was as much as 10 °C colder than today: colder periods happen about every 100 000 years, with warmer periods between. Four times in the last 450 000 years, the intervening warm periods have been warmer than today (up to 5 °C warmer around 130 000 years ago). During the nine cold periods shown in Figure 13(a), the snow that fell in winter did not melt in the following summer heat, and the ice sheets grew. 3.5 Interglacial periods and sea levelsThe EPICA ice core is a record of temperature variations in Antarctica, but what was happening in the rest of the world? Temperatures in other areas varied in a similar pattern of cycles. Other proxy data, such as from sediments found at the bottom of the oceans and lake beds, and the dating of rocks and analysis of ice cores from high-altitude mountain glaciers, show that during the cold periods a large proportion of the northern hemisphere was covered by an ice sheet that was, in places, several kilometres thick. Glaciers advanced, eroding valleys and mountains, and northern hemisphere wildlife moved south to more temperate regions. At the lowest temperatures the ice sheets covered about 10% of the entire planet – up to 30% of all the land. This meant sea levels were very much lower than today, so the area of exposed land was larger. Figure 14 shows how different the ice sheets and coastlines were.
Figure 14 The maximum extent of the ice sheets of the northern hemisphere during the 800 000 years of EPICA ice core data. Oceans are coloured dark blue and continents yellow. Ice is shown as lighter shades of blue.
A schematic map of the northern hemisphere. The oceans are coloured dark blue, and the continents are coloured yellow. On top of the landmasses and coloured in light blue are the maximum extent of the ice sheets over the last 800 000 years BP. All of Canada, Greenland and Iceland are covered in ice sheet. Virtually all of the UK and most of northern Europe is also covered in ice. A small part of Siberia and small part of Alaska remain ice free. The sea froze as far south as the northern Spanish coast, and almost all of Britain was buried beneath the ice. These periods are called the ice ages. A vast quantity of water was locked in these ice sheets, so sea level was as much as 120 m lower than today, and there was dry land between Britain and the rest of Europe. During times between these cold periods, the ice sheets melted and the water from land ice meant that sea levels rose. These are called interglacials. Note that it is only the melting of land ice that changes sea levels: melting sea ice does not change the sea level. Recall that ice is less dense than water, so it floats. As sea ice melts, it forms a smaller volume of water than the volume of ice. In fact, the volume of water formed is exactly the same as the volume of ice that was below the water surface when it was floating, so no change in sea level occurs (Figure 15). Of course, when ice on the land melts and flows into the seas, this does raise sea levels.
Figure 15 Floating ice does not increase sea level when it melts, because the volume underwater is the same as the volume of water when the whole piece of floating ice melts. Ice on the land does increase sea level when it melts, because it adds new volume to the ocean.
A simple colour diagram of a block of land ice resting on land and a block of floating ice in water shown side-on. An arrow points from the land ice into the water and this side of the diagram is labelled ‘Land ice raises sea level when it enters ocean’. The floating ice is partly submerged at the surface of the water. A dotted line box delineates the submerged part and this is labelled ‘Submerged volume of sea ice is equal to volume of water when sea ice melts’.What would happen to the size of the Arctic, as has been defined in this course, during an ice age?During an ice age, because the planet was colder and ice covered so much land, the treeline – our proxy for the Arctic definition – was much further south than today. This means that the area of the Arctic would have been much larger than at present. Activity 6 Rates of change of temperatureAllow about 10 minutesLook carefully at the Antarctic temperature record in Figure 13(a). Are there any general observations you can make about the rates of change of temperature between the relatively warm and relatively cold periods?
Figure 13 (repeated) (a) Antarctic temperature changes from the EPICA ice core from 800 000 years before present (BP) up to 1911. The vertical temperature scale has 0 °C for the mean temperature over the past 1000 years, and goes from −10 °C to +5 °C relative to this (Jouzel et al., 2007). (b) Map showing location of Dome C in Antarctica.
Figure 13(a) is a graph of temperature data inferred from the EPICA ice core going back 800 000 years. The horizontal axis is time measured in years before present (BP) and the minimum is 800 000 BP. The scale is linear and there are ticks every 100 000 years. The maximum on this axis is 0 which is the present day. The vertical axis is temperature change from the present day with a minimum of −10 °C and a maximum of 5 °C. A temperature change of 0 °C refers to the temperature of the present day which is defined as the mean temperature over the past 1000 years. The data on the graph is an irregular orange line which seems to oscillate between two states: one with a temperature close to the present day, and one approximately 8–10 °C colder. At the present day the temperature change is 3 °C, and at around 14 000 years BP the temperature change is about −10 °C. Around 130 000 years BP the temperature change is about 5 °C, but at around 140 000 years BP the temperature change is about −9 °C. At around 250 000 years BP the temperature change is about 3 °C, but at around 275 000 years BP the temperature change is about −9 °C. Around 340 000 years BP the temperature change is about 4 °C, but at around 345 000 years BP the temperature change is about −9 °C. Around 405 000 years BP the temperature change is about 3–4 °C but at around 450 000 years BP the temperature change is about −9 °C . Further back in time the temperature oscillations decrease in magnitude and the warm periods are colder than present by about −1 °C to −2 °C, whilst the cold periods are mostly around −8 °C. The final four cold periods are at times of around 550 000 years BP, 640 000 years BP, 720 000 years BP and 800 000 years BP. Figure 13(b) is an outline map of Antarctica, the landmass is white and a blue dot marks the location of the South Pole. To the South-east of the South Pole is a green dot marking the location of Dome C, the location of the EPICA ice core. The area surrounding Antarctica is shown in blue and represents the oceans: to the top left is the Atlantic Ocean, to the top right the Indian Ocean and the bottom left is the Pacific Ocean. In the bottom right is a scale indicating 1000 kilometres. Dome C is around 1700 km from the South Pole.The record in Figure 13(a) shows that the temperatures fall relatively slowly but rise relatively quickly – particularly in the most recent 450 000 years. Assuming (correctly) that the timing of Antarctic temperature changes is a proxy for the timing of changes in the amount of ice on the planet, the ice sheets in Figure 14 took about 100 000 years to grow, and yet they rapidly decreased in size – typically in only approximately 10 000 years. Consequently, sea levels fall slowly as the ice sheets grow, and rise relatively quickly as they decay again. The obvious question from Figure 13 is what causes these regular fluctuations in temperature and ice cover. One of the most influential is the Milankovitch cycles of the Earth’s orbit. You will look at this in more detail next.3.6 The Milankovitch model The amount of energy that the Earth receives from the Sun depends on its distance from the Sun. You tend to assume that this is constant, but in fact, the orbit of the Earth around the Sun is an ellipse – with the Sun at one of its foci (Figure 16) – so the distance from the Earth to the Sun varies over the course of an orbit (one year). If the Sun emits a constant amount of energy, then when the Earth is closer it will receive more than when it is further away.
Figure 16 The orbit of the Earth around the Sun is an ellipse, so throughout a year the Earth–Sun distance, and consequently the amount of solar energy received at the surface of the Earth, varies. Note that this picture shows the elliptical shape of the orbit greatly exaggerated.
A schematic plot (not to scale) of the Earth orbiting around the Sun. The shape of the orbit is an ellipse and the Sun is at one of the two foci. Hence as the Earth orbits the Sun, the distance between the two will vary.However, the shape of the ellipse also varies with time, and the Earth’s axis of rotation also wobbles, like a gyroscope. The Serbian geophysicist Milutin Milankovitch realised in 1920 that the varying energy received by the Earth as a result of these two factors could be the cause of the ice ages. Milankovitch showed that the ellipse changes shape over periods of about 100 000 years. The timing of these changes, combined with the wobble in the Earth’s rotation, matched up with data he had for the times and durations of the ice ages. He showed that the incoming energy would be at a minimum when there was an ice age and at a maximum during an interglacial. While his findings are important, modern records go back much further than the data to which Milankovitch had access, and further back in time the match is not so good. Earlier ice ages can be earlier and later than the predictions from the Milankovitch model. Clearly, there are other factors affecting the climate. You will see some of these other factors later in the course, but the differences are still not completely explained. This story illustrates another aspect of the way that science develops. The Milankovitch model was tested against new data, and found not to be fully consistent with it. The challenge was then for scientists either to completely reject that model, or to look for other effects that could be combined with the basic model to provide a better explanation of the observations. Scientific models are always subject to revision as new data are found. 3.7 The Keeling CurveThe Keeling Curve, illustrated in Figure 17, shows the trend in rising atmospheric CO2 concentrations since 1958 recorded at Mauna Loa in Hawaii. The story of atmospheric CO2 in those past 60 years is a relentless rise derived from human use of hydrocarbons: in 2008, the annual mean concentration was 383 parts per million (ppm), and eight years later it reached 400 ppm.
Figure 17 The Keeling curve: measurements from Mauna Loa of the monthly average carbon dioxide (CO2) concentration in the atmosphere, seasonally adjusted (Scripps, 2016)
This is a graph of the seasonally adjusted version of the ‘Keeling Curve’, that is it removes any seasonal fluctuations. It shows over fifty years of measurements, from September 1957 through to January 2016, of the concentration of carbon dioxide in the atmosphere, and charts its rise from 315 parts per million at the start of the period to just over 400 parts per million at the end. The curve rises smoothly and steadily over the fifty year period, but with a slightly higher gradient at the end of the period than at the start.When Charles Keeling first collected his CO2 data, he travelled around making the measurements at widely spaced locations – but he saw that apart from the daily and seasonal variation caused by local plant photosynthesis and respiration, the concentration was virtually the same wherever he measured it. Keeling quickly realised that this meant it was possible to measure the CO2 in one location, such as Mauna Loa, and it would be a reference point for the whole planet. Activity 7 How representative is the Keeling Curve?Allow about 5 minutesIs Keeling’s contention that the Mauna Loa data are a good reference for the whole planet consistent with what you have learned about atmospheric movements? Recall from the discussion of the spread of pollutants by wind (and from your own experience if you live in an exposed area!) that there are constant air movements around the planet. These movements stir up the air and mix it constantly. This constant mixing means that the concentration of CO2 is likely to be similar all over the globe. This sort of questioning as to whether methods and data are plausible is another good example of scientific method. 3.8 Ice cores and past CO<sub>2</sub> levelsAfter a few years of measurement, Keeling was the first to discover that CO2 levels in the atmosphere were rising, rather than emissions being absorbed by the oceans. The problem, of course, with the Keeling CO2 data is that they extend back only to 1958. However, ice core researchers realised that the air bubbles trapped when the ice was formed would contain atmospheric gas samples. As well as giving a proxy record of past temperatures, ice cores can give the exact atmospheric CO2 concentration for the last 800 000 years. Activity 8 Direct and proxy measurementsAllow about 5 minutesTo understand how past atmospheric concentrations of greenhouse gases have changed, are measurements of gas concentrations from an air bubble in an ice core layer a direct or a proxy measurement? Measurements of the greenhouse gas concentrations in a trapped gas bubble are direct data, not proxy data, because they are measurements of the actual quantities you wish to know about. It is perhaps surprising that it is direct, because it is a measurement of something that happened in the past! This is possible only because the actual quantity (gas) has been preserved (as bubbles trapped in the ice). This can be contrasted with, for example, measurements of the ratio of oxygen isotopes in the water, which are proxy data because the ultimate aim is to know the temperature of the planet. Here, only the proxy variable (isotope ratio) has been preserved (in the form of ice), not the temperature itself. It takes a certain period of time for the bubbles to be closed off and air to be isolated. As a result, it is not possible to measure the concentrations of gases until this has happened. In the case of the Dome C core, the most recent atmospheric CO2 concentration available is from around 100 years ago. 3.9 Global CO<sub>2</sub> levels and Antarctic temperaturesFigure 18 shows that over the last nine glacial cycles, the global CO2 and Antarctic temperature appear to be positively and very closely correlated, showing the same patterns of change.
Figure 18 Past Antarctic temperature changes (top) and global atmospheric CO2 concentrations (bottom) going back through nine ice ages, derived from the EPICA ice core (Luthi et al., 2008)
Two graphs of data from the EPICA ice core. The top panel is identical to the figure described in Figure 13, that is the temperature change inferred from the ice core over the last 800 000 years BP. The bottom panel shows how the atmospheric carbon dioxide trapped in the ice core varies over the same time period as the top panel. The two plots appear correlated, with warm temperature changes occurring at the same time as high atmospheric CO2 concentrations of typically around 280 ppm, and the cold periods occurring at the same time as low atmospheric CO2 concentrations of typically about 180 ppm. In a similar way to that described for Figure 13, the amplitude of the variation in atmospheric CO2 concentrations reduces in periods earlier than 450 000 years BP. The warm periods then have a maximum atmospheric CO2 concentration of typically 20 ppm less at around 260 ppm whilst the minimum is similar at about 180 ppm.Activity 9 Temperature and CO2 values Allow about 10 minutesAccording to Figure 18, what were the typical CO2 levels during the extreme low-temperature periods (ice ages) and at the height of the warmer interglacials? How does the value of the atmospheric CO2 concentration for 2016 (see Figure 17) compare with that in the interglacials of the previous nine cycles of the EPICA Dome C ice core? In an ice age, when the temperature is low, the CO2 is also low, typically 180–200 ppm. When the temperature is highest, in the interglacials, the CO2 is also high, at about 280 ppm for the most recent four interglacials and slightly lower at 260 ppm for the earliest five interglacials. The 2015 atmospheric CO2 concentration is around 400 ppm. This is about 120 ppm higher than in the most recent four interglacials, and about 140 ppm higher than the earliest five interglacials in the EPICA Dome C record. You now have data that you could possibly use to predict what might happen as a result of the increasing CO2 concentration that Keeling detected. You could theoretically plot a graph of temperature against CO2 concentration to highlight the correlation and, from this, read off the temperature for any given CO2 concentration. Unfortunately, there is a problem with this. The current atmospheric CO2 concentration is higher than at any time in the previous 800 000 years, so even if you had a graph of the mathematical relationship between temperature and CO2 concentration from the earlier data, it would not include the current (and much less, any possible future increased) CO2 concentration. You would have to extrapolate (that is, extend) the graph beyond the available set of values, and you do not know enough from this data alone to be sure that the relationship will hold outside these limits. This means that it is difficult to use information from these earlier periods to predict what may happen in the future. It is possible to be fairly sure that Milankovitch cycles amplified by greenhouse gases are at least partly responsible for the coming and going of ice ages; it is the best theory and one to which almost all climate scientists subscribe. But as you have seen, it is not a complete explanation, and some of the earlier cycles do not conform to this theory. To make useful predictions for the near future, and hence to suggest actions to protect the environment, you need to look for some more detailed information and more accurate scientific models. 3.9.1 Scientific methodMuch of science is concerned with gathering data, so a key part of scientific method involves scientists making observations or measurements about the world, and from these constructing theories about the causes of the observed phenomena. Using these theories or models, the scientist is then able to make predictions as to what might occur in a new but related situation to the ones previously observed. The scientist should then set up or seek such a situation, and test whether the observed behaviour does indeed occur. If it does, then the theory is supported. But if the observations do not accord with the theory, then the theory is either inadequate or possibly completely wrong. So a key part of scientific method is making testable predictions from the data. The philosopher of science Karl Popper (1902–94) was a major advocate of this approach, and brought in the concept of ‘falsifiability’. In essence, this suggested that a scientific theory would be useful only if it were possible to devise an experiment to test it, whose outcome could be in accord or not with the expectations. The results of such experiments may lead to the theory being rejected, revised, or accepted as possibly true until proved otherwise. Ideally, scientists should strive their hardest to disprove a theory rather than selectively only looking for the evidence that supports it! The story of Nansen’s expedition in the drifting ice is a spectacular example of scientific method. From the observation that trees from Siberia turned up in Svalbard, he predicted that a ship trapped in the ice would follow the same path. He then proceeded to test this theory in a very practical, but dangerous, way. The continual attempt to test, and potentially ‘falsify’ (prove to be false), theories is regarded as the essential feature of scientific method that distinguishes it from other approaches. An artist or a journalist may want to present their interpretation of a situation, but this interpretation is often only descriptive, not predictive. Some religions and similar codes make predictions and suggestions about what could or will happen, but these are rarely testable in a way that would be considered scientific. 3.10 Summary of Section 3The ice cores in Greenland and Antarctica currently provide a direct record of the snowfall going back around 800 000 years. As snow falls, impurities such as lead are trapped in the ice, so ice cores can give direct measurements of past atmospheric concentrations. By using isotope proxies such as oxygen-16 and oxygen-18, ice cores can be used to estimate temperature changes. Over the time period of the ice cores, the Earth has gone through nine cyclical temperature variations, with a cold period (ice age) approximately every 100 000 years. Trapped gases within the ice cores allow a direct measurement of atmospheric CO2 concentration, and throughout the entire Dome C record you can see that temperature and CO2 are positively correlated. Milankovitch cycles caused by variations in the Earth’s orbit, amplified by greenhouse gases, are the best current theory for the cause of ice ages, but these do not provide a sufficiently accurate model to predict the near future course of atmospheric change. 4 The end of the last ice age: the HoloceneAs noted earlier, the great ice sheets took about 100 000 years to grow and only about 10 000 years to decay. So what happened at the end of the last ice age? Figure 19 shows the EPICA ice core CO2 concentration and Antarctic air temperature for the most recent 20 000 years, which is within the last ice age. The temperature scale shows the difference from the average temperature of the last 1000 years, so 0 °C represents no change from (fairly) recent climate.
Figure 19 The global atmospheric CO2 concentration (relatively smooth line) and Antarctic air temperature change (spiky line) from the EPICA ice core over the last 20 000 years up to 1813 (CO2) and 1911 (temperature)
A graph of temperature and atmospheric CO2 concentration changes from the EPICA ice core over the last 22 000 years BP. The horizontal axis is time, measured in years before present (BP), and the minimum is 22 000 years BP with ticks every 2 000 years. The maximum on this axis is 0 which is the present day. The left-hand vertical axis is temperature change relative to the present day (mean of the past 1000 years) and the minimum is −12 °C and the maximum 4 °C, with ticks every 2 °C. There is a dashed line parallel to the horizontal axis at a temperature change of 0 °C (which refers to temperatures similar to today). The right-hand vertical axis is atmospheric CO2 concentration with a minimum of 165 ppm, a maximum of 305 ppm and ticks every 20 ppm. The temperature data shows a lot of variability compared with the atmospheric CO2 data. At 22 000 years BP the temperature change is around −9 °C whilst atmospheric CO2 is about 190 ppm. From about 18 000 years BP both the temperature and atmospheric CO2 rise until about 14 000 years BP when for a very short period of time the temperature change is −1 °C and atmospheric CO2 is as high as ~240 ppm. By 12 800 years BP the temperature rapidly falls to −5 °C below present whilst atmospheric CO2 remains approximately constant at around 240 ppm. This time period is labelled ‘Younger Dryas’. From about 11 600 years BP to 11 000 years BP the temperature and atmospheric CO2 again rise with the temperature change at around 0 °C and atmospheric CO2 about 265 ppm. There is another feature where the temperature and atmospheric CO2 fall at 8 200 years BP to −2 °C and 260 ppm, respectively. This is labelled as the ‘8.2 ka event’. From this time period onwards both temperature and atmospheric CO2 concentration increase. The CO2 data finish slightly before 0 years BP with a value of around 280 ppm and after this the temperature data show a rapid increase to around 3 °C above the mean of the past 1000 years. Figure 19 shows again the high correlation between the two variables: 20 000 years ago it was up to 10 °C colder in Antarctica, and global CO2 concentration was more than 200 ppm lower than today. Over the most recent 10 000 years the temperature was within about 2 °C of current temperatures, and this climatically stable time period is called the Holocene. Figure 18 shows that such a warm, stable period has been very unusual in the last 800 000 years, yet it is only during the Holocene that agriculture and the civilisations that rely on it have developed. Homo sapiens has flourished in the stable climate era. Figure 19 shows that up to approximately 14 000 years ago the planet appeared to be leaving the ice age, and Antarctic temperatures rose to within 1 °C of the 0 °C line. But then there was a very rapid cooling of 4–5 °C (and most of this in just a couple of decades), and lower temperatures resumed from 12 900 to 11 600 years before the present. This cold period affected most of the planet and is called the Younger Dryas, after a pretty Arctic alpine flowering plant called the white dryas (Figure 20). This species spread its geographical range as temperatures fell and the tundra biome expanded in area.
Figure 20 The white dryas. The Latin name of this flower is Dryas octopetala, meaning ‘dryas flower with eight petals’ – although it can have up to 16 petals.
A colour photograph of two white dryas flowers. The flowers are white with many petals and have a yellow centre.Another interesting event shown in Figure 19 happened just before 8000 years ago (called the ‘8.2 ka event’ where ‘ka’ is an abbreviation meaning 1000 years), when there was a definite but relatively small temperature and CO2 decrease which was associated with drier conditions in some parts of the world. This represents the largest climatic variation that civilisation has currently had to cope with. So what happened in the Younger Dryas and 8000 years ago to make the planet suddenly colder? The changes occurred too fast for the Milankovitch cycle to be responsible. It is now believed that the only way to cause that much cooling is by a sudden change in part of the global ocean circulation. Just as there are global patterns of air circulation, so there are also much slower, but enormous, movements of water around the oceans, driven by changes in water temperature and salinity, which you will look at next. 4.1 Wally Broecker’s great ocean conveyor beltThe density of fresh water decreases as its temperature rises above 4 °C. The density of salt water in the oceans likewise depends on temperature, but also on the amount of salt within it; saltier water is more dense. In the seas of the North Atlantic Ocean the surface waters are cooler and therefore denser than the lower layers. In places like the central Pacific Ocean the relatively dry, warm air increases evaporation and the surface waters are both warm and salty. All around the planet different regional climatic conditions create surface waters with different densities. Because denser waters sink, over time horizontal currents are set up similar to the processes for the winds. The result is a vast, three-dimensional circulation across the entire ocean. In the 1980s, the American climate scientist Wallace Broecker suggested that the global ocean circulation could be viewed as analogous to a conveyor belt that moved heat and salt around the planet. Broecker’s schematic picture (Figure 21) has become one of the iconic images of climate science.
Figure 21 A schematic of the great ocean conveyor that moves both heat and salt around the planet
A schematic picture of the global ocean circulation. There is a map of the world with land coloured green and the seas coloured blue. Across and joining all the oceans is a continuous band representing the continuous global ocean circulation. Warm upper waters are shaded red in the band and represent a flow from the Pacific Ocean into the Indian Ocean and finally the Atlantic Ocean, and also from the Pacific Ocean into the Atlantic Ocean through Drake Passage (south of the southern tip of South America). Once in the Atlantic Ocean the warm red current flows northwards past Europe where it cools, sinks, and changes direction to the south. The colour of the current changes to blue to represent cold, deep currents. It flows south beneath the warm current to the South Atlantic where is flows to the east before eventually returning to the Pacific and Indian Oceans where it is turned into warm currents again.It is a huge simplification, but on a global scale Broecker’s conveyor belt is excellent at helping to understand planetary processes such as the Younger Dryas and the 8.2 ka event. Heat that is carried in the ocean conveyor past Britain and up the coast of Norway towards Svalbard keeps the UK climate warmer and moister than it would otherwise be and means that the ice edge is a long way north compared with similar latitudes in North America, all year round (Figure 22).
Figure 22 Average Arctic sea ice concentration at the seasonal (a) maximum and (b) minimum from 1981 to 2010
Two maps showing the average sea ice concentration from 1981 to 2010 in March and September in a polar stereographic plot of the Arctic with the North Pole at the centre. Open ocean is shaded dark blue and the continents are shaded grey. The sea ice concentration is shown as white for high concentration through to medium blue for low concentration at its edges. The sea ice extent is far larger for March (winter sea ice maximum), where the sea ice covers the whole Arctic Ocean and slightly beyond, than for September (late summer sea ice minimum), where it covers only the central part of the Arctic Ocean. The ice edge is not linked to any particular latitude. This is particularly striking for March, where the Hudson Bay in North America is ice covered (about 51° N) but there is open water up to the north of Svalbard (about 80° N). 4.2 Stopping the ocean conveyor beltOne way to cool the planet, as occurred in the Younger Dryas or the 8.2 ka event, is to stop the ocean conveyor carrying the heat northwards. It is believed that this indeed happened as a result of large quantities of melt water from the North American continental ice sheets flooding into the North Atlantic and changing the surface density of the ocean. Once the conveyor was stopped, the climate was plunged into a cold period. Although similar events seem to have occurred further back in time, the Younger Dryas and the 8.2 ka events may have been particularly significant for human civilisation. The earliest dated farming settlements are in the Mediterranean about 13 000 years ago – around the time of the Younger Dryas. It is interesting to compare the spread of these settlements across the Middle East and Europe (Figure 23) with the temperature data of Figure 19. During the first 5000 years of human agriculture, from 13 ka BP to 8.4 ka BP, farming settlements are concentrated on the shores of the Mediterranean and Black Sea. However, after the 8.2 ka event and the collapse of the North American ice sheets, the flooding of fresh water into the Atlantic that stopped the conveyor also caused a rapid sea-level rise of around 1.4 m and large-scale flooding. After this date, the settlements rapidly spread northwards.
Figure 23 Locations and dates of sites of Neolithic farming settlements across the Middle East and Europe. The coloured dots indicate new sites that were established during each time period; grey dots represent pre-existing sites established during earlier time periods. The final panel shows the full time period (13–5.5 ka BP). (Turney and Brown, 2007)
This is a series of six Mercator maps of the Mediterranean region from North Africa in the south to southern Norway in the north, Britain, Ireland and Spain on the west to the Black Sea in the east. Each of the six maps are dated and have coloured dots representing the oldest dates that human settlements were established. In each successive panel the dots from the previous panels are shown in grey. Panel 1 is dated 13–11.5 ka BP and has fewer than ten purple coloured dots in the Levant region of the Mediterranean (modern day Israel, Jordan, Lebanon and Syria). Panel 2 is dated 11.5–10 ka BP and has approximately twice as many dots as the previous panel and coloured pink, in an expanded geographic range, but again only in the Levant. Panel 3 is dated 10–8.4 ka BP and shows more many yellow-coloured dots in the Levant but now there are approximately 10 dots on the east coast of Greece. The fourth panel is labelled 8.4–7 ka BP and shows an apparent sudden rapid expansion of human settlements. There are more than 100 green-coloured dots extending from the Levant all the way across mainland Europe, with one settlement in Britain. Panel 5 is labelled 7–5.5 ka BP and shows many more blue dots expanding across the whole of northern Europe as far as Britain and Ireland. The final panel is an amalgamation of the previous five panels giving the impression that human settlements have expanded from the Levant across Europe to Britain in the time period 13–5.5 ka BP.The exact driving factor for this human migration is impossible to determine, but it is interesting that it seemed to begin after the 8.2 ka climate event. This is an example of the kind of climate change that society will have to cope with. 4.3 Carbon dioxide (CO<sub>2</sub>), methane (CH<sub>4</sub>) and human activityIn recent decades, the understanding of the reality of climate change has moved from one of slow and gradual change over deep time to clear evidence that there have been naturally-occurring climate changes of several degrees Celsius and sea-level rises of half a metre within timescales of a decade or so (IPCC, 2013). In fact, research by Steffensen et al. (2008) on the Younger Dryas using ice core data revealed that central Greenland cooled by a staggering 2–4 °C in just 1–3 years, around ten times faster than the highest reconstructed rates of warming over the past two centuries (Box et al., 2009). While the Younger Dryas and the 8.2 ka event were entirely natural, there is an additional human contribution to consider. But when exactly did the human contribution begin? Often the phrase ‘pre-industrial levels’ is used to mean ‘before significant anthropogenic changes started’, but it is not specific. Could humans have influenced the climate before the Industrial Revolution of the 18th century? Another very significant greenhouse gas is methane (CH4), and 1 cubic metre of methane in the atmosphere can be over 25 times more effective at trapping heat than the same amount of CO2. Past atmospheric methane concentrations can also be directly measured from ice cores. Over the last quarter of a million years, CH4 concentration and the variation of solar radiation reaching the Earth attributed to the Milankovitch cycle are positively correlated (Figure 24(a)). But this correlation dramatically breaks down in the most recent data of the Holocene. The latest 5000 years of methane data show that the atmospheric concentration has risen dramatically out of synchrony with the solar radiation (Figure 24(b)). Three metres down in the EPICA ice core, dating to the 1820s, methane concentrations are as high as any period in the entire ice core record – approaching 800 ppb (parts per billion). Carbon dioxide has a similar break from the expected downward trend, although starting earlier, at about 8000 years ago. If the ‘normal’ trend of methane and carbon dioxide was downwards, along with the Milankovitch cycle, then where have the extra gases come from?
Figure 24 (a) The atmospheric concentration of methane, and changes in solar radiation reaching the Earth’s surface due to the Milankovitch cycles; (b) observed and expected atmospheric methane levels over the last 11 000 years up to 1937
This figure consists of two panels. The first panel shows the atmospheric concentration of methane and solar radiation from the Milankovitch cycle. The horizontal axis is time measured in years before present (BP) and the minimum is 250 000 years BP with ticks every 50 000 years. The maximum on this axis is 0 which is the present day. The left-hand vertical axis is methane in the EPICA ice core in parts per billion (ppb): the minimum is 300 ppb and the maximum is 900 ppb with ticks every 100 ppb. The right-hand vertical axis is solar radiation in watts per square metre: the minimum is 440 watts per square metre and the maximum is 540 watts per square metre with ticks every 20 watts per square metre. The panel has a blue-dashed line which shows the solar radiation and it is a smoothly varying curve cycling from about 440 watts per square metre to 510 watts per square metre on a period of approximately 25 000 years. An orange line shows the methane concentration in the EPICA ice core. Although it is not smoothly varying, methane concentration does cycle in a similar way to the solar radiation between the values of about 400 ppb and 600-700 ppb, such that high solar radiation corresponds to high methane, and low solar radiation corresponds to low methane. This is so throughout virtually the entire record, although this pattern is broken in the last 5 000 years, where methane levels do not reduce and instead rapidly increase to around 900 ppb. The second panel shows just methane concentration over the last 11 500 years. The horizontal axis is time measured in years before present (BP). The minimum is 11 500 years BP with ticks at 10 000 and 5 000 years BP. The maximum on this axis is 0 which is the present day. The vertical axis is methane concentration in parts per billion (ppb): the minimum is around 410 ppb and the maximum is 800 ppb with ticks every 50 ppb from 450 to 750 ppb. From 11 500 to 5 000 years BP the methane concentration in the core is on a downwards trend from approximately 700 ppb to 550 ppb. The trend is what would be expected from the previous panel and after 5000 years BP the trend is marked with an arrow continuing downwards reaching approximately 450 ppb at 0 on the horizontal axis. However, the methane concentration departs from this trend at 5 000 years BP and reverses direction to become a virtual mirror image of the previous 5 000 years, reaching greater than 700 ppb at 0 on the horizontal axis. So there is a departure from the expected trend of methane of the last 250 000 years.Carbon dioxide and methane are by-products of civilisation. You often think of these by-products as beginning with the Industrial Revolution, but their story begins far earlier. In 2003, climate scientist William Ruddiman proposed that society had been altering the levels of these gases – and therefore influencing global climate – for many thousands of years. In his own words: Human activities tied to farming – primarily agricultural deforestation and crop irrigation – must have added the extra CO2 and methane to the atmosphere. (Ruddiman, 2005, p. 48)Ruddiman suggests that these activities could have increased atmospheric concentrations by up to around 40 ppm for CO2 and 300 ppb for methane, increasing the expected natural levels by around 15% and 70% respectively (Ruddiman et al., 2016). The amounts of CO2 and methane added to the atmosphere – and therefore the degree to which human activities changed the climate – during the agricultural era are still being debated, but now, of course, the human effect on these gases is clear. As with CO2, since the Industrial Revolution the atmospheric concentration of methane has rapidly increased and currently is over 1800 ppb. Virtually all of that rise has been from anthropogenic sources, including major food production activities such as growing rice and cattle. Not only were humans possibly affected by climate change during the Holocene, but the impact by humans on the planet had already been started thousands of years before the Egyptian Pyramids were built. How then are these changes being seen today? Activity 10 Recent climatesAllow about 10 minutesHow does the Earth’s climate over the last 10 000 years compare with that of previous times, and what does this mean for humans in the future? Over the last 10 000 years, the Earth’s climate appears to have remained in a warm, stable state for longer than was normal in the preceding climate cycles. This has probably been important for humans in that they have been able to develop agriculture and other aspects of civilisation without the major disruption that would be caused by the major rapid cooling associated with ice ages. This means that when the Earth becomes warmer in future, humans may be less well adapted to the climate.4.4 Summary of Section 4Since the end of the last ice age, the climate has been uncharacteristically stable compared with the previous 800 000 years of the ice core record. This stable period is called the Holocene. The change from ice age to interglacial was not smooth, and there were two rapid cooling periods: the Younger Dryas and the 8.2 ka event. These are the most significant climate changes that humans have had to endure, and both have been linked to changes in the global ocean circulation. The 8.2 ka event coincided with the start of the spread of human settlement throughout Europe. By 5000 years ago there appears to be evidence of human influence on the composition of the atmosphere. 5 The contemporary Arctic climateThere is a remarkable seasonality in the Arctic climate. For example, the flow in some of the great rivers of Russia and North America that empty into the Arctic Ocean almost stops in winter (Figure 25). During May, ice in the rivers starts to break, and in June there is a rapid flood of fresh water followed by a fall in flow until November, when it freezes.
Figure 25 The monthly discharge on the Lena River (Russia). Each individual bar in the graph represents a monthly value for each year during 1935–99.
This is a graph of the discharge from the Lena River in Russia for the each month during 1935–1999. The horizontal axis is month of year from January to December. The vertical axis is monthly discharge from the river in cubic metres per second. The minimum is 0 cubic metres per second, the maximum is 120 000 cubic metres per second, and there are ticks every 20 000 cubic metres per second. In each month there is a series of 64 vertical bars representing the monthly discharge for each year in the period 1935–1999. From January to April there is almost no flow in the Lena and the bars are all less than 5000 cubic metres per second. In May approximately one quarter of the bars are greater than 5000 cubic metres per second, but the maximum discharge is still only around 30 000 cubic metres per second, and this only occurs in perhaps less than ten occasions in the record. In June there is a clear huge pulse in discharge and all of the bars are in the range 60 000–100 000 cubic metres per second. In July the discharge reduces to around 40 000 cubic metres per second, and in August and September it has reduced again to about 20 000 cubic metres per second. In October the discharge is typically about 10 000 cubic metres per second and finally in November and December the discharge is again less than 5000 cubic metres per second.A similar huge seasonal signal is seen in the Arctic sea ice cover (Figure 22). Most people are surprised to realise that the sea ice of the frozen Arctic Ocean is only a few metres thick. Beneath this are a few kilometres of water. In winter as much as 16 million square kilometres of the ocean freezes, and as this melts in summer, only about 6 million square kilometres remains frozen. The seasonal variation of almost 10 million square kilometres is equivalent to about 45 times the area of the United Kingdom. The contemporary Arctic climate appears to be changing. However, average global temperatures mask regional variations and the Arctic has been warming faster than the global mean (Figure 26).
Figure 26 The annual average near surface temperature from all weather stations on land relative to the average for 1961–90 for all regions from 60° N to 90° N (AMAP, 2012)
This is a graph of the average near surface temperature from all land stations from 60° N to 90° N. The horizontal axis is year from 1880 to 2010 and there are ticks every 10 years. The vertical axis is temperature change relative to the 1961–1990 mean in degrees Celsius: the minimum is −2 °C, the maximum is 2.5 °C and there are ticks every 0.5 °C. The zero on the vertical axis is the average temperature for 1961–1990 for all regions from 60° N to 90° N. There are blue dots representing the average temperature for each year. From 1900 to 1920 the points are all below 0 °C, from 1920 to 1960 almost all points are above zero, from 1960 to 1980 almost all points are below zero, and finally, from about 1980 to 2010 the points are mostly above zero and follow an upward trend.Activity 11 Recent climate change in the ArcticAllow about 10 minutesDescribe the changes in Arctic temperature that are shown in Figure 26.With the exception of a period in the 1960s and 1970s, the Arctic land temperature has been above the 1961–90 average in most years since 1920 and reached 2 °C above the 1961–90 average in the early part of the 21st century. These data came from the Arctic Monitoring and Assessment Programme. They appear to use the latitude of 60° N as their definition of the Arctic, so Figure 26 must include meteorological stations that are not in the Arctic as it has been defined in the course, which are less likely to be affected directly by changing ice and snow cover. For this reason, Figure 26 most likely underestimates the land temperature increase. Next you will look at the impact of land temperature increases.5.1 Melting ice caps and sea ice extentFigure 27 compares the surface melting on the Greenland ice cap in 1992 and 2005 as measured by satellite. For ice to form, the snow has to survive the following summer. But an increasing area of the Greenland ice cap is melting in summer, so annual snow layers are not being converted to ice in these regions.
Figure 27 A comparison of the surface melt of the Greenland ice cap in 1992 and 2005
A computer-generated map of Greenland as it would be if viewed from space from a region over north America. The ice cap of Greenland is mostly coloured white which represents ice, although around its edge there is a pink-shaded region which corresponds to the area of the ice cap that melted in the summer of 1992, and above this pink region is a red region which represents the area of the ice cap that melted in the summer of 2005. So from 1992 to 2005 the area of the ice cap that melts in summer has greatly increased. In addition there is a contour line showing the altitude on the ice cap of 2000 m. The red 2005 melting area is at a higher altitude than this.It is an extremely complex process to estimate the melt of the whole ice cap, and the current best value is that Greenland was losing around 160–270 billion tonnes of ice per year in the first decade of the 21st century. All of this melt is contributing to the predicted sea-level rise of around half a metre to a metre by 2100; a rise of a metre could affect around 150 million people worldwide (Anthoff et al., 2006). These impacts are primarily through increased flooding, rather than widespread loss of land: note the sea level changes predicted for this century are 10–20 times smaller than those of the much longer ice age cycles. The fresh water from the Greenland ice sheet could also slow Broecker’s conveyor (Section 4.1), causing other climate impacts. For the Arctic sea ice, the signal of climate change is clear: it is getting thinner and the amount of it that survives the summer is reducing. Figure 28 shows the trend in extent of sea ice in September each year from 1979–2016 (the summer minimum, of which the median for 1981–2010 is shown as a pink line in Figure 29). The sea ice minimum is decreasing at a rate of approximately 90 000 square kilometres each year. Near the start of the observations, in 1980, the September ice area was around 7.9 million square kilometres (white area in Figure 29(a)). In 2012, the September ice area was less than half this (3.6 million square kilometres, Figure 29(b)), though in subsequent years it recovered to some extent.
Figure 28 The minimum extent of Arctic sea ice in September of each year from 1979–2016
This is a graph of the minimum Arctic sea ice extent in September of each year from 1979 to 2016. The horizontal axis is year from 1975 to 2020 and there are ticks every 5 years. The vertical axis is ice extent in millions of square kilometres: the minimum is 3 million square km, the maximum is 9 million square km, and there are ticks every 1 million square km. The first data point is in 1979 and shows that for this year the minimum extent of Arctic sea ice in September was 7.2 million square km. There are points every year after this until 2016. Despite some variability, the general trend over the annual measurements is downwards, decreasing to 4.7 million square km in 2016. The lowest value is 3.6 million square km in 2012. A line of best fit is shown in orange through the points.
Figure 29 Arctic sea ice extent in September: (a) 1980; (b) 2012. These years have been selected to show the largest observed changes. The median September sea ice extent from 1981–2010 is shown as a pink line.
Two maps of Arctic sea ice extent in (a) September 1980 and (b) September 2012. The figure shows the physical landforms (grey) and ocean areas (dark blue) centred around the North Pole with the sea ice extent shown in white. Hence most of the oceanic Arctic region in the centre of the picture appears white. The sea ice extent and concentration are greater in September 1980 than September 2012. The median line for 1981-2010 is also shown, which encompasses most of the Arctic Ocean. For the 1980 the map the sea ice extent is slightly larger than the median line while for the 2012 map the extent is consistently smaller than the median. The figure has labels indicating the extent is 7.9 million sq km in September 1980 and 3.6 million sq km in September 2012.5.1.1 Gradient of a straight-line graphIn this study note you will learn how to calculate the gradient of a straight-line graph.Study note: Gradient of a straight-line graphFigure 28 shows a general trend of Arctic sea ice decreasing with time, though the annual fluctuations can be quite large. To measure this trend, a ‘best fit’ line is constructed as shown on the graph. This is drawn so that approximately the same number of data points lie above and below the line, but where there are significant fluctuations (as here) it may not pass through many or indeed any of these original points. The average rate of change of ice extent can be deduced by measuring the slope or gradient of this straight line on the graph. To do this, take two convenient points on the line and read off the values on each axis. These points should ideally be widely spaced (to improve accuracy) and will not necessarily correspond to original data points. In this example, the years 1980 and 2009 have been chosen, and the corresponding values on the vertical scale for the ice extent (according to the best fit line) are 7.8 and 5.3 million square kilometres. So the time interval is (2009 – 1980) years = 29 years.Change of ice cover is (5.3 – 7.8) million km2 = –2.5 million km2. (Note: the minus sign denotes a negative change, in other words a decrease.) This is easier to interpret if you convert millions to thousands (multiply by 1000), giving a mean rate of decrease of 90 000 km2 per year. This is the standard method for calculating the gradient of any straight-line graph, often summarised by the formulagradient = rise / run where the rise and the run are measured respectively from the change in values on the vertical and horizontal axis scales of the graph for the two chosen points. 5.2 Ice-albedo feedback loopYou have learned that the Earth’s albedo refers to the proportion of solar energy that reaches the Earth’s surface and is reflected straight back out into space. Section 2.1 explained how different surfaces on the Earth have a different albedo and so reflect a different amount of solar energy. Ice has a much higher albedo, and so reflects a much greater amount of solar energy, than the surface of the oceans. Activity 12 The changing mean albedoAllow about 10 minutesWhat is likely to be the effect of these changes in ice cover on the albedo of the Arctic region?Recall from Table 1 that the albedo of open water is 3% and that of sea ice is 40%. So the increased thawing during summer will decrease the albedo, so that less energy will be reflected back into space, and more energy will be absorbed. The effect on the albedo is actually more complex than suggested by Activity 12, but this ice–albedo feedback loop (Figure 30(b)) is potentially very important. Table 1 gives the average albedo of sea ice as approximately 40%. Sea ice is not uniform, and it could consist of a mixture of bare ice, ice with snow on (the snow could be either wet or dry) or even ponds of fresh water on the ice as it melts, and each one of these types has a different albedo. As temperatures rise there will be more bare ice, melt ponds and open water, and the overall albedo will decrease. This means that less energy will be reflected, so more solar energy is absorbed by the ocean, causing further warming and ice melting. The ice–ocean system is in a positive feedback loop, and changes such as melting ice naturally lead to more melting ice.
Figure 30 (a) Graph of the albedo of various ice categories and open water; (b) the ice albedo feedback look – an increase in absorbed sunlight leads to ice melting which lowers the albedo, causing more sunlight to be absorbed
This figure shows the ice–albedo feedback loop. On the left-hand side of the figure, part (a), a bar chart shows the albedo of different substances. The albedo of dry snow is about 82%, the albedo of wet snow is about 73%, the albedo of bare ice is about 65%, the albedo of new melt ponds is about 40%, the albedo of mature melt ponds is about 20% and the albedo of open water is about 6%. On the right-hand side of the figure, part (b), three boxes are joined together by arrows to form a loop. Box 1 is coloured yellow and labelled ‘increase in absorbed sunlight’. From this box an arrow points to box 2 which is labelled ‘melting of sea ice’. An arrow from this box points to the third box which is labelled ‘lowered albedo’ which then has another arrow back to box 1. By linking the three boxes and the albedos of the various ice components we can see that an increased melting will reduce the albedo which means more sunlight is absorbed, leading to further melting.5.2.1 Positive and negative feedback‘Feedback’ is the term used to describe the situation where the output from a process affects the input to that process. You may have encountered the ‘howl’ that can occur when a microphone is placed too near a loudspeaker; the sound from the loudspeaker feeds back to the microphone, gets amplified and is fed back again so that the volume of sound keeps on increasing until the amplifier overloads. This is an example of positive feedback. Populations of organisms can exhibit the same effect. If one generation produces more than one surviving offspring per adult, there are more organisms to produce young in the next generation, who produce more young in the next, and so on. This leads to a population explosion. Economic growth is supposed to work the same way – increased wealth this year allows us to spend and invest to produce more wealth next year, with this continuing year after year. Of course, the sound from the loudspeaker cannot get louder and louder forever, populations of organisms don’t actually go on expanding forever, and whatever economists may say, economic growth is unlikely to continue unchecked. The sound from the speaker is limited by the power available to the amplifier, and populations can be limited by their food supply. These limits can either have an effect like running into a brick wall, or be more subtle. The subtler version is the phenomenon of negative feedback, where an increase in the output from the process causes the process itself to ‘slow down’, so that output returns to a lower level. A room thermostat is a classic example. If the room warms too much, the thermostat reduces the central heating output to let the room cool to the correct temperature. Populations offer another negative feedback situation. When there are more organisms present, there is likely to be less food available per individual (or the increased population may attract more predators), so that the rate of production of young decreases (or the rate of mortality increases) and the population tends to stabilise. Negative feedback is a fundamental concept in the control of machinery and electronic devices, and there are many other examples from ecosystems. Note that the popular uses of ‘positive feedback’ and ‘negative feedback’ are praise and criticism, but the scientific meanings are not inherently ‘good’ or ‘bad’. 5.3 PermafrostClimate models suggest that, given the predictions of Arctic warming, the sea ice could disappear completely in summer by the middle of the 21st century. Given this and the changes you have observed in this course, you could not put the current situation any better than this: With sharply rising atmospheric greenhouse gas concentrations, the change to a seasonally ice-free Arctic Ocean seems inevitable. The only question is how fast we get there. The emerging view is that if we’re still waiting for the rapid slide towards this ice-free state, we won’t be waiting much longer. (Serreze and Stroeve, 2008, p. 143)The extent of snow cover in the northern hemisphere is decreasing in a similar way in another positive feedback loop, but what about the frozen ground beneath the snow that is called permafrost? Most of the global permafrost is in the Arctic and high mountain areas (Figure 31), and many cities use the frozen ground as foundations for building – and even for temporary roads in winter.
Figure 31 The permafrost distribution in the northern hemisphere. The largest area of continuous permafrost is in the Arctic and high mountain areas.
A polar stereographic plot of the Arctic regions. The North Pole is at the centre and the lowest latitude is 50° N. There are concentric lines of latitude at 60° N, 70° N and 80° N. The Arctic Circle (66.6° N) is shown as a dashed line. The oceans are coloured blue. The land is coloured according to one of five categories. With the exception of northern Scandinavia, all land within the Arctic Circle, including most of northern Russia, part of Alaska, northern Canada and the coast of Greenland, is coloured dark purple which represents continuous permafrost. Within Russia and northern Canada, continuous permafrost extends further south to below 60° N. South of these regions (with the exception of the non-coastal areas of Greenland which are covered by an ice cap) is a lighter pink region which represents discontinuous permafrost. Further south of this region and in most of Scandinavia is a yellow colour which represents sporadic permafrost. Finally there is a green region which represents isolated permafrost. Part of Iceland is shown as experiencing isolated permafrost.It should be expected that the area of permafrost will decrease, but it is difficult to measure. Virtually all boreholes into the permafrost show that Arctic warming (Figure 26) is penetrating into the ground. While frozen, permafrost provides a solid surface – a vehicle will leave no trace. As the permafrost melts, the situation is different. The State of Alaska has strict rules for vehicle travel on permafrost to prevent environmental damage. When it is too warm, travel is not allowed. The duration of allowed permafrost travel set by the Alaska Department of Natural Resources is an interesting climate change proxy (Figure 32)!
Figure 32 The annual duration of allowed winter tunda travel days set by the Alaska Department of Natural Resources, from 1970 to 2013
This is a graph of the annual duration of allowed tundra travel for oil exploration activities. The horizontal axis is year from 1970 to 2015 and there are ticks every 5 years. The vertical axis is days: the minimum is 0 days, the maximum is 250 days, and there are ticks every 50 days. The first data point is in 1970 when about 190 days were allowed. There are points for most years after this until 2012. Despite some variability, the general trend over the annual measurements is downwards. As well as the actual data point, there is a best fit line that shows the trend over time. The best fit line has a value of 203 days in 1970 and 120 days in 2013.5.4 Permafrost and tundra travel daysThe retreat of the permafrost is serious. Building foundations are collapsing, and there are ‘drunken forests’ as land beneath trees melts, subsides and slumps. Buildings require carefully built foundations, and the Trans-Alaska Pipeline was even designed with refrigerated pillars to prevent pipe fracture through permafrost thaw subsidence. Activity 13 Tundra travel daysAllow about 10 minutesFigure 31 (repeated below) shows the number of days on which travel has been allowed on the tundra (land with underlying permafrost is known as tundra). The best fit line has a value of 203 days in 1970 and 120 days in 2013. Estimate the average rate of change in the number of days over this period, to the nearest whole day per year.
Figure 31 (repeated) The permafrost distribution in the northern hemisphere. The largest area of continuous permafrost is in the Arctic and high mountain areas.
A polar stereographic plot of the Arctic regions. The North Pole is at the centre and the lowest latitude is 50° N. There are concentric lines of latitude at 60° N, 70° N and 80° N. The Arctic Circle (66.6° N) is shown as a dashed line. The oceans are coloured blue. The land is coloured according to one of five categories. With the exception of northern Scandinavia, all land within the Arctic Circle, including most of northern Russia, part of Alaska, northern Canada and the coast of Greenland, is coloured dark purple which represents continuous permafrost. Within Russia and northern Canada, continuous permafrost extends further south to below 60° N. South of these regions (with the exception of the non-coastal areas of Greenland which are covered by an ice cap) is a lighter pink region which represents discontinuous permafrost. Further south of this region and in most of Scandinavia is a yellow colour which represents sporadic permafrost. Finally there is a green region which represents isolated permafrost. Part of Iceland is shown as experiencing isolated permafrost.The number of days on which travel has been allowed on the tundra has decreased by an average rate of 2 days per year from 1970–2013.There is, however, another more worrying problem as the permafrost retreats. As the ground subsides, the depressions usually form lakes because the melt water cannot flow through the frozen ground beneath. Thawing of the permafrost at the lake bottom releases organic matter that is perhaps 30 000–40 000 years old into the water. The organic matter decomposes, giving off methane – a potent global warming gas (Figure 33). The permafrost–methane feedback cycle is another positive feedback in the system.
Figure 33 Researcher Katey Walter Anthony ignites trapped methane from under the ice in a pond on the University of Alaska, Fairbanks campus
This is a colour photograph of a scientist kneeling on ice next to a flame over two metres high that is emerging from a small vertical pipe inserted into the ice. She is looking at the source of the flame and is holding her hand back as if she had just lit the flame. The background shows evergreen forests and a darkening blue sky.5.5 Methane trapped in iceAnother potentially significant source of methane in the Arctic is trapped in the shallow seabed of the Arctic Ocean and is called methane clathrate (Figure 34). A clathrate is a lattice that contains other molecules, and methane clathrate is ice that has methane trapped within the crystal matrix. As the ocean warms, the release of large quantities of methane into the atmosphere from clathrates would be yet another positive feedback. This has been called the ‘clathrate gun hypothesis’ and it could lead to a strong amplification of the greenhouse effect that may have happened before in deep time. It could even have been responsible for previous mass animal extinctions.
Figure 34 Burning methane released from methane clathrate