The previous lecture provided the key ingredients for understanding the Greenhouse effect. The fact that solar radiation is concentrated near the visible light part of the spectrum, while Earth radiation is shifted to the infrared, relatively long wavelength band implies that there are differences in the energy input from above versus from below the atmosphere. The trace gases in the atmosphere selectively absorb the incident radiation, with gases such as carbon dioxide, ozone, and water preferentially absorbing infrared radiation. In fact, the atmosphere absorbs about 96% of all Earth radiation, with little escaping directly to space. The absorption of Earth radiation heats the atmosphere, which then radiates the energy both up and down, with some of the heat returning to the surface. If we increase the trace gas content of the atmosphere, by adding carbon dioxide, methane or water, we can enhance the efficiency of absorption of infrared radiation, further heating up the atmosphere and the surface. This is called the Greenhouse effect. (Nursery green houses are designed to let light from the sun in, but then to absorb reradiated infrared light to keep the interior warm.)
Carbon dioxide is the major Greenhouse gas, so we must ask what are the sources and sinks of this gas. What is essential is the carbon, so we must consider the carbon cycle in the Earth. Volcanic eruptions let out lots of carbon dioxide from the interior. This is how the atmosphere originally built up, and is why the atmosphere was initially much richer in carbon dioxide. There was a much stronger greenhouse in early Earth history, as well as a hotter planet overall. Plants that learned to use carbon dioxide for photosynthesis, cause reactions of CO2 with H2O, to give free oxygen plus CHO molecules, such as plant materials and sugars. This plant material is often buried in sediments and becomes part of the rock layer, with pressure and temperature turning some of the materials into coal and oil. This sequesters the carbon in the Earth, having effectively removed it from the atmosphere. Eventually some of the carbon would be recycled as the rocks were weathered and mixed into the oceans by erosion, eventually evaporating and entering the atmosphere again. However, human activity has greatly accelerated this process, and made it much more efficient. We dig up the oil and coal and burn it by recombining it with oxygen to once again produce CO2 + CO + H2O + heat energy. Combustion of fossil fuels thus returns carbon to the atmosphere.
Tracking CO2 content of the atmosphere over the past 40 years reveals a steady, 10% increase in the CO2 content, in large part due to the simultaneous effects of increased use of fossil fuels (increasing a CO2 source) and deforestation (reducing a CO2 sink). Very similar trends for other Greenhouse gases can be found, as for methane, which has almost doubled over the past 100 years, having been fairly stable for the previous 200 years.
If the increase in CO2 and other Greenhouse gases is increasing the efficiency of the Greenhouse effect, the planet could be warming up. One expected consequence of this would be that the surface ice volume should be reducing. A plot of sea level over the past 100 years that globally there has been about a 60 mm increase in sea level. This is a tough measurement to make, as any one local measurement can be influenced by local effects. But it does appear that there is a systematic increase. About half of the increase is due to melting of glaciers and about half is due to thermal expansion of the slightly warmed oceans which experience slightly higher average surface temperatures.
Overall, the present observations suggest that carbon dioxide is increasing, methane is increasing, sea level is increasing by about 1.2 mm/yr. In addition, the 5 hottest years in the 20th century occurred in the 1980's. Does this mean the Greenhouse is heating up? Are we seeing Global Warming on a rapid time scale. If so, will it lead to melting of the ice caps. That would have catastrophic effects, as sea level would rise about 80 m (240 feet). This would inundate major cities on coastlines around the world.
So, is it a done deal? Have we started heating up the planet and must now plan for the ice caps to melt? This is very hard to resolve, as the atmospheric system is a complex one, with many self-regulating mechanisms. Let's consider two of the feedbacks that keep a Greenhouse from running away.
As the CO2 goes up, the low atmosphere temperatures go up because of increased infrared absorption and reradiation toward the ground (sort of ping-ponging the heat in the near-surface environment). That makes the surface temperature go up, which causes increased evaporation. The higher water levels in the atmosphere in turn mean more cloudiness. More clouds means that more of the incident solar radiation is reflected, so surface heating from the sun goes down. Thus, the temperature may not actually rise, and a Greenhouse effect with clouds can actually lead to a cooler planet.
A more subtle effect is that the increased precipitation enhances rock weathering. As rain falls from sky water can react with CO2 to produce carbonic acid, which is very effective at leaching and eroding rock. The interaction of this acid with rock and clay can lock the carbon up in calcium carbonate materials that stay in the ground. Effectively, one can take carbon out of the atmosphere and lock it into the ground minerals by the increased precipitation. This can lower CO2 content and thus abate the Greenhouse effect.
The issue then becomes the time scale and relative efficiency of the competing effects. Humans are injecting carbon into the atmosphere in huge quantities very rapidly, far exceeding other processes other than large volcanic eruptions. Does increased cloudiness and precipitation intrinsically mitigate the effect, or do those retardants work too slowly to keep up with the pace of human activity? This is a major area of research.
The complexity of the dynamical atmospheric system is such that short-term responses are very difficult to infer, and we look to long term changes to try to understand the controlling factors that dominate in the system. This involves consideration of climate, or long term aspects of the atmospheric conditions. Recall that climate involves average properties such as temperature, pressure, and winds over a given time interval and region. There are short term, say annual, differences in climate which we call climatic variability. This is distinguished from longer time average changes, or climate change. It is recognized that climate is very erratic in space and time, with many scales, and that it has a complex response to a host of forcing functions. Studying past climates is the field of paleoclimatology, and this provides the time variations that we need to understand. Scientific study of climate variations is relatively young, and is itself evolving as we gain understanding of the problem. In a way, science is a process itself, slowly changing with time.
If we begin to examine long term climatic change, a good starting point is to delve into the current issue of Global Warming. This is linked to the Greenhouse and other effects, and is the concern that there is a secular increase in the temperature of the Earth's surface, either of human-induced origin or otherwise. This could constitute a great catastrophe, if melting of the ice caps occurs, desertification could occur and massive environmental changes would ensue. How do handle this question? Part of the approach has been mentioned above: monitor changes in Greenhouse gases, surface temperatures, sea level, etc. Another part is to study past climates to assess how much the system 'normally' varies, and to put any current changes in the context of what has gone before. Given that the current situation is of course different than any previous case, we hope to understand the system well enough to predict the expected effects of what will happen on a human time scale, and to effectuate whatever level of mitigation we can. Much of the debate is hard science, but there is also a major public policy element (how to manage forests, how to deal with changes in sea level, how to deal with potential desertification, etc.). The debate proves to be very complex at all levels.
We must pose and resolve varies questions, such as will the temperature increase everywhere, and how fast? Will cloudiness increase, and precipitation? Where? Will winds and sea level be affected? How do we measure global averages of these highly variable properties? Are there analogs in the past Earth record or on other planets? Can we develop numerical computer models that incorporate enough of the complexity that we can project future climates?
To approach the issue of past climates we must have measurements that tell us how conditions have changed with time. One approach that seems sensible is to look at large ice bodies, to see if they are progressively melting. Glaciers exist for hundreds and thousands of years, but either grow or shrink depending on temperatures and snow supply that sustain them. Growth of glaciers is comparatively hard to measure, but shrinkage is possible both by direct observations through time and by the geological deposits that are left at the leading edge of a retreating ice pack. There are several major glaciers in the Alps which have been tracked over 100-400 years, and they are all retreating. The Rhone Glacier has been receding since 1602. While still one of the 10 largest glaciers in Switzerland, during the Ice Age it had a length of 300 km and was the largest Swiss glacier. Since 1602 it has decreased in length by 2.5 kilometers. If the leading edge of the Rhone glacier is tracked as a function of time the curve followed by the retreating edge is very similar to that for other large glaciers in the French and Swiss Alps over the past century. This suggests a regional climatic effect, likely to be a warming trend. Similar measurements made elsewhere also suggest warming trends, and this corresponds well to the increase in sea level over the past century as mentioned above.
There is an active debate about whether Global Warming is actually occurring, or are these fluctuations all local effects. The majority opinion of scientists suggests that the Earth's surface has warmed by about 0.5 degrees +/- 0.2 degrees over this century, as a global average. The eruption of Mt. Pinatubo shaded the Earth enough that this might have been reversed over the past few years. There are great difficulties in measuring the global average, so this is a controversial area. The well-publicized attempt to track ocean temperatures directly by sending repeated sound waves through them is perhaps the most promising approach to a global thermometer, and should help to resolve the issue of whether there is actually a secular warming trend.
No matter what the result is, it will be important to assess whether a 0.5 degree temperature fluctuation is an unusual event or not. How much has the temperature fluctuated in the past? To test this, we must consider how paleoclimates are measured.
Reconstructing paleoclimates requires well-calibrated geologic thermometers. Examples are materials deposited in rocks that reflect local climatic conditions, of which the most important organic material is pollen. Bones of various species can indicate climatic changes, but provide less well-calibrated behavior than pollen. The most important geothermometer proves to be subtle geochemical components of the bones of organisms and rock deposits. Given a thermometer, it is critical to have good time control as well, so that the climatic change can be characterized. Once we have a record of climatic change at different positions around the world, we can consider the forcing functions, which control global climate. These are numerous, but the most important are orbital parameters of the Earth, atmospheric composition and Greenhouse state, and configuration of the continental plates which affect climate by their impact on atmospheric circulation and precipitation patterns. The effort to understand paleoclimate, and to model it quantitatively has demonstrated that the system is very complex, with coupled, nonlinear behavior that makes it difficult to isolate cause and effect. This lesson applies to the modern climate as well.
The types of geological thermometers differ in continental and oceanic rocks. On land we look for:
In the Oceans, we look at
For example, fossil pollen preserved in lake sediments reflects the windblown deposit from adjacent foliage around the lake. Variations in pollen influx as a function of time can indicate variation in the types of forest and vegetation in the region. Shifts from predominantly oak pollen to pine to spruce can indicate and calibrate the regional thermal variation of an area.
More subtle, and more important for global reconstructions, is the behavior of stable isotopes, particularly those of oxygen and hydrogen. Stable isotopes are variations in the atomic weight of a given element, involving the presence of additional neutrons in the nucleus. When the presence of the extra neutrons does not destabilize the material resulting in radioactive decay, the atom is simply heavier than other atoms of the same element, but all other chemical processes are unaltered. Oxygen has three isotopes: normal atoms with a weight of 16 (8 protons and 8 neutrons), and heavy oxygen with an atomic weight of 17 (8 protons and 9 neutrons) or 18 (8 protons and 10 neutrons). The number of protons determines the number of electrons (8 in both cases), and the degree to which the orbitals are filled determines the chemical reactions (the most stable state is to fill the orbitals with two shared electrons, thus oxygen bonds with 2 hydrogen atoms to make H2O water). Hydrogen itself has 3 isotopes (normal hydrogen with a simple proton nucleus, deuterium with a proton plus an neutron, and tritium with a proton plus two neutrons). All of these atoms are stable, and can combine to make water with mass ranging from 18 (the most common) to 24 (2 tritium atoms plus an 18O). The water behaves as water in each case, but the extra neutrons do add mass to the water molecules, so that if gravity is important in a process the different isotopes may behave differently.
Two other things are important to know about in order to understand the role of these stable isotopes. The first is that evaporation, in which solar radiation heats a water body, causing some water molecules to vaporize and enter the atmosphere as water vapor is in fact influenced by gravity. As a result the water that evaporates tends to preferentially be the lighter water molecules, relatively enriched in 16O versus 18O. The residual water then becomes enriched in 18O relative to 16O, unless the evaporation precipitates right back on the water body. If the evaporation occurs during a cold climate, there is potential for the 16O enriched water to precipitate on land as snow and be locked up into a glacier. Glacier ice is thus preferentially enriched in 16O (light oxygen) relative to the ocean. So, if we could monitor the 18O/16O ratio of the ocean as a function of time, we would see variations depending on whether glaciers are growing or shrinking. We would reference the 18O/16O ratio to a reference state to define an anomaly pattern. If the ratio is smaller (light), there is relatively more 16O in the ocean, which suggests little isolation of 16O in the glaciers, or an interglacial. If the ratio is larger (heavy), the lighter isotope is sequestered in the ice. So, how do we track the isotope content as a function of time.
The second important fact is that organisms living in the water (fish, diatoms, foraminifera, etc.) make their body parts out of the materials around them, including the water they are in. Thus, their shells and skeletons acquire the isotope ratio of the ocean in which they live. There are chemical effects of different isotopes entering into tissues, but these are fairly well understood and can be allowed for. When the organism dies, its shell or bone can settle into the sediments, preserving a time history as the sediment pile grows. If we dig up the sediments, then as a function of time track how the oxygen isotope content in the shells varies, we have a proxy for how the ocean isotope content has changed with time. This in turn reflects how glacier ice volume has varied with time. If glacier ice volume goes up, sea level goes down. Glacier ice volume varies with climate, so we have a means of tracking global climate change and sea level variations together!
Since we can go drill up ocean sediments that are as old as the oldest ocean plates (up to 190 million years or so), we can look at long time fluctuations in isotopes. If we look at the last million years, for which there are many observations, we find a remarkable pattern of oxygen isotopes varying back and forth between light isotope ratios as at present (fairly little ice pack and high sea levels) to heavier ratios, as was the case 20,000 years ago (large ice pack and low sea level). There are about 7 intervals, regularly spaced over the past 700,000 years where light ratios are found comparable to today, and about 7 intervals where heavy ratios are found comparable to 20,000 years ago. This remarkable record is a relic of the sequence of great ice ages which have regularly occurred with a 100,000 year period between them. The total fluctuation in ice pack appears to have been fairly uniform with sea level rising and falling about 150 m between peak glaciations and interglacials.
The fluctuation in sea level has actually shaped the very coastline on which we live. The up and down variations of sea level have been occurring during a period in which the coastline along the Monterey Bay has been slowly uplifting. The uplift is the result of the constraining bend in the San Andreas fault as it snakes through the Santa Cruz mountains (recall that the mountains are the result of convergence of crust into the bend, uplifting the surrounding region). With the coastline moving upward gradually, at a relatively constant rate controlled by plate tectonics motions, and the sea level rising and falling independently, a series of terraces have been cut into the ocean margin. During a period of high sea level, the ocean erodes the coastline, chewing into the margin, which is going up very slowly. When sea level drops, which it does fairly quickly at the onset of an ice age, the eroded platform is exposed, and the uplift carries it up. At the next high stand of the water, the ocean again cuts into the coast, but at a level below the last terrace because of the steady uplift. The result is a staircase set of terraces, such as the ones from downtown Santa Cruz up to Mission street (the hill on Laurel or Walnut), and then the one from Mission street up to campus. This terracing occurs in many places where there is a steadily uplifting coastline. In New Guinea there are about 9 or 10 strandlines exposed on the coast which correspond to past relative peaks in sea level over the last 300,000 years.
Why has the process not run away and completely covered the planet with ice, or alternatively melted all of the ice and stayed that way? This requires that we understand what causes the periodic ice ages of the past million years.
The last ice age ended about 18,000 years ago, but for the previous 50,000 years it is estimated that to sustain the huge ice pack of the time the Earth was 3-5 degrees cooler than now. That is about the range of thermal fluctuation the Earth has experienced over the last million years. It puts the 0.5 degree increase of the past decade into perspective: certainly we have not yet seen anything really drastic in the overall Earth temperatures, although the rate of change may be unprecedented. To understand the mechanisms responsible for a cyclical variation in temperature as large as 3-5 degrees, we analyze the detailed structure of the isotope record, determining the harmonic terms in the pattern. There are variations in the record with periods of 24,000, 41,000 and 100,000 years, meaning that the temperatures and ice volume had components of variation with each of those periods, with the strongest term being the 100,000 year term.
What about before that, does the periodicity of ice ages extend back further? Has this always been occurring? Using oxygen isotopes of older and older oceanic sediments, we can push the record back. Over the past 35 million years, the fluctuations are not much stronger than seen in the past million years, although localized intervals of low ocean temperatures correspond to events such as the first development of the Antarctic ice sheet about 10 million years ago, the onset of Northern Hemisphere ice sheets about 2 million years ago, and the first glaciers in Antarctica about 33 million years ago. About 5 degree fluctuations have occurred. But 35 million years ago, at the end of the Eocene there was an abrupt cooling, which lowered temperatures by about 5 degrees to the level of the last 35 million years. Prior to that the oceans were significantly warmed, and 50 million years ago the temperatures were fluctuating around a level that is about 10 degrees warmer than now. At that time there were no large ice sheets or mountain glaciers at all!
Pushing the record back further, we find that temperatures were hottest in the Cretaceous, about 100 million years ago, and there were high oceans which submersed about 17% of the currently exposed landmass. This included the large intercontinental seas of North America where dinosaurs thrived. Polar temperatures at the time were from 0-17 degrees C, compared to today's temperatures of -15 degrees C. On average the temperatures were 6-12 degrees warmer than now. Going back in time more, we see that the Cretaceous was a high point of the last 600 million years, and that 225 million years ago, during the Permian there were temperatures comparable to the ice ages, and the same near the onset of the Cambrian 600 million years ago. So, there have been fluctuations with time scales of several hundred million years, and overall temperatures have varied by something like 20 degrees on average. This places the Global Warming trend of today in a complex long-term context indeed.
So what causes climate variations with such a wide range of time scales? Of the mechanisms that we know of, plate motions will operate on tens to hundreds of millions of years (as continents move around and interact with atmospheric dynamics, for example one cannot build up large ice sheets unless there are continents at the poles). Long term fluctuations in carbon dioxide content associated with plant photosynthesis can also work on hundred million year time scales. The sun itself may vary in radiation output on many scales, of which we can only measure the shorter variations directly. These three mechanisms are likely to explain the longest term forcing functions affecting climate.
On the scale of 10-100,000 years, the orbital variations of the Earth appear to play a role, as was proposed by Mulitin Milankovitch. He identified 3 terms in the orbital variations which match frequencies in the spectrum of stable isotope variations. These are precession of the equinoxes, with a period of 24,000 years. The obliquity, or wobble of the tilt of the Earth relative to the plane of the ecliptic, which has a 41,000 cycle. An variation of the eccentricity or ovalness of the orbit, which has a 100,000 year cycle. The equinox are those points in the Earth's orbit when the northern and southern hemispheres are equally illuminated by the sun (12 hours of sunlight/night in both hemispheres the vernal and autumnal equinoxes, and maximum differences in daylight during the winter and summer solstices). This occurs at a slightly different position in the Earth's elliptical orbit each time, cycling through a range of positions each 24,000 years. The wobble of the tilt changes from 22 to 24.5 degrees, with the higher tilt causing more severe seasonal variations than the smaller tilt. The eccentricity controls the distance from the sun at the instant of perihelion (closest approach to the sun) and aphelion (most distant from the sun). These effects can all make small seasonal changes in the solar energy influx. The predicted effects are small, but apparently are magnified by the nonlinear coupling of the system to give complex cycles seen in the past million years.
On the shorter time scale of 1-10 million years there are variations involving volcanic emissions, which change particulate and trace gas content of the atmosphere. Finally, on the short time scale of decades, we come to human activities such as release of Greenhouse gases and deforestation.
All of this makes for a complex dynamic system which is changing with time independent of human activity. Will our behavior perturb the system from its current oscillations between glacial and interglacial periods, perhaps kicking it over into a runaway into the much hotter days like the Cretaceous. No one knows for sure, and there are many feedbacks in the system that make run away effects infrequent. But the catastrophe could be impressive if it happens.
Return to the Earth Sciences 2 Lectures Page