Earth Observatory Home NASA Earth Observatory Home Data and Images Features News Reference Missions Experiments Search
NASA's Earth Observatory
 Earth Observatory Navigation Bar
Turn glossary mode on Features
Climate Close-up
 

by Holli Riebeek· design by Robert Simmon· May 9, 2006

From the oceans’ depths to the polar ice caps, clues to the Earth’s past climates are engraved on our planet. Sea sediments reveal how much ice existed in the world and hint at past temperatures and weather patterns. Ice cores also provide a glimpse of past temperatures and preserve tiny bubbles of ancient atmosphere. Coral, tree rings, and cave rocks record cycles of drought and rainfall. Each piece of this complex puzzle must be put together to give us a picture of Earth’s climate history. Scientists’ efforts to explain the paleoclimate evidence—not just the when and where of climate change, but the how and why— have produced some of the most significant theories of how the Earth’s climate system works.

 
  Photographs of proxies used to study paleoclimate
 

The Earth’s Shifting Orbit

From the scratched rocks strewn haphazardly across the landscape and the thin layer of soil left behind by retreating glaciers, scientists learned that the Earth had gone through at least three or four ice ages. Noticing that the ice came and went cyclically, they began to suspect that the ice ages were connected to variations in the Earth’s orbit.

The Earth circles the Sun in a flat plane. It is as if the spinning Earth is also rolling around the edge of a giant, flat plate, with the Sun in the center. The shape of the Earth’s orbit—the plate—changes from a nearly perfect circle to an oval shape on a 100,000-year cycle (eccentricity). Also, if you drew a line from the plate up through the Earth’s North and South Poles—Earth’s axis—the line would not rise straight up from the plate. Instead the axis is tilted, and the angle of the tilt varies between 22 and 24 degrees every 41,000 years (obliquity). Finally, the Earth wobbles on its axis as it spins. Like the handle of a toy top that wobbles toward you and away from you as the toy winds down, the “handle” of the Earth, the axis, wobbles toward and away from the Sun over the span of 19,000 to 23,000 years (precession). These small variations in Earth-Sun geometry change how much sunlight each hemisphere receives during the Earth’s year-long trek around the Sun, where in the orbit (the time of year) the seasons occur, and how extreme the seasonal changes are.

 

Sea-floor sediments, ice sheets, corals, cave formations, ancient trees, and alpine glaciers all hold clues to past climates. Scientists have assembled a coherent picture of the Earth’s climate history by combining data from all these and other sources. [Photographs copyright (left to right) Woods Hole Oceanographic Institute, Reto Stöckli, National Oceanic and Atmospheric Administration, National Park Service, Jessica Bray, and Robert Simmon.]

Diagram of orbital parameters

In the early 1900s, a Serbian mathematician named Milutin Milankovitch meticulously calculated the amount of sunlight each latitude received in every phase of Earth’s orbital variations. His work culminated in the 1930 publication of Mathematical Climatology and the Astronomical Theory of Climate Change. He theorized that the ice ages occurred when orbital variations caused the Northern Hemisphere around the latitude of the Hudson Bay and northern Europe to receive less sunshine in the summer. Short, cool summers failed to melt all of the winter’s snow. The snow would slowly accumulate from year to year, and its shiny, white surface would reflect more radiation back into space. Temperatures would drop even further, and eventually, an ice age would be in full swing. Based on the orbital variations, Milankovitch predicted that the ice ages would peak every 100,000 and 41,000 years, with additional “blips” every 19,000 to 23,000 years.

 

Three variables of the Earth’s orbit—eccentricity, obliquity, and precession—affect global climate. Changes in eccentricity (the amount the orbit diverges from a perfect circle) vary the distance of Earth from the Sun. Changes in obliquity (tilt of Earth’s axis) vary the strength of the seasons. Precession (wobble in Earth’s axis) varies the timing of the seasons. For more complete descriptions, read Milutin Milankovitch: Orbital Variations (Diagrams by Robert Simmon.)

Graphs of change in orbital parameters for the past 2.5 million years
 

The paleoclimate record shows peaks at exactly those intervals. Ocean cores showed that the Earth passed through regular ice ages—not just the 3 or 4 recorded on land by misplaced boulders and glacial loess deposits—but 10 in the last million years, and around 100 in the last 2.5 million years.

 

The Earth’s orbit varies over tens and hundreds of thousands of years. Combined changes in eccentricity, obliquity, and precession alter the strength and location of sunlight falling on the Earth’s surface. (Graphs by Robert Simmon based on data from Berger 1992.)

Graphs of solar insolation at 65 degrees north, and relative global temperature
 

Evidence supporting Milankovitch’s theory of the precise timing of the ice ages first came from a series of fossil coral reefs that formed on a shallow ocean bench in the South Pacific during warm interglacial periods. As the ice ages came, more and more water froze into polar ice caps and the ocean levels dropped, leaving the reef exposed. When the ice melted, the ocean rose and warmed, and another reef formed. At the same time, the peninsula on which the reefs formed was steadily being pushed up by the motion of the Earth’s shifting tectonic plates. Today, the reefs form a visible series of steps along the shore of Papua New Guinea. The reefs, the age of which was well-defined because of the decaying uranium in the coral, measured out the millennia between ice ages. They also defined the maximum length of each ice age. The intervals fell exactly where Milankovitch said they would.

 

(Upper graph) The rise and fall of the intensity of sunlight (insolation) in the far North during the summer—determined by the Earth’s orbit—drives ice ages. Weak summer sunlight year after year allows snow to accumulate and glaciers to advance. The reflective ice sheets further cool the Earth’s surface, resulting in global ice ages. When the Northern Hemisphere receives more sunlight, the snow melts, ice sheets retreat, and Earth warms.

(Lower graph) Oxygen isotopes trapped in ocean sediments record cycles of ice ages millions of years into Earth’s past. This climate record matches the frequency of orbital changes, although tangled feedbacks make the relationship complex. Dips represent ice ages, and spikes represent interglacials. (Graph by Robert Simmon, based on data from Berger 1992 and Lisiecki 2005.)

 

Explaining Rapid Climate Change: Tales from the Ice

 

When scientists started to analyze the paleoclimate evidence in the Greenland and Antarctic ice cores, they found that the record also supported Milankovitch’s theory of when ice ages should occur. But they also found something that required additional explanation: some climate change appeared to have occurred very rapidly. Because Milankovitch’s theory tied climate change to the slow and regular variations in Earth’s orbit, the scientific community expected that climate change would also be slow and gradual. But the ice cores showed that while it took nearly 10,000 years for the Earth to totally emerge from the last ice age and warm to today’s balmy climate, one-third to one-half of the warming—about 15 degrees Fahrenheit—occurred in about 10 years, at least in Greenland. A closer look at marine sediments confirmed this finding. Although the overall timing of the ice ages was clearly tied to variations in the Earth’s orbit, other factors must have contributed to climate change as well. Something else made temperatures change very quickly, but what?

 
Graph of temperatures from 20,000 years ago to present, based on data from the GISP2 ice core
 

Greenhouse Gases

Scientists are now exploring a few possibilities. First, greenhouse gases probably influenced past climates. Ice cores record past greenhouse gas levels. In the past, when the climate warmed, the change was accompanied by an increase in greenhouse gases, particularly carbon dioxide. When scientists tried to build climate models, they could not get the models to simulate past climate change unless they also added changes in carbon dioxide levels. Though scientists aren’t sure why carbon dioxide levels changed, almost all believe that the shift contributed to altering the climate. Because ice cores also revealed that carbon dioxide levels are much higher today than at any time recorded in the past 750,000 years, pinning down the cause-and-effect relationship between carbon dioxide and climate change continues to be a focal point of modern climate research.

Global Conveyor Belt

Another possible trigger for rapid climate change is ocean circulation. Today, warm water from the equator is carried towards the poles on ocean surface currents. Because of the arrangement of the continents, warm water is carried far into the North Atlantic, moderating the climate in Northern Europe. As the warm surface water reaches the cold air in the north, it cools. The salty Atlantic water becomes very dense as it gets cold. The cold, salty water sinks to the bottom of the ocean before it can freeze, where it is pulled southward toward the equator. More warm water from the equator flows north to replace the sinking water, setting up a global oceanic “conveyor belt.”

 

Rapid changes between ice ages and warm periods (called interglacials) are recorded in the Greenland ice sheet. Occurring over one or two decades, the warming of the Earth at the end of the last ice age happened much faster than the rate of change of the Earth’s orbit. The last cool period (stadial), immediately before the current interglacial, began and ended suddenly, and was likely caused by changes in the deep ocean circulation. (Graph by Robert Simmon, based on data provided by Alley 2004.)

  Map of thermohaline circulation
 

This pattern helps keep Northern Europe far warmer than other locations at the same latitude. The key to keeping the belt moving is the saltiness of the water, which increases the water’s density and causes it to sink. Many scientists believe that if too much fresh water enters the ocean, for example, from melting Arctic glaciers and sea ice, the water will be diluted. Fresh water freezes at a higher temperature than salty water, so the cooling surface water would freeze before it could become dense enough to sink toward the bottom. If the water in the north does not sink, the water at the equator will not move north to replace it. The currents would eventually stop moving warm water northward, leaving Northern Europe cold and dry within a single decade.

This theory of rapid climate change is called the “conveyor belt theory,” and though many scientists do not yet agree with it, the paleoclimate record found in ocean sediment cores is beginning to support it. Recent paleoclimate studies have shown that when heat circulation in the North Atlantic Ocean slowed in the past, the climate changed in Northern Europe. Although the last ice age peaked about 20,000 years ago, the warming trend was interrupted at various points by cold spells. In a paper published in Nature on April 22, 2004, McManus and colleagues Roger Francois, Jeanne Gherardi, Lloyd Keigwin and Susan Brown-Leger at Woods Hole Oceanographic Institute and the Laboratoire des Sciences du Climat et de l’environnement in France showed that cold periods in Europe 17,500 and 12,700 years ago happened just after melting ice diluted the salty North Atlantic water, and the ocean “conveyor belt” slowed. The evidence, which they took from radioactive elements in ocean cores, is beginning to support the theory, but McManus cautions that there are still pieces to fill in before we fully understand what role the conveyor belt played in past climate change and what role it might play in the future.

    References:
  • Alley, R. (2000). The Two-Mile Time Machine. Princeton: Princeton University Press.
  • Alley, R. (2004). GISP2 Ice Core Temperature and Accumulation Data. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series #2004-013. Boulder, Colorado: NOAA/NGDC Paleoclimatology Program.
  • Berger, A. (1992). Orbital Variations and Insolation Database. IGBP PAGES/World Data Center-A for Paleoclimatology Data Contribution Series #92-007. Boulder, Colorado: NOAA/NGDC Paleoclimatology Program.
  • Bradley, R. (1999). Paleoclimatology. San Diego: Academic Press, Harcourt Brace and Company.
  • Imbrie, J., and Imbrie, K. (1979). Ice Ages. Hillside, New Jersey: Enslow Publishers.
  • Intergovernmental Panel on Climate Change. (2001) Climate Change 2001: Synthesis Report. Geneva: World Meteorological Organization.
  • Lisiecki, L., and Raymo, M. (2005). LR04 Global Pliocene-Pleistocene Benthic d18O Stack. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series #2005-008. Boulder, Colorado: NOAA/NGDC Paleoclimatology Program.
  • Rahmstorf, S. (2002). Ocean circulation and climate during the past 120,000 years. Nature, 419, 207-214.
  • Weart, S. (2003). The Discovery of Global Warming. Center for History of Physics, American Institute of Physics.
  • Weart, S. (2003, August). The Discovery of Rapid Climate Change. Physics Today, 30-36.
 

The large-scale movement of water through the oceans, called the thermohaline circulation, plays a large role in the duration of ice ages. Dense, very salty (saline) water sinks in the North Atlantic, pulling the “conveyor belt” of currents behind it. The conveyor belt carries heat from the equator towards the poles, and raises Arctic temperatures, discouraging the growth of ice sheets. Influxes of fresh water from the lands that surround the North Atlantic can slow or shut down the circulation, cooling the Northern Hemisphere.

This map shows the general location and direction of the warm surface (red) and cold deep water (blue) currents of the thermohaline circulation. Salinity is represented by color in units of the Practical Salinity Scale. Low values (blue) are less saline, while high values (orange) are more saline. (Map by Robert Simmon, adapted from the IPCC 2001 and Rahmstorf 2002.)

   
Subscribe to the Earth Observatory
About the Earth Observatory
Contact Us
Privacy Policy and Important Notices
Responsible NASA Official: Lorraine A. Remer
Webmaster: Goran Halusa
We're a part of the Science Mission Directorate