Last week I wrote about how cyclical variation in Earth’s orbit influences the long-term climate here on the surface. I also left you on a cliff-hanger promising knowledge of how we know what we know regarding climate in the deep past. This week, I give you the answer: oxygen.
Atoms or isotopes?
A very quick bit of chemistry. Promise.
Oxygen exists as a gas in our atmosphere in molecules which consist of a pair of atoms chemically bonded together. Break that chemical bond symmetrically and you get a single atom of oxygen with 8 electrons orbiting a nucleus containing 8 protons and 8 neutrons (fig. 1). In the periodic table that means oxygen has a mass number of 16 (8 protons + 8 neutrons) and is written 16O. This is the most common form (99.76%), or isotope of oxygen, with the second most common isotope (0.207%) having a mass number of 18 (8 protons + 10 neutrons) and is written 18O. the remaining tiny percentage is 17O. If the number of protons changes, then the atom is no longer oxygen: 7 protons and it’s a nitrogen isotope, 9 protons and it’s a fluorine isotope.
A weighty topic
Crucially 16O is lighter than 18O by the mass of 2 neutrons. An oxygen atom is also present in each water molecule, H2O, which means that it can be written either H216O or H218O, with the two isotopes occurring in the same proportions as in atmospheric oxygen.
A difference of 2 neutrons may not seem like much, but when water evaporates from the surface of an entire ocean the heavier molecules prefer to stay as water (more energy is needed for them to evaporate) while the lighter molecules evaporate more easily to form water vapour. This process of Rayleigh distillation decreases the proportion of 18O in water vapour compared to ocean water.
Moreover, Rayleigh distillation also means that H218O rains out first. By the time the water vapour has travelled far enough to reach the poles or high mountain ranges it contains far less H218O than when it started. When the climate is colder, snow that falls here gets locked up in glaciers and ice sheets and doesn’t return to the sea. This means that the proportion of H218O in ocean water gradually increases as the climate cools into an ice age.
Snow or Smow?
The proportion of 18O in ocean water is assessed against a standard sample of ocean water known as (Vienna) Standard Mean Ocean water, or (V)SMOW, with the difference in proportions between the two samples expressed as δ18O (delta-18O). Positive values of δ18O indicate a cold climate. This is because there is more 18O in the ocean water than in SMOW thanks to Rayleigh distillation locking up 16O in glaciers and ice sheets.
Sea-ing is believing
This is all very well, but without a long-term record of what δ18O values were thousands and millions of years ago, how can we tell what the climate was doing?
Fortunately, microscopic plankton called foraminifera, or forams (fig. 2), incorporate both oxygen isotopes from ocean water in into calcium carbonate in their shells with no real preference for one over the other. This means that the oxygen isotope signature reflects that of the ocean water the foram was living in.
When these forams die, their shells fall to the ocean floor as sediment. Over millions of years this sediment builds up and is compressed to form limestone which contains records of isotope proportions over millions of years. Scientists can then drill out cores of this rock and calculate the δ18O of ancient forams compared to SMOW to work out the state of the climate in the deep past.
In fact, scientists have found that fluctuations in δ18O occur with cycle lengths of 400,000 years, 100,000 years, 41,000 years and 21,000 years as seen for the orbital cycles discussed last week. Over the last one million years, δ18O values have followed a roughly 100, 000 year cycle coinciding with every fourth of fifth precessional cycle (fig. 3), which demonstrates the modulation of precession by eccentricity (see last week’s intro).
More importantly, the timing of the cycles in the sediment cores lag behind the orbital cycles by a few thousand years, which allows time for the climate system to really feel the impact of orbital fluctuations.
Bonus factoid: cold water varieties of N. pachyderma have shells with a left-handed coil while warm water varieties coil in a right-handed direction. Drilling sediment cores from different latitudes and looking for left-handed coils can provide a much cruder but faster estimate of global ocean temperature compared to today by showing how close to the equator cold-water varieties were living.
When all is said and done however, δ18O is really only an approximation, a proxy, for global climate. Next week, I’ll look at some other proxies that represent shorter-term climate fluctuations. I may even write something about hominins at the same time!
National Oceanic and Atmospheric Administration (NOAA). Neogloboquadrina pachyderma. United States Department of Commerce, Washington D. C. Accessed: 16/10/14.
Ruddiman, W. (2008). Earth’s Climate: Past and Future, 2nd Edition, W. H. Freeman and Company.