A brief introduction to... Ice Core Records (Part 2)

bldgblog.blogspot.co.uk; Photo by Planet Taylor

I didn't keep you waiting too long for the follow up to ice cores pt1, did I? As promised, today we will discuss the cycles of climate change throughout the Quaternary and what we think the reasons for this are. We may need to go off on a tangent on this post too, but it's worth it.

So we've established that we can use ice cores as a record of regional temperature through the Quaternary. But regional temperature is subject to many different variables. If only we had information about global temperature change? Well, the last post mentioned that we'd concentrate on just the polar ice cores. Isotope ratios from ice cores in both Greenland and Antarctica have been compared and it seems that they correlate relatively well, implying that whatever is forcing the climate in the Northern Hemisphere is doing exactly the same thing in the Southern Hemisphere, a global climate forcing. Now for the promised tangent.

We mentioned solar forcing. Cyclic changes in received solar radiation (known as insolation) are caused by the rhythmic variation in the Earth's orbit and axial tilt. These are known as the Milankovitch cycles, named after Milutin Milankovitch. The eccentricity of the Earth's orbit (how elliptical it is) fluctuates on a 100,000 year cycle and affects both the total solar radiation received in a year  (due to the change in distance from the sun) and seasonality (difference in temperature between Summer and Winter). Obliquity is the angle of the tilt of the Earth's axis. This fluctuates between 22.1 and 24.5 degrees on a cycle of 42,000 years and is another control on seasonality. If the North Pole is further away from the sun during Southern Hemisphere Summer, Northern Hemisphere Winter is longer and colder than if it were closer to the sun. Do you follow? The third cycle is the Precession of the Equinoxes. This occurs on a cycle of 26,000 year cycle and is the cycle at which the Earth's axis rotates about an imaginary axis perpendicular to Earth's orbit. The precession is caused by the combined gravitational forces of the Sun and the Moon and is another control on seasonality. For a really great tutorial on Milankovitch Cycles check out this link, it's a great introduction to the concept (you do need flash player I think).

How do these Milankovitch cycles relate to the fluctuations in the climate proxy records we see in ice cores? It is the current paradigm that these cycles in the relative position of the Earth to the Sun control the cycles in the Earth's climate from glacial conditions back to interglacial conditions.

The prevailing cycle in the climate proxy record is the 100,000 year glacial cycle. However, it has been noted by many researchers that the fluctuations in received solar radiation caused by the 100,000 year eccentricity cycles would not be enough to control the massive fluctuations in global air temperature seen between the interglacial and glacial climate extremes. Many summarise that it is probably the combined effect of the precession and obliquity cycles that control or trigger the 100,000 year climate cycle observed over the past 400-800,000 years.

There is however, a hitch. Work by Carl Wunsch analysing the temperature proxy record leads to the conclusion that most of the temperature variation over the past 400,000 years is indistinguishable from a stochastic (internally complex with unidentified feedback mechanisms) system and not a result of solar forcing. In my opinion, the jury is still out on this. Also, as highlighted in the work by Wunsch, an 800,000 year sample size isn't that big when you're only considering eight 100,000 year cycles.

So, make of that what you will! The best place to start reading about this is probably from the beginning, with the Hays et al. (1979) paper in Science, "Variations in the Earth's Orbit: Pacemaker of the Ice Ages". I know it talks about marine sediment cores, but the concept is exactly the same. It's definitely a great introduction to the concept of solar forcing of past climate.

There we have it, a brief introduction to ice core temperature proxies. Tomorrow I will discuss some of the other climate proxies we extract from ice cores: CO2, CH4 and dust flux, and what these tell us about the environments at the time.

A brief introduction to... Ice Core Records (Part 1)

Bad form, but source: wikipedia

As promised, for today and the next couple of days I will be writing about the ice core records of Greenland and Antarctica. I'll start today with an introduction to ice cores and discuss isotope signals within them. Tomorrow I hope to explain other proxies in the ice cores such as dust, CO2 and methane (CH4) and finish with a post about the controversies surrounding the conclusions based on proxies within the records, with the arguments against the validity of solar forcing (I'll get to that). So, I hope you enjoy exploring the ice core records. I've got some graphs I produced for my course to help explain the concepts, there are some garish colours in there (you have been warned).

Ice cores are simply a core of ice extracted from areas where annual accumulation of snow has resulted in a consistent record of winter precipitation over an extended period of time. These are obtainable from mountain glaciers and the polar ice caps, but I'm only going to focus on the ice caps here as these provide the longest records, and therefore a greater insight into past climates throughout the Quaternary. We know from a previous post that ice accumulation is the result of a greater snow accumulation than summer melt for any given year. Because of this annual accumulation - melt, annual layers are visible within an ice core, with clear layers signifying the melt season and cloudier layers indicating the winter accumulation. We also know that the ice will flow down slope from an ice centre in the form of glaciers. For the purpose of obtaining a complete record, it is very important to take ice cores from areas where lateral movement of ice does not occur, as this would distort or remove some of the record. Ice centres are perfect for this, however in polar regions, they are obviously extremely difficult to get to.

The three most well known ice cores from polar regions are the Vostok and Epica Dome C ice cores from the East Antarctic Ice Sheet, where ice is known to flow extremely slowly in comparison to the West Antarctic Ice Sheet; and the GRIP ice core chronology from Greenland.

I'm going to jump off on a tangent now and start talking about isotopes. Don't get scared, it's a simple concept based on the relationship between the ocean and temperature. As we know, snow is made of water, H20. The H part, hydrogen, normally consists of one proton and one electron. However, an isotope of hydrogen known as Deuterium exists, containing one proton, one electron and one neutron. This neutron effectively doubles the mass of the hydrogen molecule. Still with me? Deuterium is far less common than normal hydrogen on Earth, around 99.98% hydrogen to 0.015% deuterium. So imagine that the ocean is made up of mostly normal water, but a tiny fraction of this water is made with deuterium, and consequently slightly heavier than the rest of the water.

During the colder periods (ie. glacial maxima), there is less energy to evaporate seawater into the atmosphere. Because it takes more energy to evaporate the heavier deuterium water the clouds, and consequently the snow that lands on the ice caps, are made up of a higher ratio of normal water to deuterium water (more than 99.98% to less than 0.015%). Therefore, the ratio of deuterium to normal hydrogen in the snow for that year is less than normal. The opposite is true when temperatures are warmer, with more energy, a greater proportion of deuterium can be evaporated and is deposited on the ice caps as snow. So for warmer years, the deuterium to normal hydrogen ratio is greater than colder years.

Ok, so we're back onto ice cores again. As I mentioned earlier, each years' snow deposition is recorded in a thin band of cloudy ice. This is made up of the ratio of deuterium to hydrogen that was deposited as snow all those years into the past. When we compare these with previous years, we can see that the ratio fluctuates year on year, indicating regional temperature increases and decreases back through time. The Epica Dome C ice core goes back an astonishing 800,000 years and the data extracted from this ice core (and numerous others) is available for free to download as a spreadsheet at ncdc.noaa.gov/paleo/data.html. I find it amazing that the average Joe Public (like me) can download this data and manipulate it, graph it and make their own decisions about the climates of the past. The graph below is exactly that. It shows the deuterium ratio (delta D) for the Epica Dome C ice core. The X axis is age, youngest at the left and oldest at the right, with the deuterium concentration on the Y axis. As you can see, a pattern emerges. Sharp warming events are followed by a gradual cooling to glacial conditions.

Click image for full size

So, you've been introduced to the ice core records and the concept of isotope ratios. The use of oxygen isotope ratios are used in the same way (18O / 16O), but all you really need to know is that water made with oxygen 18 is heavier than the normal water made with oxygen 16, and the ratio of these two are used to indicate temperature in exactly the same way. Tomorrow, I'll discuss the cycles of temperature changes, how they relate to the amount of sunlight the Earth receives. On Thursday we will discuss the other measurable properties of ice cores including CO2, CH4 and dust concentrations, and what they are likely to mean.