First, you have to understand the way ice forms at the polar regions (Antarctica and Greenland, for example) is much different than what you've understood your whole life. Most of us think of the ice cubes in our freezer when someone says “ice”, especially if you grew up in Texas like I did. The ice in your freezer forms when you put water in its liquid form into the freezer. The air temperature in the freezer is below water’s freezing temperature and, boom, in a few hours you have ice cubes. The freezing temperature for the many liquids on Earth varies. For example, gasoline is a liquid. Many people don’t think about it freezing, but it can freeze. The reason we don’t is because gasoline’s freezing point is at temperatures most people never experience: around -40 to -50 degrees Celsius (-40 to -58 degrees F). Gasoline freezing is why airplanes cannot land at the South Pole in the winter when the temperatures plummet to -100 degrees F.
Back to water. Most places in Greenland and Antarctica stay below water’s freezing temperature of 0 degrees C (32 degrees F) for most of the year. Although, many coastal areas do go above freezing. South Pole is ALWAYS colder than 0 degrees C. Therefore, any precipitation that falls is in the form of snow. Because it never gets above freezing, this snow does not melt. When it snows again, this snow falls on top of the older snow and piles up. If you’ve held snow, you know it’s very fluffy and light. This is because the crystalline structure of snow (aka a snowflake) allows for a lot of air or void space to be present between two snowflakes. If you squeeze that fluffy snow together in your hand, you reduce the air or void area between the snowflakes. The same idea goes with the snow at the polar areas. Year after year, snow falls, it does not melt (this is not the case at all places, but a general statement) and it piles on top of the last year’s snow. Eventually, the weight of the snow above causes the void space between the snow crystals to reduce until there is no longer any void space and the snow is now ice. This process occurs over many years.
Scientists call the snow that is above the ice at the polar areas, firn. Porosity is also another term scientists use to describe the amount of open/void spaces in a material. More porous materials have more air pockets or open spaces. So, firn is the porous snow (see the diagram). As we move deeper into the ice sheet, the firn becomes less porous and eventual turns into the ice at a zone called the “lock-in” zone. At this lock-in zone, the air bubbles are locked into the ice. Before, because the firn is porous, the air can move freely around. Therefore, in the upper most part of the ice sheet, the air in the firn is still exchanging with the atmosphere. Once locked into the ice, the air bubbles no longer exchange with the atmosphere. That is how ice core scientists can determine the concentration of gases in the ancient atmosphere.
The time it takes for firn(snow) to turn to ice depends on the amount of snowfall or accumulation. Ice forms at a certain density (mass per volume). In order to reach that density, snow must be compacted year after year. Places that have more snowfall/accumulation per year will have a shallower firn column – or will reach the ice zone sooner (in terms of depth). For reference, a location that has high snowfall is the West Antarctic Ice Sheet Divide. This location receives about 22 cm per year of snow that will become ice. We call that ice equivalent accumulation because that is how much depth of snow we will see in the ice core. The West Antarctic Ice Sheet Divide has a lock in depth of about 75 meters. That means that until will drill down past 75 meters into the ice sheet at this location, we are still in firn(snow)! At South Pole, the yearly ice equivalent accumulation is much less at around 8 cm per year. The lock in depth at South Pole is much deeper at nearly 120 meters!
So now for a little math. If South Pole receives 8 cm per year ice equivalent of snowfall each year, how old is the snow/ice at the lock in depth of 120 meters?
Well, first we need to cover to the same units. Eight centimeters = 0.08 meters. Then we can do some basic math:
120 meters / 0.08 meters per year = 1500 years.
Wow, did you catch that?!
At the depth where the snow turns to ice at the South Pole, the snow/ice is roughly 1500 years old!
At the West Antarctic Ice Sheet Divide the math becomes:
75 meters / 0.22 meters per year = 340 years.
Okay, so know we know that the snow/ice is rather old at the depth in which it fully becomes ice. If you remember, in the firn layer, the air bubbles between the snow crystals are still able to move around and exchange with the atmosphere. That means the air in the firn is not aging at the same rate the snow is aging. There are some details that I’ll skip over here (if you want those you can email me), but basically right above the lock in depth the air bubbles in the firn have an age of zero. An age of zero means the air bubble is informing us about today’s atmosphere. All the concentrations of the gases such as carbon dioxide and methane would be roughly today’s concentrations (not exactly true, but I’ve skipped some physics just for simplicity).
So, the ice is several hundred to thousands of years old at the lock in depth, but the air bubbles are basically zero years old. This difference is referred to as the “delta age”. Delta in math typically stands for “difference” so we are comparing the difference in the age of the ice and the air bubbles. After the lock in depth, the air bubbles no long exchange with the atmosphere so they age at the same rate as the ice.
Okay, so basically we now know the air bubbles trapped in the ice cores are younger than the ice surrounding it. The “delta age” depends on the yearly snowfall accumulation at the site the ice core was drilled.
But now, how is the ice dated?
Well, we count the layers, literally. Just like the Grand Canyon, the different layers of snowfall can be observed in the ice cores when they are illuminated properly with lights (see picture). This picture was taken by John Fegyveresi at the SPICE Core core processing line at the National Ice Core Lab. You can see the small pieces of paper that he laid out to depict the different layers. You can even see the faint lines that separate the different years’ snowfall. Scientists like John visually count the layers in the ice core. Eventually, the amount of pressure of the meters upon meters of ice will start to thin the layers out (just like if you push your fist through some Play-Doh).
In addition to the annual layer counting, other measurements can tell us at what depth in the ice core a volcanic event occurred (see my earlier post about this). Many volcanoes have occurred during human history and are relatively well documented in terms of a date/time.
Combining the layer counting and volcanic peaks, scientists can come up with a robust depth-age scale (for a given depth in the ice core what is the age). Measurements of gases in the ice core air bubbles also helps tie time scales from other ice core sites together, just like fossils do in geology time scale constructions.
For our air bubbles, we basically subtract the “delta age” from the age of the ice to determine the age of the air that we analyze. For example, if I’m analyzing an ice core sample from 160 meters depth, the approximate age of the ice is:
160 m / 0.08 meters per year = 2000 years old.
We know from above that for simplicity, the air at the lock in depth is 0 years old (in reality is close to 200 years old).
The lock in depth at South Pole is 120 meters, corresponding to 1500 years old.
Our ice core sample has a gas age then of:
2000-1500 years = 500 years (with the all the physics it is more like 300 years old).
That is the quick and probably very hard to follow explanation of how scientists date ice cores and the air bubbles inside. I’m always eager to answer questions, so feel free to shoot me an email anytime.
Until next time,