Expedition 374 and ice sheet modelling Conversation with Benjamin Keisling, scientist onboard the JOIDES Resolution during Expedition 374

Benjamin Keisling, scientist onboard the JOIDES Resolution during Expedition 374

Benjamin, can you please introduce you and explain your research topics?

My name is Benjamin Keisling and I am part of the Sedimentology team onboard Expedition 374. I am a PhD Candidate in Geosciences at the University of Massachusetts Amherst in the USA.

My PhD research centers on ice sheet modelling, and is particularly focused on how we can use proxy data to constrain uncertain processes and parameters in ice sheet models.

In other words, I use geological data to test how well ice sheet model simulations represent the past. The models that best represent the available geologic data can then be used with more confidence to make future predictions of sea level rise.


It is commonly accepted that big changes in the past sea levels are linked to variations of the volume of the Antarctic ice sheet. Is that true, and in that context, how important is it to have a better knowledge of these past variations?

Big changes in past sea levels certainly reflect variations in the volume of the Antarctic ice sheet, and accurate prediction of future sea levels relies on better knowledge of these past variations.

For example, we often use the most recent warm “interglacial” period in Earth’s history (the Eemian, 125 thousand years ago) as an analogue for the present because the amount of carbon dioxide in the atmosphere was similar to preindustrial levels (around 280 parts per million). However, sea level records indicate that during the Eemian sea levels were 6–9 meters higher than they are today, which means that part of Antarctica must have melted.

But it’s very difficult from a modelling perspective to make enough of the ice sheet melt to explain those sea levels, so it means that our models are missing something, and that affects our ability to confidently model how the ice sheet will respond to warming today (when carbon dioxide levels are greater than 400 parts per million!) and into the future.

Benjamin the modeler at work


On board, sedimentologists use the lithologies of the cores to estimate advances and retreats of the ice sheet during periods warmer than today, especially during Miocene and middle Pliocene. Can you explain how these data will be helpful in your work?

We know from many records that the size of the Antarctic ice sheet has fluctuated in the past, and models do a good job of recreating some of that variability. However, we are still lacking fundamental knowledge about what caused the ice sheet to retreat during past warm periods, and the sediments we are collecting can give us insight into that.

In Antarctica today, the ice mostly loses mass at its floating ice shelves (like the Ross Ice Shelf), where icebergs calve off and warm ocean waters melt the underside of the ice. Because Antarctica is so far south, it rarely melts from the surface – it stays below freezing even during the summertime. But there is a big debate now about whether the large variations in the size of the ice sheet are triggered by melting from the ocean or melting from the atmosphere.

Through studying the sediments we are collecting on Expedition 374, we will get a better idea of what the oceanic and atmospheric conditions were like before the ice sheet retreated, and this gives us information we can use to evaluate our models. In addition, it is notoriously difficult to constrain how certain parameters in our models have changed throughout geologic time, for example, the shape of the ice-sheet bed or how slippery it is. The sediments that we are describing can give us direct insight into these important parameters, so that our models have a better shot at getting the right answers for the right reasons.

We want to know how the sea levels will evolve in the future. And Expedition 374 is looking 25 My behind us! How can having a better knowledge or our past help us to predict the future when the atmospheric CO2 concentration will increase?

Geological data provide an invaluable source of information for models, because they give us targets to test models against. We don’t know exactly what will happen in the future, but geological data can tell us what happened in the past, and right now, the future is looking more and more like the Miocene and middle Pliocene. The amount of carbon dioxide in Earth’s atmosphere hasn’t been as high as it is today since at least the middle Pliocene, and during that time the ice sheet was probably smaller and very dynamic, causing oscillations in sea level on the order of 10 meters. We need to understand what drove these oscillations and how they affected the rest of the global climate system in order to predict when and how they will happen in the future.

If you want to learn more about what is a numerical modelling of ice sheet, you can visit this webpage:


An example of ice-sheet numerical modeling is available here:


Agnes Pointu
Agnès is a French high school teacher with a passion for sharing geological sciences.
More articles by: Agnes Pointu