The last warm period: an analog for our future?

In ice core research we would love to probe a time period that was warmer than today for better predictions of our future. However, these time periods have been rare in what we can cover with ice cores, with probably one exception which we want to probe on Taylor Glacier.

With ice cores it has been possible to probe the climate as far back as 800’000 years before present. This is an extremely long time range, but only during about 10% of this time, the globe was in a state of warm climate, as we are since the last 10’000 years. All the past warm periods were either a bit colder or equally warm than the present one, but none of them significantly warmer than the present one. This was the state of the art until a couple of years ago, when research started to find evidence that the last warm period (about 125’000 years ago) was probably the one exception.

With a few degrees warmer climate than what we have today (maybe only true for certain regions), this time period has potential to give us insight in our future climate and can serve as a kind of analog for where we are heading to with human-made global warming. It won’t be the perfect example, that’s clear, because greenhouse gas concentrations have not been as high as we are driving them to (they were about 30% below the current concentrations of CO2), but other mechanism seemed to have made the globe slightly warmer than today (at least regionally). To find out what these mechanisms were, how they worked, and what it meant for the climate is what drives our interest in this specific time period.

Arial Picture of Taylor Glacier with surface ice ages.

Taylor Glacier tongue with the surface ice ages along the main flow line. The intensified lines across the flow line indicate our main work sites with the main transect at the narrow part of the valley (ka = kilo annus = thousand years before present). [credit: PhD thesis Daniel Baggenstos]

125’000 years old ice is old even for Taylor Glacier standards (see picture for a reference of the ice age). With the past few years of fieldwork it was possible to identify and nicely date a well preserved section on the glacier that covers the time period between about 6’000 and 52’000 years before present (the so called main transect). This transect is part of a unique ice fold that extends over a large part of the lower glacier tongue. However, this main fold is highly disturbed on its edges and it has been proven difficult to find older ice on the glacier.

Nevertheless, the search for older ice further down the glacier tongue has continued and it has been possible to find shorter sections of ice that date back to around 70’000 years (the so called MIS 4-5 site) and even older ice around 125’000 years of age. The ice, however, is highly compressed and the stratigraphy disturbed such that the first attempt to retrieve a record for this period was not fully satisfactory. Therefore, we are revisiting this site and hope to drill a core a couple of meters away from the original site that hopefully contains the period in a better fashion. Due to the complex stratigraphy of the glacier, in particular at the sites down the glacier, it is to some degree a try-and-error work and some luck is needed to drill in a good section.

We plan to go twice to that site: at the very beginning of the season and sometime in the middle. The plan is to drill a few tens of meters away from the original site, bring the cores back to the main camp where we can do some analytics to see what we got, and then go back again to hopefully get the best possible core from that site. Regardless whether we find a slightly better site or not, the ice we will get from there will be used to address different questions. Two examples are given here:

By measuring the Krypton and Xenon content in the trapped air it is possible to reconstruct the global mean ocean temperature back in time. With these measurements – which rely on fairly large sample and are therefore easy to perform with ice from Taylor Glacier (see previous blog for details) – we will hopefully shed some light on the question how warm the ocean was back in the last warm period (was it warmer or similarly warm than today?).

Another focus is going to be absolute dating of a certain event around that time. Dating of the trapped air based on the radionuclide Kr81 (similar concept as explained in a previous blog) is possible at Taylor Glacier, only because we can take so large ice samples. This technique will give us an independent and very precise age for the trapped air and tell us at which point in time this event has actually happened.

More happy science coming soon!

Share This:
Facebooktwittergoogle_plusredditpinterestlinkedintumblr

2 Comments on “The last warm period: an analog for our future?

  1. Thanks Berni for the background of your research.
    How exactly is the concentration of Krypton and Xenon correlated to the mean ocean temperature? And how precise can you determine the ocean’s temperatures with the found concentration? Thanks 🙂

    • The amount of Krypton and Xenon in the atmosphere-ocean system is conserved. In other words, the total amount of molecules in this system changes only very little and very slowly on time scales of millions of years. The ratio of molecules dissolved in the ocean versus molecules in the atmosphere is determined by the ocean temperature and the solubility, respectively. By measuring the changes in the atmosphere we can therefore infer the temperature of the ocean, because we know very well the solubility of the gases in the water. The changes in the atmosphere are very small though and it needs very sophisticated Mass spectrometry to measure the signal precisely enough. The overall precision has been in the range of +/-1°C (not so good if you consider an peak-to-peak ocean temperature changes of about 3°C we expect to find on our time scales). But recent work reaches precisions in the range of a couple of tens of a degree.

      Science talking 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *