Paleopollen in Antarctica

Fossilized pollen reveals climate history of northern Antarctica

Analysis of direct climate record shows Antarctic tundra persisted until 12 million years ago

HOUSTON — (June 27, 2011) — A painstaking examination of the first direct and detailed climate record from the continental shelves surrounding Antarctica reveals that the last remnant of Antarctic vegetation existed in a tundra landscape on the continent’s northern peninsula about 12 million years ago. The research, which was led by researchers at Rice University and Louisiana State University, appears online this week and will be featured on the cover of the July 12 issue of the Proceedings of the National Academy of Sciences.

The new study contains the most detailed reconstruction to date of the climatic history of the Antarctic Peninsula, which has warmed significantly in recent decades. The rapid decline of glaciers along the peninsula has led to widespread speculation about how the rest of the continent’s ice sheets will react to rising global temperatures.

“The best way to predict future changes in the behavior of Antarctic ice sheets and their influence on climate is to understand their past,” said Rice University marine geologist John Anderson, the study’s lead author. The study paints the most detailed picture to date of how the Antarctic Peninsula first succumbed to ice during a prolonged period of global cooling.

In the warmest period in Earth’s past 55 million years, Antarctica was ice-free and forested. The continent’s vast ice sheets, which today contain more than two-thirds of Earth’s freshwater, began forming about 38 million years ago. The Antarctic Peninsula, which juts farther north than the rest of the continent, was the last part of Antarctica to succumb to ice. It’s also the part that has experienced the most dramatic warming in recent decades; its mean annual temperatures rose as much as six times faster than mean annual temperatures worldwide.

IMAGE: Researchers ascertained the exact species of plants that existed on the Antarctic Peninsula over the past 36 million years during a three-year examination of thousands of grains of fossilized pollen,…

Click here for more information.

“There’s a longstanding debate about how rapidly glaciation progressed in Antarctica,” said Sophie Warny, a Louisiana State University geologist who specializes in palynology (the study of fossilized pollen and spores) and led the palynological reconstruction. “We found that the fossil record was unambiguous; glacial expansion in the Antarctic Peninsula was a long, gradual process that was influenced by atmospheric, tectonic and oceanographic changes.”

Warny, her students and colleague Rosemary Askin were able to ascertain the exact species of plants that existed on the peninsula over the past 36 million years after a painstaking, three-year examination of thousands of individual grains of pollen that were preserved in muddy sediments beneath the sea floor just off the coast.

“The pollen record in the sedimentary layers was beautiful, both in its richness and depth,” Warny said. “It allowed us to construct a detailed picture of the rapid decline of the forests during the late Eocene — about 35 million years ago — and the widespread glaciation that took place in the middle Miocene — about 13 million years ago.”

Obtaining the sedimentary samples wasn’t easy. The muddy treasure trove was locked away beneath almost 100 feet of dense sedimentary rock. It was also off the coast of the peninsula in shallow waters that are covered by ice most of the year and beset by icebergs the rest. Anderson, a veteran of more than 25 research expeditions to Antarctica, and colleagues spent more than a decade building a case for the funding to outfit an icebreaker with the right kind of drilling equipment to bore through the rock.

IMAGE: This is Rice University oceanographer John Anderson aboard the Nathaniel B. Palmer.

Click here for more information.

In 2002, the National Science Foundation (NSF) funded the project, which was dubbed SHALDRIL. Three years later, the NSF research vessel Nathaniel B. Palmer left on the first of two drilling cruises.

“It was the worst ice year that any of us could remember,” Anderson said. “We’d spend most of a day lowering drill string to the ocean floor only to pull it back up to get out of the way of approaching icebergs.”

The next year was little better, but the SHALDRIL team managed to obtain enough core samples to cover the past 36 million years, thanks to the logistical planning of marine geologist Julia Wellner and to the skill of the drilling crew. By end of the second season, Anderson said, the crew could drill as much as a meter every five minutes.

Reconstructing a detailed climate record from the sample was another Herculean task. In addition to the three-year palynological analysis at LSU, University of Southampton palaeoceanographer Steven Bohaty led an effort to nail down the precise age of the various sediments in each core sample. Wellner, now at the University of Houston, examined the characteristics of the sediments to determine whether they formed below an ice sheet, in open marine conditions or in a combined glacial-marine setting. Other members of the team had to count, categorize and even examine the surface texture of thousands of sand grains that were preserved in the sediments. Gradually, the team was able to piece together a history of how much of the peninsula was covered by glaciers throughout the past 36 million years.

“SHALDRIL gave us the first reliable age constraints on the timing of ice sheet advance across the northern peninsula,” Anderson said. “The rich mosaic of organic and geologic material that we found in the sedimentary record has given us a much clearer picture of the climatic history of the Antarctic Peninsula. This type of record is invaluable as we struggle to place in context the rapid changes that we see taking place in the peninsula today.”

###

The study was funded by grants from the NSF’s Office of Polar Programs to Anderson and Warny. Study co-authors include Wellner; Askin; Bohaty; Alexandra Kirshner, Tyler Smith and Fred Weaver, all of Rice; Alexander Simms and Daniel Livsey, both of the University of California, Santa Barbara; Werner Ehrmann of the University of Leipzig; Lawrence Lawver of the University of Texas at Austin; David Barbeau of the University of South Carolina; Sherwood Wise and Denise Kulhenek, both of Florida State University; and Wojciech Majewski of the Polish Academy of Sciences.

Get notified when a new post is published.
Subscribe today!
0 0 votes
Article Rating
53 Comments
Inline Feedbacks
View all comments
Bill Illis
July 2, 2011 4:32 pm

Jerry Dickens says:
July 2, 2011 at 8:11 am
… CO2 was about 3-4 times higher than present-day in the early Eocene, and when Antarctica had no ice.
————-
And it was 3-4 times higher when Antarctica glaciated over 33.6 million years ago. CO2 then fell and Antarctica warmed up.
There is no correlation when you look at all the evidence (and I have all the stomata and boron numbers – boron has CO2 at 71 ppm at 42 million years ago and stomata has one estimate at 298 ppm at 55.9 million years ago). There is a lot of cherrypicking involved in any effort to say CO2 drove the Antarctic glaciation. It was continental drift and the eventual isolation of Antarctica in an extreme polar climate when the Circumpolar Current started up/stopped/restarted etc.
http://img819.imageshack.us/img819/6813/allpaleoco2complete.png
http://img834.imageshack.us/img834/9665/tempc0245m.png

July 3, 2011 4:03 am

Hej Bill,
The graphs that you show are interesting, but should come with two caveats.
First, the long-term Cenozoic d18O record (constructed using benthic foraminifera) is a “mixed” signal. It is affected by temperature (presumably at high latitudes where water sinks), and the d18O of the water (which depends on terrestrial ice formation). Thus, the d18O record is not a strict recorder of temperature. It is probably a good proxy for changes in high-latitude temperature before ~35 Ma, and the onset of significant ice on Antarctica, but not afterward, because then much of the signal derives from changes in ice volume. (As an aside, you can get a more recent version at: http://es.ucsc.edu/~jzachos/Publications.html … and go to 2008). It’s also interesting to plot things to 60 Ma and through the Early Eocene.
Second, the proxies for past CO2 are neither accurate nor precise. Indeed, this remains a huge challenge (as highlighted most recently by Zeebe, Nature Geoscience, 2011).
Basically, oodles of data show that Earth’s surface temperature has changed significantly over time. In particular, the Paleogene, and notably the early Eocene, was much warmer than present-day, especially at high latitudes. This is definitely NOT because of continental drift (tectonics). As mentioned in previous posts, Antarctica has not moved much in the last 90 million years; moreover, the signal for past warmth is really obvious in the Arctic and numerous other locations. Without question, climate on Earth was much different in the past … and this demands a good explanation.
There is good evidence that pCO2 has also changed over time, and there is a first-order correspondence between temperature and pCO2 (when the records are aligned and noting the above caveats; see for example Zachos et al., Nature, 2008).
The long-term records for temperature and pCO2 generated so far are not ideal. I am open to the idea that past changes in temperature and pCO2 are not related directly. Indeed, the very fact that no model to date can explain observations for the Early Eocene strongly suggests there is some basic problem in our understanding of how Earth’s climate operates (see recent paper by Valdes, Nature Geoscience, 2011). This does not mean that temperature and atmospheric CO2 are unrelated, but likely that it is more complex than generally offered.
Jerry

Bill Illis
July 3, 2011 3:15 pm

Jerryd says:
July 3, 2011 at 4:03 am
… the d18O record is not a strict recorder of temperature. It is probably a good proxy for changes in high-latitude temperature before ~35 Ma, and the onset of significant ice on Antarctica, but not afterward, because then much of the signal derives from changes in ice volume.
———————————-
I always assumed temperatures and changes in ice volumes were correlated – well I guess we should just throw the ice core records then since they are only good before there was any ice.
Many people only like the dO18 isotope data when it shows what they want to see.
But it is the best we got and here is another longer time-scale (all 25,000 datapoints including Zachos 2008 which was just a very minor update and it was already in my numbers). Here one can see the precious Eocene in perspective. Not particularly remarkable in that period compared to other recent 5.0C warming events (the last 28 ice ages for example) and in other periods in the climate
http://img823.imageshack.us/img823/9508/tempco2570mlefttoright.png