Scientists Found a Warm Arctic Beneath a Frozen Russian Lake

About 3 million years ago, during the Pliocene geologic epoch, the Arctic was warm. Not beach party warm, but warmer on average by about 8°C (14.4°), according to a new study out today in Science. Those eight degrees are enough for it to not look anything like it does today: there were forests instead of tundra, for one, and during the warmer months, the ice cleared out, leaving open ocean.

As carbon dioxide levels dropped in the atmosphere, the planet cooled and, eventually, the Arctic became the frozen cap it is today and has been throughout relatively recent history. It's expected that we'll have open ocean once again, thanks to climate change. Figuring out how exactly these glacial cycles work in the Arctic is the work of University of Massachusetts researcher Julie Brigham-Grette, lead author of the new paper, and her colleagues.

Reaching the warm Arctic conclusion took pulling a 318 meter-long core sample from beneath the frozen Lake El'gygytgyn in Russia, and subjecting it to a variety of different state-of-the-art techniques to derive information about the region's paleohistory, including temperature, precipitation, water chemistry, and more. In effect, the core sample functions as a very long timeline with its information encoded in plant pollen and chemicals.

The story is not nearly so simple as the planet warming up and the ice melting, as if the Arctic were an ice cube in a glass of whiskey. The interconnections and feedbacks involved are daunting, and Brigham-Grette's work is just the beginning. I emailed with her about that work earlier this week.

What did 8°C warmer actually look like in the Arctic?

Arctic summers at least 8 deg C warmer than today in the Pliocene produced forest cover, probably all the way to the Arctic coasts. That means tundra was completely eliminated. We presume that the Arctic ocean was completely ice-free in summer (probably only ice in winter) and the glacial ice on Greenland  was restricted to the mountains along the eastern coast  (must have looked like southeast Alaska today). This impression comes from looking at our data and other records of the Pliocene along with climate model reconstructions.

How does the Arctic warmth during this period coincide with planetary CO2 concentrations? I see that concentrations during the Arctic's warm period are comparable to modern post-industrial concentrations, but what did the process of Arctic warming look like?

This is what we seek to learn--what were the mechanisms that drove the Arctic into glacial/interglacial cycles?

The warming of our planet now is in response to rising [greenhouse gases; rapid changes in albedo (ice to open water) is, thus, a consequence of that warming today. In the Pliocene, CO2 is thought to be in the range of 320 to 410 ppm (different groups have different measures) and this was largely because of what is called the rock weathering thermostat and Earth feedbacks. The CO2 of the Earth has been in decline since 65 million years ago when it was very, very high. (Best estimates are all over the place, but as high as 2000 ppm+.) But it’s been declining (more or less) since then as mountain ranges have been uplifted and this drove more rock and mineral weathering that consumes CO2.

Declines in CO2 also involve feedbacks and eventually thresholds were met and ice sheets expanded in "cold" orbits of the Earth around the Sun (those are the Milankovitch cycles).

To grow glaciers and develop large ice sheets requires that snow in the winter does not melt at all in the summer. Ice sheets can expand only if, year after year, the snow accumulates in mountains and on high elevation plateaus (like Baffin Island, for example) with the snow line (or Equilibrium Line ) below the accumulation zones.

So, having forested conditions at Lake El'gygytgyn and other arctic locations late into the early Pleistocene would argue that summers were warm enough to support forests and did not favor the persistence of snow. In "cold summer orbits," referring to changes in the shape of Earth's orbit around the sun, it is more likely that snow will stick around in high-elevation or high-latitude locations. But if it's too warm during the cold orbits, then it’s still hard to get ice sheet growth started.

[This is] our future work--involving collaborations with climate modelers--determining what the thresholds were.

Can you explain a bit about how we can get this kind of information from lakebed samples?

Some of this is explained in the supplement to the paper. But what we do is study the style of sedimentation, the amount of organic matter and biogenic silica reflects productivity in the lake (more life means warmer summers usually), and we scan the core for all sorts of elemental data that tells us about the geochemistry of the lake in cold vs. warm intervals. Most importantly, in our study we looked at the pollen in the lake sediments because it reflects what is living around the lake and in the region. Palynologists can use "modern analog" methods to then determine the summer [and] winter temperature habitats, and the precipitation needed to create these biomes. This is very powerful information.

Maybe an odd question, but with so much energy devoted to reconstructing a landscape through so many years, do you spend much time trying to imagine what it looked and felt like around this now icy presumably fairly brutal lake?

Yes—it's a great question. In 2003, I camped next to the lake for 10 weeks—June to September—and marveled at it! It is not a stretch to imagine the lake surrounded by forests. We also have evidence that there were large mammals around too (we find digestive algae from their guts in the lake sediments) so imagine boreal and mixed forests with Mammoth and bison and horse etc. Very fun.

Reach this writer at michaelb@motherboard.tv.