Draining glacial lakes cause temperature drop

[General Zod]

http://arstechnica.com/science/news/201 ... -chill.ars

For most of our current geologic period, the Holocene, temperatures seem to have fluctuated within a narrow range, indicating a fairly stable climate. But the end of the glacial period that preceded the Holocene saw a series of rapid temperature changes, with swings of up to 10°C in as little as a few centuries; the world would seem to be just about out of the glacial period, only to be plunged back into the cold for centuries. In the last few weeks, however, a couple of papers have indicated that these cold snaps may have had a single underlying cause: the sudden draining of massive glacial lakes left behind by the ice.

The idea that the melting that produced glacial lakes could ultimately have an enormous cooling effect on the climate may seem counterintuitive, but it's based on fairly straightforward geography. Large areas of North America drain northwards, through the St. Lawrence River basin or into Hudson Bay. During the last glacial period, these outlets were trapped under a massive ice sheet, leaving the water nowhere to go. So it built up into enormous glacial lakes, such as Lake Agassiz (shown below), until the lake surface reached a level where drainage would take place to the south.

But, as the ice sheets that kept these lakes in place melted, the remaining, weakened glaciers came under increased pressure from the enormous reservoirs they held in check. When the ice failed, staggering amounts of fresh water flowed into the ocean in a short time period.

Geologists think that the massive volume of fresh water dumped into the North Atlantic would be sufficient to interfere with the normal circulation that brings warm water north from the tropics. This in turn would give the Northern Hemisphere a chill, allowing the ice sheets to re-form; the expanded ice would reflect more sunlight back into space, enhancing the cooling. The net result, the hypothesis goes, is an extended cold snap that would persist until the general warming trend overwhelmed it.

That may sound like a tidy explanation, but it hasn't been universally accepted. A number of researchers think they've found evidence that the impact of a comet or meteor coincided with the biggest of the chills. Others, however, contest those findings. So what we seem to have is a genuine scientific controversy, and it sets the stage for the recent papers.

The first paper, in today's issue of Science, takes a look at a relatively minor cold snap occurred approximately 9300 years ago—minor being a drop of 2°C, about twice the magnitude of what we call the Little Ice Age. The authors use sediment records from a number of sites near Lake Superior to determine when various small bodies of water were connected to Superior itself. Based on the elevation of these sites, they can make inferences about the altitude of Lake Superior's surface.

Putting the records together, the authors conclude that the lake experienced a sudden and significant drop in surface levels. Lake Superior seems to have lost about 45m worth of water in a single event that coincides with the temperature drop that occurred 9300 years ago. (The authors found evidence that the water bypassed Lakes Erie and Ontario, and took a more direct route to the sea across southern Canada.) So, that seems to support the proposal that dumping a lot of fresh water into the Atlantic can drive a significant chill.

Draining Agassiz

Taking a bit off the top of Lake Superior may be a pretty big deal, but it was probably nothing on the event that may have set off the Younger Dryas, a chill that lasted over 1,000 years. That is thought to have been triggered by the emptying of Lake Agassiz. At various times, Agassiz covered portions of two Canadian provinces, and extended south into the US. The one problem here: nobody had conclusively identified the path the waters took on their way to the North Atlantic.

A paper from a recent edition of Nature provides a good explanation for that: they didn't exit into the North Atlantic. Instead, its authors argue that they've found physical evidence that the flood dumped the water into the Arctic near the Canada-Alaska border, along the course of the Mackenzie River. According to their calculations of land and ice sheet elevations, this route allowed the water to skirt the edge of the ice sheet for most of its trip to the sea.

If they're right, that's good news for the general idea that draining glacial lakes can have a huge impact on the climate. But it would seem to throw a bit of a monkey wrench into matters, since the fresh water would end up in the Arctic Ocean, where it's unlikely to have an immediate, direct effect on the Atlantic currents. The authors cite earlier work that had suggested, given the lower ocean levels and positions of other ice sheets, the only place for the fresh water to exit the Arctic was past Greenland and into the North Atlantic. So, the end result would have been the same, even if the route was different.

Combined, the two papers make the case that freshwater entry into the Atlantic can have a major impact on the hemisphere's climate and, with strong enough feedback effects, alter the global climate.

I'm not really sure what this means for combating global warming if anything, but it's an interesting read all the same.

[Darth Wong]

It actually seems rather obvious that glacial melting would cool down ocean water, to the point that one questions how anyone could possibly deny such a mechanism, and what rationale they could possibly use. If you have a huge chunk of ice (which may include bodies of meltwater lakes) resting mostly on land and it starts to melt and run off into the ocean, then duh, it will cool the ocean down.

[Sarevok]

The only question is does it make a quantifiable difference in global temperature that is worth the cost of such an undertaking. Without knowing the answer to that it is at best an interesting curiosity.

[Darth Wong]

The only question is does it make a quantifiable difference in global temperature that is worth the cost of such an undertaking. Without knowing the answer to that it is at best an interesting curiosity.

"Cost of such an undertaking?" I don't think we're talking about deliberately melting glaciers to cool down the world; they're just talking about what might have influenced temperature changes in the past.

[tim31]

Glaciers aren't exactly easy to tamper with anyway. As I recall the 1964 Alaskan earthquake, a genuine 9.2, had a net effect of fuck all on glaciers running into Prince William Sound.

This theory has been around for a while, and inspired the chain of events in The Day After Tomorrow. Water density is important of course, as that cold water going into the North Atlantic is going straight down until it meets colder water, resulting in a stacking effect that disrupts the thermohaline current. Given that this is a 1,600 year cycle, it's easy to see what a powerhouse it is and how much it'd wreck the show to have a spanner in the works.

[Kuroji]

I don't know about you guys, but unless we see global warming jump to the point where Louisiana and Florida are genuinely being flooded, I don't think anyone should be wanting to get those glaciers into the ocean. A two degree difference would be enough to basically ruin global crop cycles as we know them, I imagine.

Then again, if an earthquake of that magnitude didn't disrupt the glaciers, nothing people can do (short of burying a nicely sized nuke in one end of a glacier and blowing it out) would affect them anyway.

[Sarevok]

The only question is does it make a quantifiable difference in global temperature that is worth the cost of such an undertaking. Without knowing the answer to that it is at best an interesting curiosity.

"Cost of such an undertaking?" I don't think we're talking about deliberately melting glaciers to cool down the world; they're just talking about what might have influenced temperature changes in the past.

Ugh, upon further reading you are right.

That said it would be pretty cool if humans could lower the temperature of planet Earth. There loads of pie in the sky proposals for cooling down Venus or warming up Mars. It would be nice if someone had an actual workable scheme that would at least work for Earth.

[Simon_Jester]

Most of the ones I've heard of revolve around deliberately spreading dust in the stratosphere. I'm not sure if they'd actually work or not.

[Sky Captain]

Most of the ones I've heard of revolve around deliberately spreading dust in the stratosphere. I'm not sure if they'd actually work or not.

If you get enough dust into stratosphere it would have a cooling effect. When Tambora volcano blew up in 1815 it created Year without summer by injecting hundreds of millions tons of fine dust into stratosphere. However problem is to get required quantities of dust that high. Only means I can think of that could work on required scale involves large amount huge thermonuclear bombs and that would spread radioactive fallout everywhere.

[GrandMasterTerwynn]

Most of the ones I've heard of revolve around deliberately spreading dust in the stratosphere. I'm not sure if they'd actually work or not.

Not quite dust. Much smaller particles of sulfur-bearing aerosols. As has been written, we'd basically replicate with SCIENCE what large volcanic eruptions do naturally (i.e. inject these sulfur aerosols into the stratosphere, which are responsible for causing the cooling associated with volcanic winters.)

[LadyTevar]

Large areas of North America drain northwards, through the St. Lawrence River basin or into Hudson Bay.

I know the Red River still flows north into Lake Winnepeg, which would certainly be a watershed for Lake Agassiz were it not for careful comparison the map of the Lake and the current watershed for the Red River. The southern border of Lake Agassiz totally covers the watershed for the Red River. This makes it very possible that the Red River was formed as the Lake drained and left a new landscape behind. What we can tell from the map of the Red River's flow is that there is a ridge of high ground that separates it and it's tributaries from the Missouri River watershed. Once again, this could just be the effect of Glaciers pushing dirt in front of them and leaving a mound of material, or it could be that Lake Agassiz left behind a huge hole. If so, then the Red River is doing no more than draining what's left of the lake-bed, as it has probably been doing since the Lake drained, while the Missouri River is traveling the untouched shoreline.

The Saint Lawrence Waterway is another river that flows North as it drains the Great Lakes out into the Atlantic. However, it's been proven that the Saint Lawrence watershed was formed after the Ice Age, and is simply following a valley formed by the flow and retreat of the Glaciers. If it was connected to Lake Agassiz at all, it would have been a minor drainage point.

So, what other river in the area was flowing North that could be Feeding the Lake? The New River flows north through North Carolina, Virginia, and West Virginia, before cutting west and flowing into the Ohio River and down the Mississippi. The New is totally misnamed, as its path cuts through the Appalachian Mountains as if it was cutting through the mountains as they were rising up. There is a theory that before the last Ice Age the New River continued flowing north through Pennsylvania or Ohio and then into the Hudson River Basin. The theory states that the last Ice Age blocked the New's passage north, and forced it to cut west along the boundary of the glaciers, setting the current riverbed. This theory remains unproven, but the New River may be one of the tributary feeders.
(Side Note: Interestingly, the native names for the New River and the St. Lawrence are similar -- "Kanawha/Konhaway" and "Kahnawáˀkye". In theory, "Kanawha" means "River of little death" for the whirlpools that could and would swallow boats whole.)

[CaptainChewbacca]

This theory has been around for a while, and inspired the chain of events in The Day After Tomorrow. Water density is important of course, as that cold water going into the North Atlantic is going straight down until it meets colder water, resulting in a stacking effect that disrupts the thermohaline current. Given that this is a 1,600 year cycle, it's easy to see what a powerhouse it is and how much it'd wreck the show to have a spanner in the works.

Quick clarification, a thermohaline disruption would most likely be caused because water flooding into the north atlantic was FRESH water, not because it was cold.

[Big Phil]

Given that this is a highly complicated subject, I don't suppose there's any simple way to explain the impact of climate change on the global climate, is there?

For example, global warming==> glaciers melting in Alaska ==> colder water in California ==> no rainfall in the midwest ==> duststorms and famine

I know this is likely beyond most climate change deniers, but it would still be a helpful exercise

[CaptainChewbacca]

Given that this is a highly complicated subject, I don't suppose there's any simple way to explain the impact of climate change on the global climate, is there?

For example, global warming==> glaciers melting in Alaska ==> colder water in California ==> no rainfall in the midwest ==> duststorms and famine

I know this is likely beyond most climate change deniers, but it would still be a helpful exercise

The simple truth is we DON'T KNOW exactly how climate will be affected. We can guess, but there's still too many variables for 100% accurate predictions.

[Narkis]

We cannot accurately predict the weather past a few days. Our best minds and computers have been trying to solve this issue, and their success has been very limited. Climate is an inherently chaotic system, and thus we'd need a staggering amount of data to predict a chain of events as specific as you seem to want. General trends are far easier to do though, and that's how we now that something Bad is going to happen if nothing changes.

This is the reason that studying past climate changes is so important, btw. That way we get more data, and we need them to better refine our models and predictions.