DR. JAMES OVERLAND >>...things that we've never really seen before.

And this one relates to very warm temperatures, especially in January and February.

Here's a map of the Arctic, looking down at the North Pole. 

North America's on the right and the regions that had extreme temperatures are in red, in the central Arctic, north of western Russia and also in Alaska.

What sort of caught my eye on this is normally when I look at monthly average maps, like this, the scale normally goes from plus or minus 4 and to put the stat on there, we had to go from plus to minus 10. So it really was a major event.

In scoping this out, we average the temperature over the Arctic. And the top line here, is the plot of temperatures for January. The second line is a plot for February. 

And to make those maps of Arctic change over the whole Arctic, we actually interpolate the data onto our...is interpolated into what's called the reanalyses where they basically interpolate the data. So, not sure that just one mapping is the correct one, so we put on four different reanalyses, or way of averaging for the Arctic.

If you look at the last point on the upper right, if you average those, we have a 5 degree temperature anomaly over the whole Arctic compared to climatology. And the second largest value is 3 degrees, so not only did we surpass the previous record, but we blew away. We just surpassed it by a lot. 

Then to double check, we looked at some of the actual measurements at certain stations around the Arctic. That's the third line. 

And Scandinavia was not involved with the warming but three other Arctic stations here are all at 5 degrees.

And the bottom line is the monthly anomalies for every month, going back to 1980. 

And you can see how the months in January through April were all large on the bottom right.

So you know the point where we've had a value that's larger than we've ever seen before, there's a real philosophical question of how you interpret that. Is that part of just what's been going on in the climate beforehand? 

Or is it really something different out there? 

And people have been debating that since the 1800s in the Enlightenment. And there's no real clear way of thinking about that, especially when we have fairly short records compared to recent changes.

So if you want to compare the outlier with a statistical test for the null hypothesis, you can't really say that the new value is any different than you've had before. Well, you can't prove it.

But there's the other way of looking at it...that if you think more in risk analysis, like insurance companies, you know, is this something different?

You also cannot prove that it's *not* something separate. So for using NOAA and information that things in the Arctic really are changing, I tend to go with the latter interpretation.

And we can look at other changes besides just temperatures to add more credibility to the fact that something's really changing.

So let's look a little bit at what caused those temperature changes in January and February.

And meteorologists tend to look at weather maps. If you look at for what's driving the winds at sea level relative to a level surface, you can look at differences in pressure and those drive the winds, if you're looking near the surface.

Meteorologists, however, when you go up in the atmosphere, turning that around rather than pressure differences on a constant surface relative to a constant surface, they pick a pressure level and then see how the different elevations of that height surface vary. And those differences in heights drive the wind. So it's really the same thing. Pressure differences are driving all the winds, it's just whether which way you look at it.

So on the upper left figure here is the height of the 700 millibars surface which is about a third of the way up in the atmosphere. And you're looking down at the North Pole and Asia is at the bottom of the figure. And North America is at the top of these figures. 

And the winds tend to follow the constant height values of this height surface. So if you look at the age of the blue, that's basically where the tropospheric polar vortex would be going all the way around the globe. 

And you see there's some wavy features in it. They're called a ridge over north of Russia where the winds are heading towards the north and then head south again. And likewise there's southerly winds coming into Alaska and North America and then south in eastern North America. 

What was really different here is if you look at the two purple regions, there's polar vortex actually split so you had two different cells of lower heights or lower pressure with the winds going counterclockwise around each one. And then there's higher heights in between the two cells so the separation of the polar vortex in the two different pieces was somewhat unusual. And the second figure in the middle is the same figure as the first one only we've subtracted out the average to see, okay, what was different about January and February. In blue you see the low height areas from North America out over the Atlantic on one side and then south into the Pacific in the other. So the polar vortex that's normally fairly tightly controlled around the Arctic had these really strong lobes further south. And then they had this region of higher height signal in the middle that are separating the two different pieces.

The third picture shows the value of the southerly winds and in the light...in the orange colors. And we have southerly winds coming into the Arctic from western Russia and in over Alaska. 

And so those southerly winds are bringing the warm air into the Arctic. So the main...one main conclusion of the talk was the Arctic didn't warm just from internal changes of it. It really warmed because it was coupled with the rest of the hemisphere. And also when you bring in warm air, warm air is less dense and so the separation between these height levels is greater when it's warmer. 

So bringing warm air into the middle over the North Pole help increase these heights and help maintaining these two separate cells of the polar vortex. So that probably had a lot to do with why this pairing stuck around for a couple of months rather than normal random weather chaos of every couple weeks that we...the warm air helped reinforce the heights and separated these two cells.

If we look at a slice down here along western North America, here's that ridge of higher heights that's from winter 2015. Here it is in 2016. If you think of a wavy jetstream you would have a ridge here following, going south into the north Atlantic in both of these years.

If you go back to earlier in the decade, this ridge was actually over the center of the North Pacific Ocean. 

So when you got the wavy pattern, it brought cold temperatures into Florida and that sort of thing. 

But what's happened in the last couple years is that wavy pattern shifted eastward so that the bottom of the wave is now over the North Atlantic and feeds into the Arctic.

So that's what's happened this year. It looks like a one event that was tied with the weather and the mid-latitudes that helped bring the warm air from mid-latitudes into the Arctic. There wasn't anything special going on in the Arctic this year.

But as we know longer term that the Arctic is warming up faster than the rest of the planet. In this figure showed the average temperature for the Arctic in blue and the globe in red and the Arctic is warming up faster than the rest of the planet by at least a factor of two.

And there's a fair...and that's called Arctic amplification and that's a fairly well-known process. And there's a nice study that looks at why is the Arctic warming faster and breaks it up into different causes and how big they are.

And here's CO2 direct effect in the Arctic. This would just be normal greenhouse blocking over the Arctic would contribute about this much. 

But then the albedo effect where when sea ice melts...when sea ice is around it reflects the sunlight back to space. But if you melt the ice and just have the dark ocean you capture that energy for the planet in the newly sea ice-free areas and that's how big this effect is.

Another effect is when things are warmer the air holds more moisture and that tends to trap some of the heat as well. And then these two other terms here, called the lapse rate and the Planck, are a little more subtle. When you add more heat to a location, you radiate more energy out to space based on that increasing heat with the temperature being higher. But the rate that you lose it to space depends on what the background temperature is, not just what the increase is. And so in the Arctic, since the temperatures are lower than they are in the tropics, when you add more heat to the Arctic, it captures more of it compared to the tropics that tends to lose it. So all of these are special to the Arctic--why the Arctic is warming up faster than the mid-latitudes.

So we've seen some other surprises as well. I'm sure you all remember looking at sea ice extent in summer and that's shown on the red line here with the years along the bottom. And there's a big drop in the amount of ice extent we had in 2007. And even early in summer we didn't really know this was coming. So this was a big drop. And there was a lot of talk back then about a tipping point that, you know, I mention this albedo effect that if you lose ice then you get more heating in the ocean and it melts more ice and so on that some people thought this might be a tipping point in 2007 where we might just lose all the ice by a self-perpetuating loss.

So that didn't happen in '07 and '08. The ice recovered but it tended to stay down at this lower amount.

We had another loss in 2012.

If you look back to 2007, the meteorology helped reduce the sea ice. It wasn't just the albedo effect. So now looking back that had some one-year effect, just like the winter temperatures this year.

In 2012, about half was an albedo effect and about half was a meteorology effect. So that tends to seem to be how things are changing in terms of just extent.

But it's not just extent that's important for the Arctic sea ice and climate. We've also lost about 40% of the thickness of the ice. So if you combine both the area loss and the thickness loss, we've lost about two-thirds of the whole ice...the amount of ice that we used to have. And from satellites you can tell if the ice had been there a while or whether it just froze last year. 

If it just froze last year there's more salt still in it. It hasn't all drained out where multi-year ice, it's drained out. And so they can see that on satellites and the thick multi-year ice is showing in red and the first-year ice is shown in blue. And here in winter in 2008, after 2007, you see the much more extensive blue area than the red area that we had in years before that. 

But what we've seen is when we had those major ice loss and in the following year, some of that first-year ice is now second-year ice. And two years later some of that first-year ice is third-year ice. So after we had these real minimums of multi-year ice, you tend to rebound a little bit and that seems to be what we're seeing again from 2012 as well. 

But the big point here is the difference between a decade ago and what we've been seeing now is we've lost two-thirds of the amount of ice. Old thick ice was very stable. It was kind of the flywheel for the planet.

And so we've gone over to mostly first-year ice and it'll be very hard for us to go back to where we were before.

You wondering about this year--if the maximum amount of ice you have in an annual cycle normally occurs in February, but this year that maximum amount was less than we've ever seen before and so people were wondering, well, what's going to happen this summer? 

And on this graph the dotted line shows the loss of sea ice during the summer months and the blue line is for this year. So coming out of winter we had less ice than we ever had before but then in the last month or so, the weather has not been conducive to lose more ice rapidly. There's been more of a low pressure so the loss of the ice this summer has slowed down and is now tied with the previous record.

So whether we end up with a new minimum in the summer or not, we don't know, but it will certainly be one of the lower values. But again, we don't see any kind of runaway loss of sea ice again.

And that's one of the other points that I have.

Greenland is another indicator of changes in the Arctic. In 2012, the surface of the whole Greenland ice cap melted. That had never been seen before. And if normally when we lose mass from the ice sheet in Greenland, you lose it from the edge. And so this figure on the left keeps track of the loss of ice sheet, Greenland ice sheet, around the edges. And there's a long-term loss every year that as I said before, the wavy pattern lined up earlier in the decade with the high and the ridge over North Atlantic, sorry, North Pacific. And then if you look at the minimum and then the maximum of that, that also shooting warm air into west Greenland. And so five years ago, we had a period of real rapid ice sheet loss in Greenland, but again, just like sea ice, there's a long-term loss. But then when the meteorology lines up you can speed that up and that's what happened about five years ago. But then as I said before, the last couple years, the phasing of the waves has sifted further east, so the loss of ice sheet on Greenland has slowed down in the last few years.

So that's some of the complication between the weather and the physics of the Arctic that we're looking at more closely in the last couple of years.

Warming in the Arctic also changes the plant cover. From satellites, they have a measure of greenness.

And that, basically for the Arctic, how the Arctic changes from tundra vegetation to low shrubs and plants and that can be seen from satellite and the trend in that is shown here in these green areas.

So virtually over the whole Arctic you see areas where we are transitioning from tundra to shrubs.

Each one of those is a walrus. 

And it's not a good time to be a walrus in the Chukchi Sea, north of Bering Strait.

In the past, every walrus had his own ice floe...his or her own ice floe...and their own environment, the clams off the bottom of the Chukchi Sea.

And so they're all spread out, where without the ice floes there, they can tend to have to go to shore. And then when they go out to forage and hunt, they're going into the same region so it's not a good system for them.

And if we look forward for using climate models to project what temperature changes we're going to have, the left figure shows for the Northern Hemisphere and the right shows for the Arctic.

And then we have two scenarios. The pink line is business as usual if we don't do anything to slow down greenhouse gas increases.

The blue is if we do substantial but still rather modest mitigation.

And so on the left you see with the blue line out at 2100 or so, we can hit a two-degree increase in Northern Hemisphere temperatures with mitigation but without that we continue up.

If you look in the Arctic, however, we're still looking at six degree and I don't even want to think about it if we don't do anything in terms of mitigation.

The other point from this slide is that both of those emission scenarios give the same answer out for the next twenty or thirty years. That the CO2 we've already put in the atmosphere or what we're scheduled to put in in the next ten or twenty years, you're frontend loading the temperature response, so there's not much difference between these two cases out 2040s.

So even if we do mitigation starting now your children aren't going to see it, it's going to be your grandchildren that are going to see the eventual effect. So, you know, I'm thinking we're looking at four degrees here in the next 25 years.

But what I showed you at the beginning of the talk about what happened this winter and where we have surprises, this is the gradual increase. It doesn't include these surprises from the waviness of the wind interacting with the physics in the Arctic. So that this may be that this number may be too small. This is just the CO2 direct...an albedo effect without the coupling between the winds and the Arctic physics.

So on that positive note...

Yeah, so we saw a new all-time extreme in the Arctic to beat the previous extreme by two degrees or so.

And it was persistent. It wasn't just a one or two week event but certainly it lasted two months.

It wasn't really caused by the internal physics of the Arctic, it was really initiated by warm winds coming up from the south and helping them maintain themselves.

So are we going to have a runaway Arctic? Doesn't look like it based on the data. 

We don't see any real tipping point occurring but what we do see is when we have events like January and so on, it's warming the Arctic and there could be some memory that goes into the Arctic system that increases the rate of temperature increase and that's basically not included in the climate model projections.

So thank you very much!

[APPLAUSE]

MAN 1 IN AUDIENCE>> So, of course we had a pretty strong El Nino this year.  And there's been some previous work that has shown that when that happens there's more, uh...using the term loosely...[indistinct] wave energy coming out of the lower latitudes into the high latitudes and kind of disrupting the circulation.

Are you aware of anybody who's really looked into that as a kind of mechanistic way...[indistinct] mechanism for this particular event?

DR. OVERLAND>> Yeah. Yeah.

If you look at the previous strong El Nino cases, there's not a central Arctic warming that occurs at the same time.

However, that connection to the North Pacific, you know, we move the warm SST blob from the center of the Pacific in the ridge there over towards the West Coast and that looks ENSO-like and PEO-like and so ENSO probably had a contribution to the Alaska part of the warming but not the major warming in the central part of the Arctic.

MAN 2 IN AUDIENCE>> Yeah, so [indistinct] past graphs of disappearance of sea ice in the Arctic from basic numerical models that have run and oftentimes an individual model will go along and then the sea ice will drop off catastrophically and then bump along for a while. 

Are those catastrophic drops from those models associated with this kind of circulation in the atmosphere? Do we know that?

DR. OVERLAND>> Not necessarily the warm winter but they're associated with higher pressure and more of a clockwise flow during the summer. Like 2007 had a high pressure north of Alaska and clear skies and let more sun in and diffuse the ice more but that's more of a summer effect.

People who have looked have not really seen that what happens in the winter is a good predictor for what's going to happen in the summer.

But that's the question now with this thin ice. If the ice is, like, 70 meters thick then it doesn't feel much of what's happening at the surface.

But...and if it was 2 or 3 meters thick, like in the old days, it would never ever feel it. But the question here is around the edges north of Russia and Norway.

It was the ice thinning up there so when we brought in all this warm air from the south that it actually helped melt out some of that ice and so we did have a minimum winter ice amount at the same time. It's a question when it went away. The ice will...thin ice...if you start with thin ice, it'll grow faster again when it tries to recover and so we can't say that this year had a real effect on the ice but it's the kind of situation if we start having more of that again, we can start accelerating and having sea ice memory for thinning happen throughout the whole year.

MAN 2 IN AUDIENCE>> Thank you.

WOMAN>> Well, thank you, Jim.

DR. OVERLAND>>All right. Thank you everyone.