When it was initially apparent that a period of mild cooling was taking place in the high Arctic one month ago, I decided to reach out to Walt Meier of the National Snow and Ice Data Center. I wanted him to get on the record about what to me looked like a shift from years past. What at the start of our exchange was the longest period of days cooler than average in the DMI record has since grown from about 60 days to about 100.
Always affable, always reasonable, always intelligent, and yet somehow always alarmist at the same time, Walt was kind enough to oblige. Among the surprising pieces of his end of our e-mail exchange: His willingness to jump on the story that the North Pole cam was capturing something significant in terms of melting. (At first, that was Walt’s claim – he backs down from it later, as you’ll see.) Walt easily, and skillfully, frustrates my desire to get him to show any optimism (or even recognition) of the cooling, and quietly beats the all-melt, all-warming drum throughout. Interestingly, Walt has actually left NSIDC and gone to work at NASA/Goddard Institute for Space Studies just in the last week.
Below is our exchange; I’ve bolded a few elements. It’s a lengthy exchange, and sometimes one of us is responding to an e-mail one or two earlier in the exchange, but overall the back-and-forth underscores the experience of trying to get a scientist squarely in the AGW-is-a-disaster camp to acknowledge any evidence to the contrary.
(July 12, 2013, 5:52 a.m.)
I saw that according to DMI, temperatures above of 80N have sat beneath the green average line for more than 60 days straight, the first time in the entire record that this is the case.
I am not claiming a cooler-than-average year, nowhere close. I do personally find it of note, however, that the 60-day cool spell has taken place at all and that it includes, so far, the warmest part of the year when temps at this latitude are above freezing (which they are this summer as well, but slightly less than usual).
Have you been surprised by the 60-day stretch of cooler-than-average temps? And do you attribute it to multiple weather phenomena? Do you find the cool spell in any way hopeful?
Thank you in advance.
July 12, 2013, 5:26 p.m.
Thanks for the question. It has been a cool start to the summer in the Arctic. Alaska had unusual snows in May (though by last month they had record high temperatures), Greenland melt has been slow to start, and sea ice decline has not been as fast as in recent years – particularly north of Alaska. And as you mention, the DMI temperatures have been a bit lower than normal (though up north of 80 latitudes, temperatures never get much above freezing).
These aspects show the influence of natural variability in the climate system, particularly over the shorter term. Even in a warming world, cool spells can still happen, especially in a given region. This spring and early summer in the Arctic, the main reason being low sea level pressure over the North Pole. This has made the region cloudier and thus cooler during the early melt season.
What this means for the rest of the summer is an interesting question. Greenland melt can jump quickly. As mentioned above, Alaska warmed quickly after the cold start. And sea ice loss sped up and is now that far above last year at this time with a lot of thin ice available yet to melt.
So bottom line, natural variability can slow things down for a bit – we don’t expect new records every year. But over the long-term, the recent cold spell in no way changes the conclusion that there will be continued warming, melting and loss of ice.
July 24, 2013, 9:09 a.m.
Surprised by this story:
“In early July, temperatures were 2 to 5 degrees Fahrenheit (1 to 3 degrees Celsius) higher than average over much of the Arctic Ocean, according to the National Snow & Ice Data Center. – See more at: http://www.livescience.com/38347-north-pole-ice-melt-lake.html#sthash.KrWpLjjt.dpuf“
You were acknowledging the cool summer so far around the same time — can you help me understand?
Also, is there a link to NSIDC’s version of Arctic temps that I can be checking?
Thanks for all!
July 24, 2013 7:11:13 PM EDT
It was simply due to timing. It had been a cool start to summer, but it did start warming up by early July. When I responded I hadn’t looked at the latest temperatures, which had started to become warmer around the beginning of July.
We use the NCEP/NCAR Reanalysis fields (925 mb level air temperatures to avoid errors due to the boundary layer). Here is the link to create plots – we use a custom polar stereographic projection.
July 24, 2013 7:21:41 PM EDT
Are you surprised that DMI continues to show temps at or below average through the period?
July 24, 2013 7:27:09 PM EDT
It’s interesting, but I don’t put a whole lot of stock in the DMI temperatures. These are model reanalysis fields, which are not particularly good near the surface. Everything is near the melt/freeze point at the surface, so there isn’t much variation.
It’s pretty clear from the North Pole camera that there is a lot of melt going on in the high Arctic:
(the station has drifted south from the pole, but still at ~87 degrees N).
July 24, 2013 7:22:28 PM EDT
How are the reanalysis fields obtained — by satellite?
[I’m referring to the NCEP/NCAR fields here]
July 24, 2013 7:28:29 PM EDT
The reanalysis fields use in situ station data, some satellite data, and models.
July 24, 2013 7:30:33 PM EDT
Well, in terms of “a lot of melt” you seem to be suggesting that it’s air-temperature related but also that measuring temperature at the surface yields bad data? Do I have that about right?
I have your response re data source for reanalysis fields — that’s pretty similar to DMI actually.
You’re saying they make a mistake by giving surface temps and that higher elevation tells you when melt will occur at the surface?
According to DMI, the surface is cooler than usual this summer so if melt is above usual that would suggest melting from below -agreed?
July 24, 2013 7:32:22 PM EDT
[*next is a 2nd e-mail in a row from me]
what latitude was used as basis to describe summer temps as significantly warmer than usual?
July 24, 2013 7:38:17 PM EDT
The thing is that at the surface during summer, the surface is melting, so it’s pretty isothermal – right at the melting temperature. Any additional energy (e.g., solar radiation) goes mostly towards melting ice and the temperature doesn’t increase. So surface temperature doesn’t really tell you much. That’s why we use the 925 mb level, about 3000 feet altitude, which is far enough above the surface to give a better sense of the temperature anomaly relative to average. We look at maps of the entire Arctic (roughly north of 50 N latitude) to assess the temperatures.
July 24, 2013 7:41:05 PM EDT
surprised that you use temps as much as 16 degrees south of the Arctic Circle
what’s the thinking on that?
July 24, 2013 7:42:09 PM EDT
Attached are 925 mb temperature anomalies for the month of June and for July 1-21. You can see the cooler than average temperatures (blue shades) over the Arctic in June, and the anomalously high temps (orange/red) in July.
July 24, 2013 7:44:28 PM EDT
and you are on record attributing anomalous north pole melting to these 925 mb temps — even though DMI has the region above 80N as below average — correct?
July 24, 2013 7:44:58 PM EDT
As shown in the images attached in my previous email, we plot maps to 50 N. It depends on the time of year for where we focus. In winter, we’ll look at more southern latitudes, near the ice edge. Now we’re looking mostly within the Arctic Ocean, north of 70 N.
We don’t calculate a single number like DMI does – we look at the regional patterns, which provides a better view of what’s happening in the Arctic than the single number.
July 24, 2013 7:49:17 PM EDT
[* second e-mail in a row from Walt, and it’s an important one]
It’s not clear how anomalous the north pole melt is this year. This is something we’ll look at in the future. The camera does seem to show a lot of water, but melt ponds aren’t uncommon in the high Arctic at this time of year.
The 925 mb temps being higher than normal indicate more heat in the Arctic. in my view, the DMI temperatures are not a terribly good indicator of temperature anomaly, at least not once melt has started (i.e., the temps go above the blue). I think they’re more useful for winter temperatures and perhaps the timing of when melt starts – i.e., when the temp rises above the blue line – which does appear to have been later than normal due to the cooler than normal conditions in June.
July 24, 2013 7:54:59 PM EDT
It’s interesting that you mention heat
wouldn’t you need accurate Relative Humidity values, over time, in order to assess how much heat is in the system relative to average?
this is something that drives me batty about the AGW debate generally
temp not equal to heat …
so, to state the obvious, if the 925 temps for July happen to correspond to an anomalously dry period at the 3,000-foot layer then you could just as well have *less* heat in the system, at that layer, than usual
why do you get to assume that RH stable?
July 24, 2013 8:19:27 PM EDT
Yes, humidity is an important factor. However, there’s not a lot of water vapor up in the Arctic, and particularly higher in the atmosphere, so it’s not likely to be a major factor – or at least not as big of a factor as it can be at lower latitude.
- e-mail exchange ends here –
There are many interesting things going on in the exchange. Among them:
- Walt initially claims the webcam shows “a lot” of Arctic melting, then backs away from that claim.
- During the same exact time period in early July, DMI has the North Pole and environs colder than average and NCEP/NCAR has the region warmer than average.
- DMI’s numbers are pegged to surface temperatures, and NCEP/NCAR’s are 3,000-foot temperatures.
- NSIDC indicated for a news article this summer that the Arctic, specifically the North Pole, was unusually warm, based on temperatures at the 3,000 foot level, but DMI hasn’t had one above-average day all summer so far.
- Walt agrees that relative humidity is worth considering when evaluating heat in the system but then says that this does not vary, even though he appears not to have data to support this. I’ll reach out to Walt and see if he wishes to chime in on this as a commenter.
- It has been a cool summer in the highest reaches of the Arctic, the coolest on record so far as the Danish Meteorological Institute is concerned, and this appears to be something that one of the highest ranking scientists at the National Snow and Ice Data Center does not recognize.