I’ve been investigating the NOAA OLR (Outgoing Long-wave Radiation) data to try to find where things have changed by more than the global average. I’ve found a ‘step change’ in the high latitudes which occurred between 2002 and 2006.
figure 1: OLR anomaly 60-90N 2000-2012
Curiously, this increase appears to be mainly in winter rather than summer:
figure 2: OLR 60-90N 1975-2012
This seems to fly in the face of received wisdom on increased summertime melt exposing the relatively warm ocean to the cold September air, which only gets a few hours daylight, reducing the diurnal TOA insolation of 100W/m^2 or so. Note this will be changed by cloud cover.
figure 3: TOA insolation through the year in W/m^2
So what is the cause of this? Reduced winter ice cover in the polar region? Warmer north Atlantic ocean surface? Less wintertime cloud cover? increased co2? The Arctic ocean SST anomaly rose over that period, but is this enough to cause the OLR anomaly? Why doesn’t the previous rise from 1995 do the same?
figure 4: Arctic SST anomaly: Courtesy of Bob Tisdale
It’s a bit of a mystery to me at the moment, but someone may have a ready answer. I just wish I’d spotted this before our visit to Cambridge.
Tallbloke's Talkshop 
Whenever I read of step-changes in polar temperatures the first thing I think of is that it could be due to the phase change of water freezing. Of course, to examine that further we properly need absolute temperatures, not temperature anomalies.
What is the data source for figure 1?
“Curiously, this increase appears to be mainly in winter rather than summer”
Forgive me if I’m way off the mark, Roger, but this stuff from DMI always shows the greater temperature variability (in the FAR north +80degN) is in the winter not the summer.
http://ocean.dmi.dk/arctic/meant80n.uk.php
Anthony:
I used KNMI’s online graphing facility. It’s NOAA data.
Thanks marchesarosa, the modeled re-analyses doesn’t seem to show anything very remarkable for the 2002-2006 period, but it’s a good resource anyway.
I am always interested in the Arctic so I decided to do a bit of investigation, here is result without any comment from me:
Models with thicker ice in the annual mean
typically simulate less net longwave heat loss at the surface
during the cold season months (November through March).
This is particularly true for winter (DJF) when the correlation
between annual mean thickness and net longwave
radiation is 0.7. If the sea ice were primarily responding to
the net longwave radiation, thinner ice would be associated
with a less negative net longwave budget (less cooling), a
negative correlation. Thus, the positive sign of the correlation
suggests that ice conditions are influencing the net
longwave forcing instead of responding to it. Thicker ice
generally inhibits upward heat conduction through the ice,
resulting in a lower surface temperature. Interestingly,
there is no significant relationship between annual ice
thickness and the downward or upward flux components—
the relationship only holds with respect to the net longwave
flux. The reason is that the surface temperature responds to
both the incoming radiation (which has little ice thickness
dependence) and the conductive flux through the ice
(which is strongly thickness dependent). Consequently, in
models with thicker ice, there is less outgoing longwave
radiation to counteract the incoming flux, giving rise to a
significant relationship only with the net longwave flux.
Click to access week_10_holland_et_al_2008.pdf
Increased jetstream meridionality bringing warmer air into the polar regions more often and for longer.
Most noticeable during the winter months when the temperature contrast is greatest between the cold generated at the winter pole and the temperature of the warmer inflowing air from warmer latitudes.
Meanwhile the outflowing cold air cools the mid latitudes.
If one looks at the South Pole station data a similar situation is revealed:
http://www.antarctica.ac.uk/met/gjma/
There seems to have been cooling from 1958(-48.9) to 1998 (-50.5) during a period of increasing zonality.Though there are outliers of -50.6 in 1988 and -51 in 1983.
Then warming from 1998 (-50.5) to 2011 (-49.0) during a period of increasing meridionality.
Obviously some months have changed more than others due to seasonal windflow variations but breaking it down that way confirms my diagnosis.
I’ve been saying for over 4 years that the zonality / meridionality change occurred around 2000 and there we see the effect even at the South Pole.
In the decade pre 2000 6 years went below -50 at the South Pole.
In the decade after 2000 only 1 did so.
In the first 18 years of the record none did so.
Vulcevic, I went looking for trend information regarding Arctic Winter temperatures, with limited success.
While it only gives a crude measure, it does indicate that Arctic warming since 1996 has taken place mostly in the Winter, with Summer temperatures mostly unchanged. A more detailed look at the winter temperature data, (which you can probably find more easily than me) might throw more light on this, especially if a step change in winter temperatures matches Mr. Watts’ suggested step change in OLR.
[Reply] It’s not “Mr. Watts’ suggested step change in OLR” it’s mine. He asked me where the data came from above.
Coincides with the zero crossing of the PDO? AMO has finished it enso driven step increases, is flat and should soon start cooling (dumping heat). Is there data that shows the last zero crossing of the PDO which coincides with the AMO starting this last ramp up? Seems like a regime shift, opposite to the 77′ shift.
Anything happening with arctic ozone , solar uv, polar vortex? Maybe this is normal for the negative cycle of the pdo…a jet stream shift that accompanies it, or lower local winter cloud cover. Probably linked with the stall of the AMO, initiated by the PDO flip.
I think a warmer open ocean surface is the most likely answer to this. I’ll try some narrower OLR datasets to see if I can pin down where the extra energy left Earth from.
Glad you have given this your attention Roger.
As I understand from various texts the equatorial area to about 40 degrees in each hemisphere is in radiation surplus and from 40-90 is in radiation deficit compared with the temperature experienced. Therefore heat which enters at the equator and thereabouts is redistributed to warm the rest of the globe.
and exits most easily in high latitudes due to low water vapour.
I see this step change as a response to the global step change somewhat earlier. As global temperatures have been rather flat, perhaps the thermostat is working.
On the other hand you could read it as the signal of (permanent) arctic warming which we have been promised.
If global temperatures were still rising rapidly that is how I’d see it, actually, but with the record as it is, and much of the equatorial ocean with very unremarkable temperature anomalies at the moment and having some faith in the low solar outlook I think this is the reponse to a rather sudden upward perturbation in the 1990s. Can’t prove it though.
OLR is under “daily fields” at KNMI Climate Explorer.
After finding it and plotting it, note the following option over in the right-hand menu:
“Investigate this time series
View per month, season, half year or full year”
Here’s the “view per month” – (not sure this link will work):
http://climexp.knmi.nl/plotseries.cgi?id=someone@somewhere&TYPE=i&WMO=noaa_olr_0-360E_60-90N_n_0p&STATION=NOAA_OLR_0-360E_60-90N&NAME=Index&KIND=month
And “view per season”:
http://climexp.knmi.nl/plotseries.cgi?id=someone@somewhere&TYPE=i&WMO=noaa_olr_0-360E_60-90N_n_0p&STATION=NOAA_OLR_0-360E_60-90N&NAME=Index&KIND=season
Let’s see if direct links to the graphs on those pages will work:
by Month:
by Season:
“I think a warmer open ocean surface is the most likely answer to this.”
Isn’t that inconsistent with the largest effect being in winter when there is no open water ?
To get the observed result we really need an increase in warm air advection towards the pole in winter which is where jetstream meridionality comes in.
Paul, thanks a lot, the first link of the second set works fine and give monthly views, most helpful.
We can see the seasonal cycle here, with the increased OLR from August through to January inclusive beginning back 1995 (nod to Geoff C).
The amplitude of the Feb-June rise is much smaller, so this does seem to be an late-summer/autumn/early-winter phenomenon. Which is what we’d expect with reduced ice taking longer to cover up the ocean again after the September minimum.
The next step I think, is to try to narrow it down further by checking the amplitudes of increased OLR for various longitudinal sectors, and comparing that result with cryosphere maps of where the ice had diminished markedly from 1995. Suggest we split it into Canadian Arctic, N.Atlantic amnd Scandinavia, and Siberia.
Then look at sst’s for those areas compared to others, and see if we can divine anything about the cloud cover situation from the types of movements we see.
Then there are wind patterns to consider in order to test Stephen’s idea on jets.
Maybe WUWT will have some useful dataset links?
Is there similar data for OLR from the South Pole ?
That could help distinguish between a water heating effect and any effect from more advection of warm air.
Stephen Wilde (September 26, 2012 at 12:03 pm) wrote:
“To get the observed result we really need an increase in warm air advection towards the pole in winter which is where jetstream meridionality comes in.”
Absolutely. I am very pleased to see you refining your narrative to distinguish unambiguously between the fundamentally different circulatory morphologies of winter & summer. Excellent!
Now, the next step is to ponder STEEP land-sea temperature contrasts along the winter Eurasian east coast. (The North American east coast gradient is also highly noteworthy, but it’s not nearly as strong.)
For those of you working on cross-narrative synthesis, you can dovetail in Bob Tisdale’s KOE “spin-up”, which multidecadally amounts to PDO & -NPI. (Gulf Stream for the smaller continent / ocean pair.)
For beautifully clear exposition on equator-pole gradients & consequent flow, see the work of Jean Dickey (NASA JPL).
–
Stephen Wilde (September 26, 2012 at 12:22 pm) wrote:
“Is there similar data for OLR from the South Pole ?”
KNMI Climate Explorer. You’ll find it’s user-friendly.
Dammit, lets try that again.
“I think a warmer open ocean surface is the most likely answer to this. I’ll try some narrower OLR datasets to see if I can pin down where the extra energy left Earth from” Tallbloke
Alternatively it may be an area effect.
The Arctic ice extent figures for April show a sustained decline between 2003 and 2006. January and March show a similar decline, though it is not as clear for february.The extra open sea area would have a warming effect.
Eastman & Warren, J of Climate, May 2012, seem to have the data regarding cloud cover. A 39-Year Survey of Cloud Changes from Land Stations Worldwide, 1971-2009.
Cloud cover seems to have INCREASED over the Arctic during this time, but whether it is winter or summer, dunno.
[ snip ]
[ snip ]
[ no emails please, co-mod]
They would probably be able to sort data by season.
Entropic man,
“The extra open sea area would have a warming effect.”
Don’t you mean the open sea radiates more so would have a cooling effect showing in more OLR??
Hi TB. I don’t know if this is so, but, to me, this looks like an acceleration in the hydrological cycle at the N Pole.
Besides the issue of summer/autumn cloud albedo change, cloud droplets ‘freeze’ in winter. Thus, a possible increase in OLR from the phase change of greater summer cloud cover during the winter period.
Best regards, Ray.
OK, here are the anomalies for the zones I outlined:
Note the Y axis scale changes
Siberia 30E-150E
Canada 150E-270E
Atlantic 270E-30E
Looks like most of the action is in the Atlantic and to a lesser extent Canada. This makes sense, as the currents bringing warmer water from the south predominantly come up through the Fram Strait. But there is more marked ‘step change’ in the Canadian sector in 2002-3.
Since the effect is clear over land masses such as Canada and Siberia doesn’t that suggest warmer air being advected across those regions ?
That would give more clouds in polar regions notwithtanding reduced cloudiness in the subtropics.
Suricat’s sggestion of a faster hydrological cycle works for me.
Since the Arctic tends to have a strong inversion through the winter, I would tend to think that the changes are largely atmospheric, that being where the OLR is emanating from. Firstly from an increase in negative AO and NAO from 1995 causing increased exchange of Arctic air with warmer air from beyond the Arctic circle:
http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/month_ao_index.shtml
http://www.cpc.ncep.noaa.gov/products/precip/CWlink/pna/month_nao_index.shtml
you can see it working the other way in the early 1990’s with +AO and +NAO and less OLR.
And secondly the increase in Sudden Stratospheric Warming events since 2000 providing more atmospheric mixing when the vortex breaks down. In fact they should be able to compound each other, or work against each other, which may be the case in 2000 and 2002 where AO/NAO are positive during SSW events.
http://www.geo.fu-berlin.de/en/met/ag/strat/produkte/northpole/index.html
SSW data
Lower Troposphere North Pole Ocean monthly anomaly since Dec 1978:
(point 200 on the scale is June 1995)
Source: http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
Thanks Stephen and Ulric. In general, lower troposphere temperature lags behind sea surface temperature b a few months. So maybe the there is a combination of ‘solar change from above’ coupled with SST’s going on which then lead to changes in the much smaller heat capacity AO and NAO air.
Solar cycle 23 peaked in 2002 and then started the long slow tailoff. The energy comes back out of the ocean as the sun goes quiet. That energy heats the air. The air then loses that energy to space as OLR.
Hi, I was also looking at Arctic but from the point of view of TOTAL annual ice cover data, rather than the usual blinkered look at one day of the annual min.
I found that if you look at all the daily data there was a change in pattern during 1997-2007 and after that year the earlier cyclic behaviour returned with an almost identical form.
I thought to compare it to full NH lower tropo, a nine month shift aligns the variation fairly well… except for the period that was seeing streadily accelerating melting, where the correlation was weak or even inverse.
While looking at the plot I noticed that the peaks and troughs in the UHA TLT plot were almost as regular a clockwork. About 7.5 years. I’m very caucious since gaussian filters tend to make any data look like nice cosine curves but the regularity of the events is notable.
Now I thought that is right up tallbloke’s street, so I came here to post it and found this article which ties in quite nicely.
ice data is from:
ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/north/daily/data/
UAH TLT from:
http://vortex.nsstc.uah.edu/public/msu/t2lt/tltglhmam_5.4
Maybe there is some explanation or tie-in of the breakup and restoration of the cyclic pattern in the energy discussion here.
I think the graph shows that the accelerated melting finished pretty abruptly in 2007 if well look at all available data , not just one day a year.
Hope it’s of interest.
Tallbloke, a suggestion:
Take a look at OLR for lower NH latitude bands.
The impact in the Arctic shows fascinating thresholds & nonlinearities (paradoxical statistical misinterpretation trap livina & lenton fell into), but the origin is further south.
Stephen’s multidecadal perspective is entirely compatible with your annual insights. There’s tight ((multi-year-ice-damped) seasonally nonlinear) equator-pole coupling, but Stephen & I should be a little more careful about winter advection.
Apologies for economy of words — busy week. Thanks for initiating a very interesting exercise.
Kukncat
Winter maximum ice area is down from 15millionKm^2 to 13million Km^2 since 1979.
http://arctic.atmos.uiuc.edu/cryosphere/arctic.sea.ice.interactive.html
2012 Winter temperatures ran 5C above the long term average, while the Summer temperatures were average.
http://ocean.dmi.dk/arctic/meant80n.uk.php
The minimum ice extent trend slopes more steeply from 2002 to 2006.
Vulcuvec describes of the link between thinner ice and higher Winter temperatures. I suggest that we are seeing more warm water moving into the Arctic in Winter. This could cause a decrease in ice thickness, allowing more radiation to escape upwards from the water beneath, and a decrease in ice area, allowing more direct radiation. Both would increase OLR.
Anyone got a temperature record for the water under the Arctic ice? It would be interesting to look for a step change around 2002
PS UAH data is from monthly “anomalies” , full NH , this may affect the lag. I will have to try to add back his “climatology” to get real data.
Given that the freezing of water (salt or fresh) releases a very large amount of energy it would seem likely that the ‘extra’ ice being made each year could explain the ‘extra’ OLR in the winter. It does seem like a sensible candidte anywy.
tallbloke says:
“In general, lower troposphere temperature lags behind sea surface temperature a few months. So maybe the there is a combination of ‘solar change from above’ coupled with SST’s going on which then lead to changes in the much smaller heat capacity AO and NAO air.”
The air warmer air would be coming in from outside the Arctic. More likely the SST change would occur at the same time as the AO/NAO changes.
I’m becoming more and more confident that the primary causation in the first instance is an effect on the shape of the polar vortices from chemical changes in the atmosphere (more pronounced above the poles) induced by solar variability acting differentially on ozone amounts at different heights.
When the sun is active the polar vortices are strong vertically and contracted horizontally which pulls the jets and climate zones poleward allowing more energy into the oceans.
When the sun is inactive the vortices are weak vertically but flop about horizontally pushing the jets and climate zones equatorward allowing less energy into the oceans.
The consequences work through both air and ocean circulations on different timescales but overall an active sun gives less clouds and warming oceans whilst an inactive sun gives more clouds and cooling oceans.
The system response is to increase the speed of the hydrological cycle to eliminate extra warming when the sun is active and to decrease the speed of the hydrological cycle to try and preserve warmth when the sun is less active. In other words whatever happens the system response is negative and not positive as proposed by AGW theory.
As far as I can see that general mechanism covers all known observations including the OLR changes which are the subject of this thread.
RichardLH says:
September 27, 2012 at 5:29 pm
Given that the freezing of water (salt or fresh) releases a very large amount of energy it would seem likely that the ‘extra’ ice being made each year could explain the ‘extra’ OLR in the winter
Yes, but nonetheless, wintertime OLR is much less than summer time OLR as fig 2 shows. And winter ice diminished over the time period.
Stephen,
sounds about right. The active sun’s destruction of ozone would let more radiation out from the troposphere, without it being buffered in stratospheric ozone. So, cooler stratosphere as warming occurs, but not for the reason the AGW theorists posit.
Following up Paul Vaughan’s recommendation, here are the plots for the 30-60N regions
Note vertical axes vary.
‘Siberia’ 30E-150E 30-60N
‘Canada’ 150E-270E
‘Atlantic’ 270E-30E
If the Arctic is the ‘canary in the coalmine’ for global warming, then so it is for global cooling as well. Notice that compared with the previous three plots, the earlier ones from 60-90N are showing a marked reduction in OLR from 2010 for the Canada and Atlantic regions.
Ulric Lyons says:
September 27, 2012 at 3:05 pm
Lower Troposphere North Pole Ocean monthly anomaly since Dec 1978:
http://snag.gy/RwQte.jpg
Interesting that the jump in LT occurs around the same time as the big spike in OLR. I had assumed it was a data splice. Now I’m not sure. What happened in 1995? Big solar event?
I cant find much detail, but a significant solar flare was recorded in 1995.
http://www.networkworld.com/community/blog/biggest-solar-shot-1995-earth-bound
There’s are also two references to a flare dated November 19th 1995. Whether it’s the same flare is not clear.
Entropic, Nov ’95 looks a bit late. Thanks though.
According to P Solar’s graph there was a big swing in rate of change of sea ice extent at the start of ’95
There was a flare in January 1995. I’m not sufficiently au fait with this area to judge its significance.
Click to access 2008_mnoras_1905.pdf
tallbloke says:
September 27, 2012 at 8:44 pm
RichardLH says:
September 27, 2012 at 5:29 pm
Given that the freezing of water (salt or fresh) releases a very large amount of energy it would seem likely that the ‘extra’ ice being made each year could explain the ‘extra’ OLR in the winter
Yes, but nonetheless, wintertime OLR is much less than summer time OLR as fig 2 shows. And winter ice diminished over the time period.
Is ihere any spectrum breakdwn of the OLR yearly pofile? f so the vaious contibutions should be exposed surely?
Stephen Wilde says: September 27, 2012 at 1:39 pm
“Suricat’s sggestion of a faster hydrological cycle works for me.”
Slow down Stephen, this is a complex issue and, AFAIK, can only be derived from first principles. Let’s look at the differences between ‘ice’ and ‘no ice’ scenarios at the N Pole with the hydrological cycle in mind.
For Ocean:
With the polar cap covered with sea ice;
the ‘partial pressure’ (roughly, the energy needed to break molecular bonds and change the phase of a compound [rationalised as latent heat]) for the evolution of ‘water vapour’ (WV) from ice is high, ‘melt pools’ during summer months provide better opportunity for the evolution of WV, ice inclusions (e.g. black carbon) assist ice melt via albedo change during periods of insolation (post industrial era), sea ice is thermally protected from ocean heat by the -4C expansion anomaly exhibited by H2O in its fluid phase, old (broken) ice improves the ‘impeller’ action that facilitates movement in the planar centrifuge that generates the Polar Climate Cell.
With a water covered polar cap;
WV is more easily evolved from the sea surface (the ‘partial pressure’ is lower), black carbon sinks in water and only alters albedo for the more penetrating spectra of albedo, atmosphere is protected from ocean heat by the -4C expansion anomaly exhibited by H2O in its fluid phase (the ‘insulation’ factor provided by a layer of ice is gone and the ocean is in direct contact with the atmosphere), open ocean (or new ice) provides a poor ‘impeller’ action that facilitates movement in the planar centrifuge that generates the Polar Climate Cell.
I’m sure you’ll recognise differences there that imply changes in the hydrological cycle from the ‘ocean forcing’ side of the scenario. 😉
For Atmosphere:
With the polar cap covered with sea ice;
sea ice inhibits the evolution of WV (by means of its insulating boundary) and reduces the ability of energy transfer by H2O’s most recognised mechanisms (sensible and latent heat [which is mostly observed in the lower tropo]), little cloud cover from this scenario interprets to a minimal insolation thermal capacity (the ‘mass’ isn’t there to provide an appreciable thermal capacity), the ‘impeller action’ from ‘old ice’ enhances the energy imparted to the Polar Climate Cell and amounts to the energy received from the surface (ocean) and the atmospheric reactance.
With a water covered polar cap;
H2O is abundant. All hell lets loose, I’ll deny any such post. 😉
Best regards, Ray.
P. Solar (September 27, 2012 at 5:17 pm) wrote:
“PS UAH data is from monthly “anomalies” , full NH , this may affect the lag. I will have to try to add back his “climatology” to get real data.”
It’s reassuring to see careful thinking about interpretation of stats.
tallbloke (September 27, 2012 at 9:30 pm) asked:
“Now I’m not sure. What happened in 1995? Big solar event?”
Have a look at (a) solar proton flux & (b) total column ozone:
(a) http://htmlimg2.scribdassets.com/22o1pro2yo1n12si/images/7-0ee31fdd29.jpg
(b) http://i45.tinypic.com/bfxn4.png
Big proton fluxes eat ozone viciously (to put it loosely in plain language).
Someone with time/interest might start looking at ozone by region. I’ll follow along and if time/resources ever permit, I’ll attack this with a vengeance.
I’m not yet convinced that Corbyn’s right about ozone being the agent. Based on the total collection of data exploration I’ve done to date, I can’t rule out the possibility that ozone’s just a strong symptom.
(So many questions. Not enough time/resources.)
tallbloke says:
September 27, 2012 at 9:30 pm
“What happened in 1995?”
Apart from the spike itself, plasma speed dropped well through 1995/96/97:
Paul Vaughan says: September 28, 2012 at 4:08 am
“I’m not yet convinced that Corbyn’s right about ozone being the agent. Based on the total collection of data exploration I’ve done to date, I can’t rule out the possibility that ozone’s just a strong symptom.”
AFAIK, ozone in the stratosphere IS “just a strong symptom” of the insolation of the day. UVc and UVb are the expected antagonists for the production of O3 in the stratosphere and this process is mostly aligned with ‘sunspot activity’, when UV is strongly propagated.
The atmospheric chemistry of O3 takes place secondary to the evolution of O3. If the O3 wasn’t there, the atmosphere couldn’t react to it. 🙂
Best regards, Ray.
I realize North Atlantic catches eyes — not aiming to spoil that enduring party but want to suggest we keep our eyes on the GLOBAL ball …so here’s some counterbalance:
“Figure 1. (a) Annual mean PDO and SOI indices, (b) annual mean zonal wind stress along equatorial Pacific (averaged over 5°S–5°N, 130°E–100°W) derived from different atmospheric reanalysis products, and (c) annual mean detrended Fremantle sea level anomalies.”
Feng, M.; Böning, C.; Biastoch, A.; Behrens, E.; Weller, E.; & Masumoto, Y. (2011). The reversal of the multi-decadal trends of the equatorial Pacific easterly winds, and the Indonesian Throughflow and Leeuwin Current transports. Geophysical Research Letters 38, L11604. doi:10.1029/2011GL047291.
Click to access 2011FengGRL.pdf
abstract: http://www.agu.org/pubs/crossref/2011/2011GL047291.shtml
“The equatorial Pacific easterly winds have rejuvenated and experienced a multi-decadal strengthening trend since early-1990’s (Figure 1b) [Feng et al., 2010]. There have been reports on the recovery of the strength of the Pacific subtropical cells during the period of early-1990’s to early-2000’s [McPhaden and Zhang, 2004].”
(I illustrated this coupling strengthening some time ago, but the university took down my graphs when I moved to the private sector.)