Guest post by Dr. Walt Meier, National Snow and Ice Data Center
Winds, Temperatures, and Arctic Sea Ice Extent
As the summer sea ice melt season gets into high gear, I thought I’d do a post on sea ice processes and other tidbits about sea ice that may be useful as people watch the seasonal sea ice extent decline. My thanks to Anthony for the opportunity to share this information.
Often, much of the focus in the news is on the effect of warming air temperatures on observed decline in Arctic sea ice extent, such as in the The Economist article. Others have suggested, such as in last Saturday’s post, that winds are the key to understanding the extent decline. These are not competing viewpoints, but reflect complementary contributions to changes in sea ice extent. For a full description of how sea ice changes – day-by-day, month-by-month, and over the years and decades – both wind and air temperatures (along with other factors, e.g., the oceans) need to be considered.
Winds and daily variations in extent
Winds primarily affect sea ice extent by pushing ice around, either spreading the ice out over larger area (increasing extent) or compressing it into a smaller area (decreasing extent). Often, day-to-day changes in sea ice extent are primarily due to changes in winds and not freezing or melting. The winds can also open areas of water within the ice-pack, called leads, if they push floes of ice apart. Thus, even during winter, there are open water areas or areas of thin ice (as leads begin to re-freeze) throughout the ice-pack. It is this feature that has allowed submarines to surface at the North Pole since the 1950s, even though the overall sea ice thickness was much greater in the 1950s compared to today. (In other words, surfacing subs at the North Pole are not an indicator of Arctic sea ice conditions.)
Winds and interannual changes in extent
Winds are variable, blowing at different directions and speeds. Thus over time, the effect of the winds settles into an average pattern and their net effect on extent is smaller relative to temperatures. However, average wind patterns can themselves vary over longer periods of time due to large-scale climate oscillations, most notably for the Arctic Oscillation (AO). During the late 1980s and early 1990s, the AO was often in a positive mode that favors the motion of older, thicker sea ice out of the Arctic. The remaining younger, thinner ice cover was more easily melted completely in the subsequent summers. This contributed to some of the summer extent decline during that period, as was noted in papers by Rigor and Wallace (2004) and Rigor et al. (2002). However, in recent years, this relationship appears to have broken down. After very strongly negative AO winters in 2009-2010 and 2010-2011, the summer sea ice again reached low levels (Stroeve et al., 2011).
Winds and summer extent
Even over a season, variation in the winds can play an important role. They were a key factor in the record low extent of 2007, as noted for example by Ogi and Wallace (2012) and Zhang et al. (2008) , who found that ~30% of the record low extent could be attributed to unusual ice motion (driven by the winds). According to Ogi et al. (2010), 50% of the year-to-year variation in extent can be explained by the variation in winds. Ogi and Wallace (2012) noted that if the wind patterns were similar to 2007, the minimum extent during 2010 and 2011 would have likely been as low as or lower than 2007.
Effects of winds and temperature on long-term changes in sea ice
Winds can also influence the long-term trend in extent. Ogi et al. (2010) estimated that up to 33% of the trend for 1979-2009 could be explained by winds. One mechanism for this long-term influence is via long-term changes in the winds, which have been noted by Ogi et al. (2010) and Smedsrud et al. (2009). Another effect on extent due to winds is in how effective winds are pushing the ice around. Spreen et al. (2011) noted that while some increase in wind speed is observed (in agreement with the Ogi and Smedsrud papers), the speed of the ice increased much more. In other words, the winds are becoming more effective at pushing the ice around.
The motion of sea ice is affected not only by winds (and other smaller factors), but also by the ice itself. Thinner ice is more easily pushed around by the winds than thicker ice (Haas et al., 2008). And the sea ice cover has been getting substantially thinner through the loss of older, thicker ice (e.g., Maslanik et al., 2011; Kwok and Rothrock, 2009). Zhang et al. (2008) found that thinner ice cover was a crucial factor in the 2007 ice loss and if the ice pack were thicker, a record low would not have occurred under the same winds. As mentioned above, some of this loss can be ascribed to the positive AO of a couple decades ago. However, since then the AO has been in a mostly neutral or negative mode and yet older ice has continued to be lost. For those interested, a nice animation of changes in ice age can be seen at the NOAA Climate Watch website.
Thus, the long-term thinning trend is primarily a reflection of additional energy from globally warming temperatures. Thick ice still moves out of the Arctic (or melts within the Arctic), but the additional energy in the Arctic prevents the replenishment of thicker ice at the same pace. The system is out of equilibrium and older, thicker ice continues to decline (though with some year-to-year variability). The additional energy is not always indicated by warmer local air temperatures though, especially in ice-covered areas. The Danish Meteorological Institute (DMI) air temperature estimates for north of 80° N, shows summer temperatures just above freezing during summer and there is little or no trend. This is because the additional energy is used to melt the surface of the ice and not warm the atmosphere, which stays near the melting point through the summer.
Conclusion
So, overall, the long-term decline in sea ice is mostly due to increasing temperatures leading to thinner ice cover that is more easily melted completely during summer. Winds accelerate or slow the long-term decline through the motion of thick ice out of the Arctic for period of up to a few years. The effects of winds may have longer-term consequences because their effect on ice motion increases as ice thins. The interplay between the two factors – wind and temperature – is perhaps best exemplified by the estimates for September sea ice extent in the recently released Sea Ice Outlook. There is a wide spread between different outlooks – about 500,000 square kilometers; even the uncertainties of a single method are on the same order of magnitude. That 500,000 square kilometer uncertainty reflects uncertainty in how the winds will vary this summer. However, all of the outlook contributions are more than 1.5 million square kilometers below normal, which demonstrates the effect of the long-term warming trend.
References
Haas, C., A. Pfaffling, S. Hendriks, L. Rabenstein, J.-L. Etienne, and I. Rigor (2008), Reduced ice thickness in Arctic Transpolar Drift favors rapid ice retreat, Geophys. Res. Lett., 35, L17501, doi:10.1029/2008GL034457.
Kwok, R., and D. A. Rothrock (2009), Decline in Arctic sea ice thickness from submarine and ICESat records: 1958–2008, Geophys. Res. Lett., 36, L15501, doi:10.1029/2009GL039035.
Maslanik, J., J. Stroeve, C. Fowler, and W. Emery (2011), Distribution and trends in Arctic sea ice age through spring 2011, Geophys. Res. Lett., 38, L13502, doi:10.1029/2011GL047735.
Ogi, M., K. Yamazaki, and J. M. Wallace (2010), Influence of winter and summer surface wind anomalies on summer Arctic sea ice extent, Geophys. Res. Lett., 37, L07701, doi:10.1029/2009GL042356.
Ogi, M. and J. M. Wallace (2012), The role of summer surface wind anomalies in the summer Arctic sea ice extent in 2010 and 2011, Geophys. Res. Lett., 39, L09704, doi:10.1029/2012GL051330.
Rigor, I.G. and J.M. Wallace (2004), Variations in the Age of Sea Ice and Summer Sea Ice Extent, Geophys. Res. Lett., v. 31, doi:10.1029/2004GL019492.
Rigor, I.G., J.M. Wallace, and R.L. Colony (2002), Response of Sea Ice to the Arctic Oscillation, J. Climate, v. 15, no. 18, pp. 2648 – 2668.
Smedsrud, L. H., Sirevaag, A., Kloster, K., Sorteberg, A., and Sandven, S. (2011), Recent wind driven high sea ice area export in the Fram Strait contributes to Arctic sea ice decline, The Cryosphere, 5, 821-829, doi:10.5194/tc-5-821-2011.
Spreen, G., R. Kwok, and D. Menemenlis (2011), Trends in Arctic sea ice drift and role of wind forcing: 1992–2009, Geophys. Res. Lett., 38, L19501, doi:10.1029/2011GL048970.
Stroeve, J. C., J. Maslanik, M. C. Serreze, I. Rigor, W. Meier, and C. Fowler (2011), Sea ice response to an extreme negative phase of the Arctic Oscillation during winter 2009/2010, Geophys. Res. Lett., 38, L02502, doi:10.1029/2010GL045662.
Zhang, J., R. Lindsay, M. Steele, and A. Schweiger (2008), What drove the dramatic retreat of arctic sea ice during summer 2007?, Geophys. Res. Lett., 35, L11505, doi:10.1029/2008GL034005.
==============================================================
Latest graphs are on the WUWT Arctic Sea Ice page
Related articles
- The Economist Provides Readers With Erroneous Information About Arctic Sea Ice (wattsupwiththat.com)
- Global warming: Arctic sea ice extent dips toward new lows (summitcountyvoice.com)
- Will Arctic Sea Ice Reach Record Low This Year? (news.discovery.com)
From Dr. Meiers article and from a number of the comments that correctly point out dozens of other factors that influence sea ice, it is quite clear that even something as seemingly simple as ice melting, is, in fact, extremely complex. Very difficult to model or come up with a complete understanding. Given that we only have 30 years of data, it is obviously impossible to say at this point that we know what the cause or future trends will be. Very large error bars should be assigned. With another 15 years of study, however, we will be in a much better position to determine the causes and to make predictions. As a scientist myself, I want to see more data before I come to a conclusion. That’s really the only responsible position to take.
Thank you, Dr Meier. Your description matches almost perfectly with what I have been thinking. I also appreciate your use of the term “global warming” rather than AGW. Most of us accept that the planet was, in fact, warming for most of the satellite data period up to around 2006. This, no doubt, led to some thinning of the ice that allowed winds to be more effective than they would have been otherwise.
I have personally predicted that we will not see a recovery of the ice until 2020 at best even with the current leveling off of temperatures. It will take a long time to replenish the ice when it can so easily be moved around. I’m sure I’ll be watching it closely when I’m not spending time doing my other equally exciting hobby … watching grass grow.
I also appreciate Walt’s willingness to engage here. I find the phrase “long-term decline” tough to swallow, nonetheless. Why? Because waves striking the shores of northern Greenland have been shown to have been generated by a mostly open Arctic Ocean during the Holocene Climatic Optimum. These same waves are not currently striking northern Greenland, because the Arctic Ocean is too clogged with ice to allow them to form. So, unfortunately, a few decades’ worth of satellite imagery is taken as indicative of “climate change” when we know perfectly well that the entire system has been in flux for millions of years, including during the remarkably stable Holocene. Walt and I have corresponded enough for me to know that he is utterly convinced that the three decades’ worth of evidence is proof of a human fingerprint, worrisome, and probably unique. I respectfully disagree on all three counts.
Dr Meier, I don’t understand your quantum leap in attribution. You wrote well about the impacts of winds, and then suddenly make the statement: “Thus, the long-term thinning trend is primarily a reflection of additional energy from globally warming temperatures” You invoke dubious subtraction to imply greenhouse warming by ignoring the warmer waters that lie below the surface which are pumped into the Arctic. The volume of warm water entering the Arctic has been demonstrated by moorings to have increased in accord with a positive PDO and NAO. Now that the PDO has gone negative that source of warm water entering through the Bering Sea has slowed and cooled, and accordingly we also see tremendous growth of ice in the Bering Sea. As demonstrated by Shimada’s 2006 paper “Pacific Ocean inflow: Influence on catastrophic reduction of sea ice cover in the Arctic Ocean”, the loss of sea ice allows the winds to couple the surface withe deeper warmer layers. Despite the cooling of upstream waters from the Bering, there was continued sea ice loss as wind turbulence brought warmer waters to the surface. Likewise the great loss of sea ice in the Barents Sea is due to warm Atlantic intrusions that similarly generated the similarly warm temperatures in the 1930’s. (See BENGTSSON 2004) Despite the changes in the wind that initiated the ice loss, the winds are still churning the open waters and heat is being brought to the surface and vented. This process would create higher temperatures in the winter ice seaon.The lack of a summer trend of rising temperatures is further testimony that high Arctic temperatures are due to heat ventilation due to less ice, not as you imply that less ice due to more heat. That ventilation also suggests a deeper cooling has been taken place and predicts that thicker sea ice will soon return.
Pardon me if I sound ignorant on this because I’m by no means an expert on oceanography, but it seems to me that you are missing part of your equation. The atmosphere above the ice is important I’m sure, but wouldn’t the effects of the sea under and around the ice be just as important to understanding what’s really going on? I would think it would be premature to conclude that the temperature is warming and that’s causing the sea ice to shrink without being able to rule out any possible negative effects from the currents themselves, and they can take decades or centuries to turn over, so 40 years of data wouldn’t really help you to establish a pattern there.
“The Danish Meteorological Institute (DMI) air temperature estimates for north of 80° N, shows summer temperatures just above freezing during summer and there is little or no trend. This is because the additional energy is used to melt the surface of the ice and not warm the atmosphere, which stays near the melting point through the summer.”
I have always had a problem with this concept. The same thing has been argued about the oceans absorbing the heat energy that increasing CO2 is generating and that is why the air temperatures are not warming.
Doesn’t the whole process start with a warming atmosphere? How can the additional energy be “used to melt the surface of the ice and not warm the atmosphere” when the additional energy (as heat) is in the atmosphere? That is like having a pot of boiling water on the stove that is heated to exactly 100 degrees C. Dr. Meier and others appear to be arguing that it is possible to make the water boil faster by adding heat to the pot, but not have the pot get any hotter, just the water inside the pot. Seems to me, if the pot isn’t getting any hotter, neither is the water inside of it.
Am I missing something?
Warmer Atlantic waters are reaching deeper into the Arctic Ocean which could in the future keep the Arctic ice free even in the dark winter. New species will move in. Open water means erosion of coastal areas which were previously protected by coastal ice. The much feared melt of permafrost which seems now inevitable and the consequent scape of CH4. There is far too much going on and nothing to stop it. The Siberian tundra has again paid the price of a massive temperature anomaly which has been over the area for the last 2 months… this has now resulted in massive Tundra fires which are dwarfing those in 2010.
Warm says:
June 21, 2012 at 12:53 am
Dr Meier addressed this issue:
“The additional energy is not always indicated by warmer local air temperatures though, especially in ice-covered areas. The Danish Meteorological Institute (DMI) air temperature estimates for north of 80° N, shows summer temperatures just above freezing during summer and there is little or no trend. This is because the additional energy is used to melt the surface of the ice and not warm the atmosphere, which stays near the melting point through the summer.”
==============================
Thank you for pointing this out Warm, but I did see that. The question I was asking is what effect CO2 will have on the variability of temperatures. They should become more uniform as they push upwards, yes? Since there should be warming occurring, wouldn’t one logical conclusion of that warming be more uniform temperatures as the melt season should begin sooner, expanding the length of time the less variable temperature exerts its influence on the record? Otherwise it sounds like, “It’s warming when it’s not.” which is kind of redundant.
Dr. Walt Meier (or anyone else) interested in the Arctic temperatures may find useful to know what NASA’s Jet Propulsion Laboratory has to say
http://www.vukcevic.talktalk.net/Arctic.htm
We are doooommmmed!!!! References would be nice next time. 😉
By the way your Russian fire link says it is mostly man-made with a little help from above.
Thank you for your post Dr. Meier! One small comment on “…surfacing subs at the North Pole are not an indicator of Arctic sea ice conditions”:
Most of us understand that ice free conditions at the North Pole don’t necessarily mean much. Kindly pass this information on to Dr. Serreze, the IPCC, and the media.
I’m always happy to see Dr. Walt here. I don’t always agree with him, but he plays fair, supports his arguments well and in detail, is genuinely interested in having a respectful dialogue, and doesn’t get overly excited (unlike some of his colleagues) in extrapolating short term trends.
Thank you for your calm assessment Dr. Meier.
I have long maintained that receding Arctic sea ice is a negative feedback with the albedo decrease being far exceeded by ocean energy loss by the open water losing energy to the atmosphere and thence to space.
It’s natures way of cooling the oceans.
DaveE.
‘The Danish Meteorological Institute (DMI) air temperature estimates for north of 80° N, shows summer temperatures just above freezing during summer and there is little or no trend. This is because the additional energy is used to melt the surface of the ice and not warm the atmosphere, which stays near the melting point through the summer.‘
Does anyone else feel that this last bit is just thrown in there? Is there any OBSERVATIONAL data that shows surface melting of the ice in the Arctic at those ‘just above freezing’ temperatures?
My experience with ice melting is rivers and streams in upstate NY…it generally doesn’t actually melt from above, usually the water below the ice increases (because it is not freezing when it falls, and the comparatively warm earth melts snow from below), warms up (it is almost always already above freezing), the slowly eats the ice from below. Only when Spring temps reach the high forties (F) does are temp begin to affect the thick winter ice.
Anyone else know anything about this? Any Arctic experts here at WUWT?
The DMI site shows, using the old eye=o-meter, that the last five years or so summer air temps have been a bit below the 40 yr average.
TomRude says:
June 20, 2012 at 10:54 pm
“So, overall, the long-term decline in sea ice is mostly due to increasing temperatures leading to thinner ice cover that is more easily melted completely during summer”
3) If arctic ice state was a novelty, unprecedented interrupting a state of perennial stability how do you explain the absence of 100y old ice, 50 y old ice in the 1980s?
It wasn’t absent just substantially diminished, e.g. the Ayles Ice sheet has recently broken and is a shadow of its former self.
4) Since you wish to minimize modes of atmospheric circulation to the role of 0.8 C global temperature rise (on land, if we can trust HadCrut etc…) how come the shape of summer arctic sea ice is always the same?
It isn’t! See here for example: http://nsidc.org/data/google_earth/index.html#sea_ice_index_sept
Kip Hansen,
Your exactly right and your intuitive sense is being confirmed. Most of the melting is coming from below due to warm water intrusions associated with positive PDO and NAO. A 2011 paper “Fate of Early 2000s Arctic Warm Water Pulse” by Igor V. Polyakov et al wrote :
These recent observations show the warm pulse of Atlantic Water (AW) that entered the Arctic Ocean in the early 1990s finally reached the Canada Basin during the 2000s. The second warm pulse that entered the Arctic Ocean in the mid-2000s has moved through the Eurasian Basin and is en route downstream. One of the most intriguing results of these observations is the realization of the possibility of uptake of anomalous AW heat by overlying layers, with possible implications
for an already-reduced Arctic ice cover.
…
from several cross-isobath sections spanning 43° to 185°E show consistently higher temperatures in the OL (overlying layer) in eastern sections than in the western sections. Since the AW layer is the closest source of heat, this leads us to suggest significant upward heat flux from the AW to the OL during the transit along the slope.
Jim Cripwell says:
June 21, 2012 at 6:23 am
An excellent paper. What I always hope is that when people like Walt seem to have such an excellent grasp on a situation, in this case Arctic sea ice extent, they dont use this knowledge to forecast what will happen in the near future. Canadian Ice Services has, for two years, put in a forecast to ARCUS for minimum Arctic sea ice extent. One wonders why NSIDC is absent from the list of ARCUS forecasters. Surely their knowledge should be second to none, and their forecasts amongst the most accurate.
They have indeed submitted a forecast, authored by Julienne Stroeve, Walt Meier, Mark Serreze, Ted Scambos, & Mark Tschudi, their average predicted extent is 4.59. Their June 2011 forecast was pretty much bang on.
Thanks for writing the article Dr Meier.
“the long-term thinning trend is primarily a reflection of additional energy from globally warming temperatures”
What is really being meant here when you say “globally warming temperatures”, temperature of what exactly? You go onto say the local air temperature is not increasing, so is it:
a) additional radiant energy from GHG (CO2) in the atmosphere above the ice.
b) additional energy due to warmer sea temperatures in the Arctic.
c) additional energy due to warmer temperatures elsewhere (but how is this energy transported to the Arctic?)
d) something else?
Maybe I’m just missing something obvious.
This post 3 years ago shows data published by Levitus et al on an exact corellation between Barents sea temperatures and the AMO.
http://wattsupwiththat.com/2009/10/08/new-paper-barents-sea-temperature-correlated-to-the-amo-as-much-as-4%C2%B0c/
If – as Dr Meier states, Arctic ice decline is not from cooling air temperatures (current DMI arctic temps are normal or just below) then maybe the decline of recent decades is simply the upswing of the AMO pushing more north Atlantic drift water up to the Arctic. It looks due for an overturn to a downward slope.
Kip Hansen says:
June 21, 2012 at 1:13 pm
My experience with ice melting is rivers and streams in upstate NY…it generally doesn’t actually melt from above, usually the water below the ice increases (because it is not freezing when it falls, and the comparatively warm earth melts snow from below), warms up (it is almost always already above freezing), the slowly eats the ice from below. Only when Spring temps reach the high forties (F) does are temp begin to affect the thick winter ice.
======
After thinking about it, It crossed my mind that the proposed melting mechanism doesn’t match what we see here in the Champlain Valley either. Many years, we enter March with anywhere from 50cm to a meter of snow on the ground and the lakes frozen. As the sun gets higher in the sky and temperatures rise, we start to lose frozen moisture directly to the air (sublimation). As the days get longer and warmer, there is direct melting. But what doesn’t happen is a noticable tendency for daytime temperatures to hold at freezing (because energy is sucked from the air to melt snow and ice). Of course things might be different in the Arctic where sun angles are much lower than they are at 44 degrees North.
@ur momisugly Jimbo
Nevermind the fact that the snow cover is dissapearing earlier and the Tundra is losing its permafrost. TheTundra is catching fire because as its exposed to sun radiation for longer (that snow cover melts earlier now, remember?) its vegetation of mainly dwarf shrubs, sedges and grasses, mosses, and lichensis is dry like tinder.
http://www.arctic.noaa.gov/report10/ocean.html
Notice this is from 2010 and that it talks of historical record warm water. I’d expect that if the ice melts this year as much as in 2007, the same will occur.
All the best.
has the ocean temperature warmed in the arctic since 1979?
I dunno ask Tisdale
http://bobtisdale.files.wordpress.com/2012/03/12-arc.png
Does anyone here think that warmer water will create more ice?
A nice depiction of air temperatures from a typical buoy. As it warms the air temperature is thresholded at around 0C.. guess why
http://imb.crrel.usace.army.mil/2011J.htm
http://imb.crrel.usace.army.mil/2011M.htm
http://imb.crrel.usace.army.mil/2012E.htm
More stuff here
http://nsidc.org/noaa/moored_uls/index.html
http://nsidc.org/cgi-bin/get_metadata.pl?id=g02141
Rigor and Wallace 2004 noted a dramatic decline in old sea ice which occurred at the end of the 80s as a result of a paradigm shift in the Beaufort Gyre and the TransPolar Drift. During the 80s the BG was large and covered most of the Arctic between the Pole and the Bering Strait. Old ice circulated within it for up to a decade or more. The TPD ran mostly parallel to the Lomonosov Ridge. At the end of the 80s the BG contracted and shifted toward North America and the TPD moved to track that followed almost exactly along the prime meridian and tangential to the tighter BG circulation. The old ice that through the 80s had remained west of the pole at minimum and was 80% of the total was broken up and flushed out through the Fram and in only a couple seasons old ice was down too 30% and max age had changed from 10+ yrs to 3-4 yrs.
That circulation pattern remained mostly the same for nearly two decades
http://iabp.apl.washington.edu/pdfs/RigorWallace2004.pdf
At this point it’s probably doubtful that much ice is enduring for even 4 years. In the last couple years the BG seems to be quite variable and although it might return to its 80s pattern the Arctic drift maps don’t seem to indicate a consistent pattern yet, at least to my eye. Unless or until the BG does go back to what it was in the 80s, we will probably continue to see variations around the new baseline that was established in the early 90s and, although temps will undoubtedly have some effect, the end results will continue to be dominated by circulation patterns driven by winds, currents, and pressure fronts which, from what I’ve seen, are mostly unrelated to anything anthropogenic.