It's half time in the Arctic, and with the Summer Solstice and the first half of the melting season behind us, it's time for an assessment. It's clear that PAC-2013, this year's persistent Arctic cyclone, kept the Arctic in a cold and cloudy grip, causing a very slow start to the melting season. In this post I want to explore how the conditions in the past two months of May and June compare to those of previous years, and get a feel for just how slow the start to this melting season has been.
When reading Chris Reynolds' second entry in the June Status series on his Dosbat blog, describing atmospheric patterns for the previous month in the 2004-2013 period, I remembered doing a similar thing last year in a blog post called Arctic atmosphere June-July 2012. In that post I divided the period from June 1st to July 15th up in three parts by using maps from the daily mean composites page from NOAA's Earth Science Research Laboratory.
This time I've combined maps showing sea level pressure patterns from 2007, 2010, 2011, 2012 and this year, divided into four periods spanning two weeks in May and June. I've chosen these years because they have been the biggest contenders for the records and have the most interesting features that help us understand the period after the shift in 2007. And of course, I already had the images that I used for last year's post.
Below the maps I have added Cryosphere Today sea ice area graphs (CT SIA) which will allow me to discuss certain correlations. The image is pretty big, so I hope it fits on everyone's screen when they want to have a closer look by clicking on it:
Before discussing the two months one by one - and bear with me, because it's going to get interesting -, here's a quick explanation on how to interpret these sea level pressure (SLP) maps:
The green, yellow and orange stand for high pressure, the blue and purple for low pressure. In high-pressure areas the winds blow in a clockwise fashion. In low-pressure areas it's the other way around: winds blow in an anti-clockwise fashion. The direction and intensity of winds are usually represented by isobars (dividing lines between different pressures), but in this case the different colours make it clear enough.
The important thing to remember is that the ice follows the wind, with a bit of a lag and not as fast. With clockwise blowing winds the ice gets pushed together, under anti-clockwise blowing winds ice floes get dispersed (see image on the right). This has to do with the Coriolis effect caused by the Earth's rotation.
Another important difference between high-pressure and low-pressure areas is that highs make for clear skies, and thus lots of insolation during summer (or radiation out to space during the dark winter months). Lows cause cloudiness, and thus less sunshine reaches the ice pack during summer, making the air generally colder. During winter clouds can block outgoing radiation and thus increase air temperatures, though still below freezing. Depending on their position, highs and lows can pull in warm or cold air from the continents.
What makes area/extent decrease fast in general, is when highs dominate the American side and centre of the Arctic (the right half of the maps) and lows dominate the Siberian side of the Arctic (left half of the maps), because together they act like cogwheels - one turning clockwise, the other anti-clockwise - that push all the ice floes together and towards the Atlantic where they melt out. This phenomenon is known as the Arctic dipole anomaly.
Right, and now onto the analysis:
2011 (purple trend line) starts out low at the beginning of May and stays low because of lots of high pressure all around the Arctic. It is soon joined by 2010 (green trend line) that has a very quick start to the melting season due to a big high over the Chuckchi and Beaufort regions that weakens slightly after the first half of the month. You can clearly see the slight lag on the CT SIA map, because as I said it takes a while for the ice pack to respond when weather patterns stay put
for a while, especially early in the season.
2007 and 2012 (blue and orange) aren't far from 2010 and 2011 at the end of May, also having enjoyed some high pressure during the first half of the month, turning into a Arctic Dipole (AD) during the second half of the month. The reason that 2012 drops slightly faster than 2007 most probably has to do with the fact that the ice in 2012 was a lot thinner on average after the 2010 and 2011 melting seasons.
2013 (red trend line) clearly stands out on the May CT SIA graph. It starts out alright, but the ice decrease soon stalls under the influence of the persistent cyclone that will dominate the Arctic for a couple of weeks, bringing in clouds and cold, but at the same time dispersing the floes in the interior of the ice pack.
2013 remains at a fair distance from the other years because of PAC-2013. A large low-pressure zone even envelops Greenland, which make for antithetical conditions compared to jaw-dropping 2012. This first half of the melting season is clearly very different from anything we've seen in recent years.
During all of June 2007 (orange) and 2011 (purple) keep dropping steadily due to some pretty stable conditions with highs continuing to dominate much of the area over the Arctic Ocean.
2010 has a bit of a lull during the first half of the month, but drops fast again when a relatively big high-pressure system takes over the Central Arctic. 2012 produces an even bigger high during the first half of the month, causing the trend line to plummet on the CT SIA graph.
And that brings us to the most interesting period of last year, a period that in some ways was more interesting to me than the period with the Great Arctic Cyclone that would cause havoc with a capital H a month later. In the second half of June the Arctic became dominated by a widespread low, and although the sea ice area decrease slowed down, it didn't stall completely like it did in previous years under similar circumstances (see last year's blog post on atmospheric conditions for further explanation).
To me this was a sign that the ice was so weak in large parts of the ice pack that it didn't really matter what the weather was, the ice melted out anyway. Especially, of course, after a warm winter in the Kara and Barentsz Seas where the ice melted out in record time during a very decent May and first half of June, combined with lots of insolation over the Beaufort Sea. Even though the lows persisted during July (image on the right), 2012 never gave away the lead.
We could be witnessing the same thing this year, even though 2013's start of the melting season doesn't come close to those of previous years, let alone 2012. We have to keep in mind though that this year did start out with a total volume that was as low as 2011 and 2012, the highest amount of first-year ice on record (due to last year's record minimum), and with PAC-2013 that has impacted the interior of the ice pack. We still haven't seen the full effects of that event.
There are basically three set-ups that determine the rate of decrease:
- Highs dominating the Arctic (negative Arctic Oscillation) -> fast decrease, especially if the highs are on the American side of the Arctic and stay put for a while, causing a lot of compaction.
- Lows dominating the Arctic (positive Arctic Osciallation) -> average decrease, faster when the pressure is very low. Lows on the Siberian side of the Arctic, combined into an Arctic Dipole with highs on the other side, cause the fastest decrease. 2007 was so special because the AD was persistent and relatively stable for almost all of the melting season. Were that to happen again in coming years, the Arctic could very well go ice-free before the end of the melting season.
- Neither fish nor flesh, neutral Arctic Oscillation, unstable weather systems that don't stay into place -> slow decrease.
What has changed compared to the past, is that the gaps between these three set-ups have become smaller, and this obviously has to do with the thinning of the Arctic sea ice pack. A less intense high can do as much damage as an exceptional high during the 80's and 90's. A large cyclone tears the ice pack apart, whereas the ice pack was much better able to withstand it in the past. Thinner ice is now dampening weather dominance.
And so we are entering that period of the new melting season, where area and extent keep decreasing steadily, even if the weather isn't all that conducive to transport, melting and compaction. With some favourable weather conditions in coming weeks 2013 might even join the top years, or possibly even have a shot at 2012's title (more on that in the next ASI update). It would be a remarkable feat after the unusually slow start to the melting season.