I thought it was time to look at the sea ice data as the summer Arctic melt proceeds.
The image below shows the mean sea ice extent 1989-93 on 22 July, the sea ice extent 22 July 2017, and the difference between the two data sets. Reds imply less sea ice than the mean 1989-93, and blues an increased sea ice extent.
As we would expect, compared with the 1989-93 data the sea ice edge is consistently further north. Things that stand out for me are the virtually open water in the Barents and Kara Seas. This region was very slow to freeze over in the autumn and winter of 2016, so I would have expected the sea ice there to be relatively thin at the end of the Arctic winter.
Overall the area of the Arctic Ocean covered with sea ice is low. Some will note it is currently above the record low in 2012, but it's only 90,000 km2 above that. To me that is not that significant.
What is significant is the sea ice 22 July 2017 is ~1.7 million km2 below the median extent from 1981-2010.
I looked a couple of days ago at the sea ice in the North West Passage (19 July 2017), and it is starting to open up.
The yellow line traces out possible ship routes through the North West Passage and whilst there is still ~1200 km of sea ice on that route, when you compare the region to the longer term data you can see how low this is compared the historical record.
It may be this year that the passage may not open at all, but taken together the two plots are a a good example of how we can expect the north west passage to become consistently open as the Arctic continues to warm.
This is the Antarctic Peninsula and the outline of A68 from the satellite image on 14 July 2017 shown in black. The ice front is from the Bedmap2 data set (so a little out of date), and the bathymetry from the IBCSO data set.
In the map below, the shaded colour is the distance of any point on the sea bed to the closest actual depth measurement.
So the dark blue stripes labelled in the Weddell Sea are actually ship tracks - and the dark colours are good depth data. These measurements will have been made by icebreaker.
Just in front of A68 there is a very large area where no ship has been within ~80 km.
One small note on the size. I digitized the iceberg from a satellite image (a KML File can be downloaded). On twitter today there were satellite images showing fractures already.
But Martin O'Leary of the MIDAS team posted today on twitter that to the untrained eye looks like iceberg, is very likely fast ice (so thick sea ice that is "fast" to A68 - but only a few metres thick.)
There is not a consistent trend in Antarctic sea ice extent, and much regional interannual variability. The plot below shows the sea ice extent on 13 May for each of the years 1989-95, and 13 May 2017.
The image above shows the sort of variability we expect in the Antarctic sea ice extent. It is helpful too to see where the sea is currently is and isn't compared with the mean from 1989-93.
The regions in May 2017 with the greatest deficit of sea ice remain the Amundsen and Ross Sea, and the Eastern Weddell Sea and off the coast of Dronning Maud land. As I said in my last Antarctic sea ice post it is likely the freeze up is delayed because of the heat gained by the ocean in the Antarctic summer of 2016/17.
You can also see in the South West Weddell Sea the Ronne Polynya I wrote about in March 2017 is still seen in the sea ice concentration data. In the visible satellite data you can also see this open water.
In the Sentinel 1 SAR data from the 15 May (From PolarView), the growth of the sea ice in the polynya is clear.
This ice growth is important for the ocean as it means the salinity of the waters just beneath the sea ice will be increasing.
I'll keep watching the polynya to see if and when it closes up. And I will also keep looking at the sea ice.
Approaching the middle of May and well into the Arctic sea ice retreat we can see that the sea ice extent (area of ocean with at least 15% sea ice) is still well below the mean over the satellite record.
I like a geographic perspective, so this is the mean sea ice extent 1989-93 on 13 May, the sea ice extent 13 May 2017, and the difference between the two data sets. Reds imply less sea ice than the mean 1989-93, and blues an increased sea ice extent.
There appears to be a general trend of the Arctic sea ice edge retreating between the two data sets, but I think this is in places meteorological - that is the winds are compressing the sea ice. I think this because there is a lot of blue (i.e. more sea ice than the 89-93 mean) just north of the sea ice edge.
The Bering Sea appears relatively sea ice free at this time.
Sea ice is still relatively low in both the the Arctic spring and Antarctic autumn. A geographical perspective always helps so here is the status of the sea ice concentration 23 April 2017 for both polar regions.
Here is the sea ice concentration 23 April 2017 compared with the 1989-1993 mean on the 23 April. Red shades = less sea ice than the 1989-93 mean on 23 April, and Blue shades = more sea ice than the 1989-93 mean on 23 April.
The stand out regions for me are once more (as in my post in January), the Northern Barents Sea is relatively low, along with the Bering Sea and the Sea of Okhotsk. There is a consistent retreat of the ice edge almost everywhere, and comparatively a lot of open water in Hudson Bay.
The polynya I saw forming in early February is still clear, and very large in the Southern Weddell Sea. At the moment it is more than than 80,000 km2, although there is clearly a lot of young sea ice covering a large part of the polynya.
In my original post I said this was likely formed by winds from the Ronne Ice Shelf.
Well Dr Stef Lhermitte (Delft) has put together the most amazing movie showing the development of the polynya over January and February. It shows satellite sea ice data with winds from the ECMWF overlain.
You can clearly see the winds pushing the sea ice away from the ice shelf as time progresses.
Trying to understand the geographic nature of the very low Antarctic sea ice extent I made the following animation:
The highlight issues in the graphic are the clear lack of sea ice in the Amundsen/Ross Seas and the Southern Ocean off Dronning Maud land. (If you are not familiar with the names off the seas / locations see the map below).
The Weddell Sea is has a relatively compressed sea ice cover this year - but it's clear there is large inter-annual variability.
Off Wilkes Land the sea ice is heavier this year - and it's easy for this to get lost in the headline story around the very low extent. There are several Antarctic research stations along this coast:
I noticed yesterday that a polynya had formed in front of the Ronne Ice Shelf over the last 2 weeks.
In that image it is about 27,000 km2 in area.
I mapped the opening of the polynya from MODIS imagery over the last two weeks. There is cloud in the images but the opening of the polynya is fairly clear.
On 31 January 2017 there is no open water, but then over the 16 day period it opens to the ~27,000 km2 in area. If you're eagle eyed you can see that there is thin frazil ice forming in the open water in front of the ice shelf at the end of the sequence.