Facts About the Arctic in May 2023

At the beginning of May AWI’s high resolution AMSR2 extent metric is at the top of the historical range:

After a period of melt in the East Greenland Sea, export of sea ice from the Central Arctic via the Fram Strait has increased recently:

Sea ice area in the Barents Sea peaked in the first week of April and has declined steadily since then:

SMOS data is flowing again, and the gaps have been filled. Here’s the final pre melt ponding map of the spring:

It reveals large areas of thin ice that have recently developed in the Kara Sea. Here’s a pseudo colour glimpse of the region through the clouds:

“False colour” image of the Kara Sea on May 1st from the MODIS instrument on the Terra satellite

and the regional sea ice area graph:

An area of thin ice has also developed off the Chukchi Sea coast of Alaska.

[Edit – May 4th]

The PIOMAS gridded thickness data for April show that the 2023 maximum volume of 23,320 km3 was reached on April 23rd:

Here too is the thickness map for April 30th:

In Arctic weather news there is currently a cyclone spinning over the Laptev Sea which is forecast by GFS to reach a minimum MSLP of 971 hPa later today:

It remains to be seen what effect the inclement weather has on the ice in the region.

[Edit – May 5th]

As Neil points out below, the May edition of the NSIDC’s Arctic Sea Ice News has been published:

Air temperatures at the 925 hPa level in April were near average to below average over most of the Arctic Ocean. This helps to explain the slow rate of sea ice loss during April. While temperatures were modestly above average over the Barents Sea, these areas are already largely free of sea ice:

The sea level pressure pattern was characterized by fairly high pressure over most of the Arctic Ocean. The clockwise winds around a separate high-pressure center over Scandinavia brought warm winds from the south over the Barents Sea, consistent with the above air average temperatures in that area. Similarly, below-average temperatures over Alaska were driven by the combination of a strong low pressure area in the Gulf of Alaska and high pressure in the Beaufort Sea, driving air generally southward from the ice-covered Arctic Ocean.

The NSIDC article also mentions a new open access paper by Sledd et al. entitled “Clouds Increasingly Influence Arctic Sea Surface Temperatures as CO2 Rises“. Here’s an extract from the abstract:

Under pre-industrial CO2 concentrations, summer clouds have little direct effect on maximum annual sea surface temperatures (SST). When CO2 concentrations increase, sea ice retreats earlier, allowing more solar radiation to warm the ocean. Clouds can counteract this summer warming by reflecting solar radiation back to space. Consequently, clouds explain up to 13% more variability in maximum annual SST under modern-day CO2 concentrations. Maximum annual SST are three times more sensitive to summer clouds when CO2 concentrations are four times pre-industrial levels.

[Edit – May 7th]

Since Neil is wondering about the recent “slow melt”, here’s an animation of Arctic sea ice drift over the last month. Click the image if your bandwidth supports viewing a 13 Mb video:

Click to animate (13 Mb!)

[Edit – May 11th]

Melt ponds (and a patch of open water) are now clearly visible along the coast of the Coronation Gulf in the Canadian Arctic Archipelago:

“False colour” image of the Coronation Gulf on May 11th from the MODIS instrument on the Terra satellite

The Swedish icebreaker Oden is now in the marginal ice zone west of Svalbard. It won’t come as a surprise to regular readers that the Art of Melt 2023 expedition blog reports that:

We encountered the first ice on the first day, which quickly became quite challenging to break, even for Oden, as there were many ridges and bumps on the ice.

[Edit – May 13th]

Above zero temperatures are forecast for the Beaufort Sea by the end of the weekend:

The weather is already warm in the Canadian Arctic Archipelago. Here’s the report from Kugluktuk for the last 24 hours:

Melt ponds are now visible in Franklin and Darnley Bays:

“False colour” image of the Amundsen Gulf on May 12th from the MODIS instrument on the Aqua satellite

[Edit – May 15th]

Melt ponds are now visible in the Beaufort Sea off the Mackenzie Delta:

“False colour” image of the Mackenzie Delta on May 15th from the MODIS instrument on the Terra satellite

[Edit – May 17th]

The flow of the Mackenzie River at Tsiigehtchic (Arctic Red River) has suddenly gone off the charts:

The raw data does have a big gap in it, so there are doubts about the veracity of the recent numbers:

However continuous data from the middle channel in the Mackenzie Delta shows a similar pattern:

There are also reports of flooding at Fort McPherson on the nearby Peel River:

The Hamlet of Fort McPherson, N.W.T., declared a state of emergency over flooding on Tuesday evening. 

Elders, infants and people who are sick are being evacuated out of Fort McPherson, N.W.T., as early as Thursday morning, even though flooding in the community has started to drop, according to the community’s chief. 

“They won’t be back until the state of emergency is lifted,” said Tetlit Gwich’in First Nation Chief Elizabeth Wright. 

The hamlet of about 700 people on the banks of the Peel River declared a state of emergency on Tuesday evening, after water flooded cabins at a popular fishing site south of the community and threatened its fresh water supply.

Photo: Dean Charlie

[Edit – May 19th]

Melt ponds are now visible on the southern shore of the Kara Sea:

“False colour” image of the Kara Sea on May 18th from the MODIS instrument on the Terra satellite

Bottom melt is now apparent in the western Beaufort Sea at 74.56 N, 155.4 W, the location of ice mass balance buoy 651330:

Assuming that no further thickening occurs the ice reached a maximum thickness of 1.35 m.

[Edit – May 20th]

The clouds have cleared over the Mackenzie Delta, revealing the river’s dark waters flooding the fast ice:

“False colour” image of the Mackenzie Delta on May 19th from the MODIS instrument on the Terra satellite

The confluence of the Arctic Red River with the Mackenzie is visible in the bottom left of the image, with the Peel River and Fort Macpherson above that.

[Edit – May 22nd]

The mid month PIOMAS gridded thickness data has been released. Here’s the Arctic sea ice thickness map for May 15th:

and here’s the calculated volume graph:

The clouds cleared over Chaunskaya Bay yesterday. Evidently surface melt has begun on the East Siberian land fast sea ice:

“False colour” image of Chaunskaya Bay on May 21st from the MODIS instrument on the Aqua satellite

Here’s the recent weather record from Pevek:

[Edit – May 25th]

There was a fine view of Chaunskaya Bay from on high this morning:

“False colour” image of Chaunskaya Bay on May 25th from the MODIS instrument on the Terra satellite

The melt ponds on the fast ice are now clearly visible.

[Edit – May 29th]

Water is now starting to flow through the Lena Delta, and surface melt has begun on the fast ice in the Laptev Sea further south:

“False colour” image of the Lena Delta on May 29th from the MODIS instrument on the Terra satellite

Barents Sea ice area has experienced a steep decline over the last few days:

Further west on the Atlantic side of the Arctic Ocean, the Greenland Sea ice area is finally back within its previous historical range:

Watch this space!

27 thoughts on “Facts About the Arctic in May 2023

  1. NSIDC May 3rd.
    The rate of sea ice loss for April 2023 was slow, owing to cool conditions across the ice-covered Arctic Ocean and below-average to near-average temperatures near the ice edge.

  2. Jim, you left out part of the NSIDC comment on the Sledd paper:
    “Looking to the future, their findings suggest that when the Arctic becomes seasonally ice free, the maximum sea surface temperature becomes three times more sensitive to clouds than in the pre-industrial era.”

    Their modeling analysis was conducted assuming four times pre-industrial CO2 levels. That’s an RCP8.5 scenario (at least) with over 1,000 ppm, and has essentially zero probability. Sorry, more junk science to scare the kiddies.

    1. Whereas you seem to have missed the part where the abstract says:

      When CO2 concentrations increase, sea ice retreats earlier, allowing more solar radiation to warm the ocean. Clouds can counteract this summer warming by reflecting solar radiation back to space. Consequently, clouds explain up to 13% more variability in maximum annual SST under modern-day CO2 concentrations.

      1. Noted, but irrelevant to my point about speculation on what happens ‘when the Arctic is seasonally ice-free’.

  3. November month end sea ice extent fell BELOW 2010’s average for the first time since March 2022.
    It remained BELOW 2010’s average until April 2023

    Jan.31.—-16th.lowest. +333,064 sq.km. above 2010’s average.
    Feb.28.—-11th.lowest +116,764 sq.km. above 2010’s average
    Mar.31.—-10th.lowest -127,470 sq.km. BELOW 2010’s average
    Apr. 30.—-14th.lowest +16,886 sq.km. above 2010’s average
    May. 31.—15th lowest +395,097 sq.km. above 2010’s average
    Jun. 30.—–10th.lowest +35,744 sq.km above 2010’s average
    Jul. 31.—–12th lowest +228,905 sq.km. above 2010’s average
    Aug.31.—-11th.lowest +203,902 sq.km. above 2010’s average
    Sep.30.—-–7th lowest +115,918 sq.km. above 2010’s average
    Oct.31.– 11th lowest, +269,119 sq.km. above 2010’s average
    Nov30.——5th lowest, -183,803 sq.km. BELOW 2010’s average
    Dec.31—–4th.lowest,-138,700 sq.km. BELOW 2010’s average
    Jan.31—-2nd.lowest, -302,000 sq.km. BELOW 2010’s average
    Feb.28——9th.lowest, -36,000 sq.km. BELOW 2010’s average
    Mar.31—–7th.lowest, -274,000 sq.km. BELOW 2010’s average
    Apr.30—–11th.lowest, +50,500 sq.km. above 2010’s average
    May 6—-17th.lowest, +298,000 sq.km. above 2010’s average

    Month end April sea ice has just risen ABOVE 2010’s average.
    Can anyone explain the apparent rise in sea ice extent for May.
    As you may know I think it is related to the AMO.
    Does anyone know why the AMO data has not been updated since jan.2023

    1. Just looked at the University of Bremen map, for this date in 2019, this year has more ice in the Bering sea between Svalbard and FJ land, wind driven export we saw last year. The Greenland sea has a lot more ice, again ice leaving the central Arctic for warmer waters.
      Also at this time in 2019 there was no ice beyond the Bering strait, was the north Pacific warmer the preceding winter?

  4. Here’s a “false colour” satellite image of north-west Canada revealing the smoke from numerous wildfires on the left and ice free stretches of the Mackenzie River at the top right:

    According to CBC:

    The Alberta government has declared a provincial state of emergency as out-of-control wildfires force more residents from their homes.

    Premier Danielle Smith made the announcement at a media briefing Saturday afternoon…

    Nearly 25,000 Albertans have been forced from their homes as wildfires continue to rage in north and central Alberta, while 5,200 residents are on evacuation alert.

    The number of active fires increased to 110 from 103, as hot and dry conditions led to fast-spreading wildfires that threatened homes, businesses and lives.

  5. Noted with some interest that PIOMAS has reported on April. They again had the opportunity to cut and paste their observation about ‘no trend in 12 years’. It’s notable because April is the peak volume month, and April volume is now nearly 2 SD above the long-term trend. Per DMI, temps above 80 degrees have been at or below the long- term average for most of the Spring. Surely something is changing up there.

    1. The PIOMAS team have also cut’n’pasted this observation once again:

      The energy required to melt the 16,400 Km³ of ice that are lost every year (1979-2010 average) from April to September as part of the natural annual cycle is about 5 x 10²¹ Joules. For comparison, the U.S. Energy consumption for 2009 (www.eia.gov/totalenergy) was about 1 x 10²⁰ J. So it takes about the 50 times the annual U.S. energy consumption to melt this much ice every year. This energy comes from the change in the distribution of solar radiation as the earth rotates around the sun.

      To melt the additional 280 km³ of sea ice, the amount we have have been losing on an annual basis based on PIOMAS calculations, it takes roughly 8.6 x 10¹⁹ J or 86% of U.S. energy consumption.

      However, when spread over the area covered by Arctic sea ice, the additional energy required to melt this much sea ice is actually quite small. It corresponds to about 0.4 Wm⁻². That’s like leaving a very small and dim flashlight bulb continuously burning on every square meter of ice. Tracking down such a small difference in energy is very difficult, and underscores why we need to look at longer time series and consider the uncertainties in our measurements and calculations.

      1. Jim, thanks, and yes, I am a faithful reader of all their commentary. I do appreciate scientists who point out how critical solar energy is to climate variability. Some climate scientists struggle with that concept.

        Additionally, and tellingly, since volume has been stable for the last twelve years, does that imply (using their own numbers) that there has not been an additional 8.6 x 10^19 joules of energy/year input to the Arctic climate system that previously caused the ice loss scenario? Did someone turn off all those tiny bulbs? Or are we in a period of lower solar output? Any papers on that question?

        1. At the highest latitudes, the sun tends to be low in the sky, except for a brief few weeks either side of June 20.
          Some over on ASIF have speculated wether that brief Arctic summer is long enough to produce a so called blue ocean event, but it might also tell us why it’s more difficult for extent to fall below a certain level, hence the drop in extent stalling around 4 million km2 or so.
          Of course the deciding factor could be the gradual rise in ocean temperatures.

          1. Thanks, Jim, but your graph of solar insolation at latitude is a geometric model that assumes constant solar output. I was referring to the variations in solar output from cycle to cycle, which would make those crisp lines a bit ‘smudgier’ ( if that’s a word) in any given year/solar cycle. I’m sure you’ve read Longhurst’s book (second edition now out). Chapter 8 cites some interesting data wrt the Arctic.

          2. Your certainty is misplaced Taylor!

            I hadn’t read “Doubt and Certainty in Climate Science” before, but I’ve now skimmed Chapter 8 of the first edition. Thanks for the tip. Which part(s) do you think are relevant to the current discussion?

          3. For example Lean (2000), shows data that would indicate that solar insolation has varied nearly 2 Wm^2 cyclically from 1600 to the present.

          4. Have you checked the scale on my TOA flux delta map above?

            Solar insolation that’s reflected instead of absorbed at the surface varies by a lot more than that from one summer to the next.

      1. Interesting anomaly map using a 1978-2000 baseline, one of the coldest periods in the last 70 years. Matt’s graph, using the 1958-2002 period (from DMI) shows temp for 80+ North below normal all spring. I don’t suppose the climate community could agree on a single baseline so we could compare anomalies, do you?

  6. As I just pointed out to Matt, the current low pressure system has sucked some warm air into the Pacific side of the Arctic Basin.

    GFS seems to have settled on a minimum MSLP of 978 hPa:

    Plus a sprinkling of rain in the vague vicinity of IMB buoy 651330.

  7. Hi Jim,
    Sorry to trouble you again about the image displayed on my posts.
    I noticed you recently changed yours.
    I am getting a little older these days
    Thought my image should reflect this.
    How do I make a change?

      1. Thanks very much Jim.
        Found the site. Put in into my favourites.
        Now looking for a more up to date photo showing my advancing age.
        I will still be holding a glass of wine

Leave a Reply

Your email address will not be published. Required fields are marked *

The maximum upload file size: 8 MB. You can upload: image. Links to YouTube, Facebook, Twitter and other services inserted in the comment text will be automatically embedded. Drop files here

This site uses Akismet to reduce spam. Learn how your comment data is processed.