MASIE: “high-resolution, accurate charts of ice conditions”
Walt Meier, NSIDC, October 2015 article in Annals of Glaciology.

Update February 4, 2017 Below

The home page for MASIE (here) invites visitors to show their interest in the dataset and analysis tools since continued funding is not assured. The page says:
NSIDC has received support to develop MASIE but not to maintain MASIE. We are actively seeking support to maintain the Web site and products over the long term. If you find MASIE helpful, please let us know with a quick message to NSIDC User Services.

For the reasons below, I hope people will go there and express their support.

1. MASIE is Rigorous.

Note on Sea Ice Resolution:

Sea Ice Index (SII) from NOAA is based on 25 km cells and 15% ice coverage. That means if a grid cell 25X25, or 625 km2 is estimated to have at least 15% ice, then 625 km2 is added to the total extent. In the mapping details, grid cells vary between 382 to 664 km2 with latitudes.  And the satellites’ Field of View (FOV) is actually an ellipsoid ranging from 486 to 3330 km2 depending on the channel and frequency.  More info is here.

MASIE is based on 4 km cells and 40% ice coverage. Thus, for MASIE estimates, if a grid cell is deemed to have at least 40% ice, then 16 km2 is added to the total extent.

The significantly higher resolution in MASIE means that any error in detecting ice cover at the threshold level affects only 16 km2 in the MASIE total, compared to at least 600 km2 variation in SII.  A few dozen SII cells falling below the 15% threshold is reported as a sizable loss of ice in the Arctic.

2. MASIE is Reliable.

MASIE is an operational ice product developed from multiple sources to provide the most accurate possible description of Arctic ice for the sake of ships operating in the region.

Operational analyses combine a variety of remote-sensing inputs and other sources via manual integration to create high-resolution, accurate charts of ice conditions in support of navigation and operational forecast models. One such product is the daily Multisensor Analyzed Sea Ice Extent (MASIE). The higher spatial resolution along with multiple input data and manual analysis potentially provide more precise mapping of the ice edge than passive microwave estimates.  From Meier et al., link below.

Some people have latched onto a line from the NSIDC background page:
Use the Sea Ice Index when comparing trends in sea ice over time or when consistency is important. Even then, the monthly, not the daily, Sea Ice Index views should be used to look at trends in sea ice. The Sea Ice Index documentation explains how linear regression is used to say something about trends in ice extent, and what the limitations of that method are. Use MASIE when you want the most accurate view possible of Arctic-wide ice on a given day or through the week.

That statement was not updated to reflect recent developments:
“In June 2014, we decided to make the MASIE product available back to 2006. This was done in response to user requests, and because the IMS product output, upon which MASIE is based, appeared to be reasonably consistent.”

The fact that MASIE employs human judgment is discomforting to climatologists as a potential source of error, so Meier and others prefer that the analysis be done by computer algorithms. Yet, as we shall see, the computer programs are themselves human inventions and when applied uncritically by machines produce errors of their own.

3. MASIE serves as Calibration for satellite products.

The NSIDC Background cites as support a study by Partington et al (2003).  Reading that study, one finds that the authors preferred the MASIE data and said this:

“Passive microwave sensors from the U.S. Defense Meteorological Satellite Program have long provided a key source of information on Arctic-wide sea ice conditions, but suffer from some known deficiencies, notably a tendency to underestimate ice concentrations in summer. With the recent release of digital and quality controlled ice charts extending back to 1972 from the U.S. National Ice Center (NIC), there is now an alternative record of late twentieth century Northern Hemisphere sea ice conditions to compare with the valuable, but imperfect, passive microwave sea ice record.”

“This analysis has been based on ice chart data rather than the more commonly analyzed passive microwave derived ice concentrations. Differences between the NIC ice chart sea ice record and the passive microwave sea ice record are highly significant despite the fact that the NIC charts are semi-dependent on the passive microwave data, and it is worth noting these differences. . .In summer, the difference between the two sources of data rises to a maximum of 23% peaking in early August, equivalent to ice coverage the size of Greenland.” (my bold)  For clarity: the ice chart data show higher extents than passive microwave data.

The differences are even greater for Canadian regions.

“More than 1380 regional Canadian weekly sea-ice charts for four Canadian regions and 839 hemispheric U.S. weekly sea-ice charts from 1979 to 1996 are compared with passive microwave sea-ice concentration estimates using the National Aeronautics and Space Administration (NASA) Team algorithm. Compared with the Canadian regional ice charts, the NASA Team algorithm underestimates the total ice-covered area by 20.4% to 33.5% during ice melt in the summer and by 7.6% to 43.5% during ice growth in the late fall.”

From: The Use of Operational Ice Charts for Evaluating Passive Microwave Ice Concentration Data, Agnew and Howell  http://www.tandfonline.com/doi/pdf/10.3137/ao.410405

More recently Walter Meier, who is in charge of SII, and several colleagues compared SII and MASIE and published their findings October 2015 (here).  The purpose of the analysis was stated thus:
Our comparison is not meant to be an extensive validation of either product, but to illustrate as guidance for future use how the two products behave in different regimes.

The abstract concludes:
Comparisons indicate that MASIE shows higher Arctic-wide extent values throughout most of the year, largely because of the limitations of passive microwave sensors in some conditions (e.g. surface melt). However, during some parts of the year, MASIE tends to indicate less ice than estimated by passive microwave sensors. These comparisons yield a better understanding of operational and research sea-ice data products; this in turn has important implications for their use in climate and weather models.

A more extensive comparison of MASIE from NIC and SII from NOAA is here.

4. MASIE continues a long history of Arctic Ice Charts.

Naval authorities have for centuries prepared ice charts for the safety of ships operating in the Arctic.  There are Russian, Danish, Norwegian, and Canadian charts, in addition to MASIE, the US version.  These estimates rely on multiple sources of data, including the NASA reports.  Charts are made with no climate ax to grind, only to get accurate locations and extents of Arctic ice each day.

Since these long-term records show a quasi-60 year cycle in ice extents, it is vital to have a modern dataset based on the same methodology, albeit with sophisticated modern tools.

Summary

Measuring anything in the Arctic is difficult, and especially sea ice that is constantly moving around.  It is a good thing to have independent measures using different methodologies, since any estimate is prone to error.

Please take the time to express your appreciation for NIC’s contribution and your support for their products at MASIE  home page.

Update February 4, 2017

In the comments Neven said MASIE was unusable because it was biased low before 2010 and high afterward.  I have looked into that and he is mistaken.  Below is the pattern that is observed most months.  March is the annual maximum and coming up soon.

As the graph shows, the two datasets were aligned through 2010, and then SII began underestimating ice extent, resulting in a negative 11-year trend.  MASIE shows the same fluctuations, but with higher extents and a slightly positive trend for March extents.  The satellite sensors have a hard time with mixed ice/water conditions (well-documented).

More on the two datasets NOAA has been Losing Arctic Ice

Russian Convoy Takes Advantage of Reduced Arctic Ice Extent

In the first such crossing since Soviet times, the convoy had earlier delivered supplies for the world’s first floating heat and power plant to be assembled in Chukotka, Russia’s most easternmost region, after a successful journey from Arkhangelsk to Pevek lasting from 14 December to 7 January.

The ease of the sailing was seen as a sign that climate warming in the Arctic can open up shipping lanes even in midwinter.  From the Siberian Times, Blow to Northern Sea Route as voyages of two icebreakers are… broken by ice

Reports of Arctic Ice Demise Prove Premature

But, despite significant temperature rises across the northern latitudes in recent years, the vessels became quickly stuck in thick, compressed ice on their return journey.

Initially there were hopes that the icebreakers could force their way out and continue their voyage within a week, and aerial reconnaissance was deployed in a search routes from the ice clog.

They are currently trapped by sudden thick ice around Chukotka’s northernmost cape Shelagsky, some 24 nautical miles from Pevek, in some of Russia’s most exposed waters.

Ruslan Nazarov, chief of Chukotka’s emergencies service, said: ‘The ice around Cape Shelagsky is at 10 points. The ice fields are more than one metre thick. The ice compression is strong and hummocks are higher than 2.5 metres.

Embarrassing But Not Life-threatening

The situation is not critical, Nazarov said, stressing that the regional ministry of emergencies and Chukotka government keep a close eye on it. ‘All the ships have enough fuel, food and all other necessities.’

A spokesman for Rosmorport has announced the icebreakers will delay a return until probably May or early June. ‘The vessels will remain for the winter because of the very heavy severe ice conditions,’ he said.

All the vessel got out of the ice, and three of them – Captain Dranitsyn and the two cargo ships – returned to Pevek. The Admiral Makarov moved further east to continue working for Rosmorport in clearing sea routes.

Officials said the icebreakers could have gone further through the ice but there was ‘a very high risk of significant damage’ to the supply ships, and it was decided to postpone the return to Archangelsk.

Arctic Ice Picks Up the Pace Everywhere

After a slow recovery in October and December, ice extent has picked considerably in recent weeks.

The graph shows how 2017 has surged to approach 2016 and the 11-year average, while exceeding 2006 (the decadal low year) by almost 500k km2.  Sea Ice Index (SII satellite product) lags behind by more than 300k km2, as it has throughout January.

The table shows ice extents across the seas within the Arctic area.

Region	2017025	Day 025 Average	2017 -Ave.	2006025	2016-2006
(0) Northern_Hemisphere	14023232	14217519	-194287	13473190	550042
(1) Beaufort_Sea	1070445	1070111	334	1069711	734
(2) Chukchi_Sea	966006	966001	5	966006	0
(3) East_Siberian_Sea	1087137	1087131	6	1087103	35
(4) Laptev_Sea	897845	897835	10	897773	71
(5) Kara_Sea	914838	918784	-3946	897490	17348
(6) Barents_Sea	390190	547198	-157007	310539	79651
(7) Greenland_Sea	646404	589533	56871	516605	129799
(8) Baffin_Bay_Gulf_of_St._Lawrence	1328099	1275341	52758	1087283	240815
(9) Canadian_Archipelago	853214	852982	232	852715	499
(10) Hudson_Bay	1260887	1260842	45	1260779	108
(11) Central_Arctic	3138059	3210975	-72916	3109874	28185
(12) Bering_Sea	488459	689158	-200699	817517	-329058
(13) Baltic_Sea	32910	78733	-45822	53850	-20939
(14) Sea_of_Okhotsk	908522	712635	195887	470477	438045
(15) Yellow_Sea	20850	26566	-5716	11722	9128
(16) Cook_Inlet	7952	11204	-3252	8941	-989

On the Russsian side, Chukchi through Laptev are all at maximums, as evidenced in the reports from the Siberian Times.  Kara is also now average and well ahead of 2006.  Barents continues to be in deficit to average, but in surplus to 2006.  On the Canadian side, all seas are above average, with Greenland and Baffin well into surpluses.

The main difference between 2017 and the average is Bering Sea down by 200k km2.  Bering Sea is the only place with less 2017 ice than 2006.  On the other hand, Okhotsk is 200k km2 above average and 400k km2 above 2006.

Summary

Just when they thought it was safe to go back in the winter water, the ice returns.  As Siberian Times concluded:

But the saga shows that despite all the talk of climate change and warming in the Arctic, thick ice can prevent convoys from crossing the Northern Sea Route in deep winter.

 

With the end of December, Arctic ice is rebuilding in the dark up to its annual maximum before the beginning of dawn in March.  Since many of the seas are already at their maximum extents, the coming months will only add about 2M km2 to the approximately 13M km2 of ice in place.

The regrowth of Arctic ice extent was slower than usual until recently. After showing resilience in September, ending higher than 2007, ice growth lagged in October, but has been ramping up toward the averages.  The map above shows the deficit of ice is mainly in two marginal seas: Bering in the Pacific and Barents in the Atlantic.

In December, 2016 ice extent has grown by 85k km2 per day, compared to the 10-year average 66k km2 per day.  As of Dec. 31, 2016 ice extent is ~3% less than average (2006 to 2015).  The chart also shows the variability of ice extent over the years during this month. (Note: Dec. 31, 2016 result is actually day 366, but the 31 days of December are compared properly.)

2015 did have the highest ice recovery rate in the last decade, but ended up just below average.  2010 had the lowest year end extent in the last decade, matched by 2016. 2011 March extent was about average at 14.819 and higher than both 2015 and 2016.

The chart also shows 2016 Sea Ice Index (SII) from NOAA has been lagging behind by  ~300k km2, but closing the gap lately.

The table below shows this year compared to average and to 2011 for day 001.  Since several years in the dataset were missing day 365, I am making the comparison a day later.

Region	2017001	Day 001 Average	2016-Ave.	2007001	2016-2006
(0) Northern_Hemisphere	12857418	13223710	-366293	12991512	-134094
(1) Beaufort_Sea	1070445	1070111	334	1069711	734
(2) Chukchi_Sea	966006	966001	5	966006	0
(3) East_Siberian_Sea	1087137	1087131	6	1087137	0
(4) Laptev_Sea	897845	897835	10	897845	0
(5) Kara_Sea	875975	898092	-22118	928941	-52966
(6) Barents_Sea	199976	469542	-269566	340349	-140373
(7) Greenland_Sea	464142	592432	-128290	545210	-81068
(8) Baffin_Bay_Gulf_of_St._Lawrence	1153387	997051	156336	863809	289578
(9) Canadian_Archipelago	853214	852979	235	852731	483
(10) Hudson_Bay	1260887	1231119	29768	1217437	43450
(11) Central_Arctic	3085289	3216691	-131403	3214432	-129143
(12) Bering_Sea	228020	506683	-278663	512000	-283980
(13) Baltic_Sea	14808	31992	-17184	16	14793
(14) Sea_of_Okhotsk	674769	376649	298120	478831	195938
(15) Yellow_Sea	5761	12180	-6419	0	5761
(16) Cook_Inlet	5283	9040	-3758	15902	-10619

The main deficit to average is in Barents and Greenland Seas on the Atlantic side, and in Bering Sea on Pacific side.  The Canadian and Siberian sides are locked in ice, with sizable surpluses in Baffin Bay and Okhotsk Sea.

The Arctic Ice Extent Plateau Continues

As I have pointed out before, the annual average ice extent is a better indicator of climate variation, since the seasonal changes are so sizable and extents vary with weather activity.  By averaging all daily extents over the calendar year, 2016 came in at 10.389M km2 compared to 10.414M in 2007, a virtual tie.  And the trendline remains slightly positive, though again, virtually flat.

The data above comes from MASIE, the most accurate Arctic ice dataset with unparalleled resolution at 4km, compared to SII which uses 25km cells.  Alarmists are unhappy with MASIE because it shows more ice, and it has been certified as “reasonably consistent” since 2006.

Alarmists are making much ado about 2016 being lower than 2007, and hoping for no future bounces as happened in 2008 and 2013.  Will the long-predicted decline finally ensue in 2017, or will the ice make a comeback as before?

No one knows what will happen to Arctic ice.

Except maybe the polar bears.

And they are not talking.

Except, of course, to the admen from Coca-Cola

Summary

There is no need to panic over Arctic ice this year, or any year.  It fluctuates according to its own ocean-ice-atmospheric processes and we can only watch and be surprised since we know so little about how it all works.  Judah Cohen at AER thinks the much greater snowfall in October and since will make for a very cold winter.  We shall see.  It is already adding more mass to the Greenland ice sheet than in previous years.

https://rclutz.wordpress.com/2016/07/06/warm-is-cold-and-down-is-up/

 

 

The darkest time of year at the North Pole is the Winter Solstice, this year on December 21. There has been no sunlight or even twilight since early October. The darkness lasts until the beginning of dawn in early March.

The regrowth of Arctic ice extent was slower than usual until recently. After showing resilience in September, ending higher than 2007, ice growth lagged in October, but is now rapidly ramping up toward the averages.  The map above shows the lack of ice is mainly in marginal seas close to the Atlantic and Pacific oceans.

In the last 19 days, 2016 ice extent has grown by 100k km2 per day, compared to the 10-year average 70k km2 per day.  As of day 355, 2016 ice extent is ~3% less than average (2006 to 2015).  The chart also shows the variability of ice extent over the years during this month.  2015 was the highest ice recovery rate in the last decade, while 2006 ended up the lowest.  The chart also shows 2016 Sea Ice Index (SII) from NOAA has been lagging behind by  ~300k km2.

The table below shows this year compared to average and to 2006 for day 355.

Region	2016355	Day 355 Average	2016-Ave.	2006355	2016-2006
(0) Northern_Hemisphere	12239317	12586206	-346889	12148183	91133
(1) Beaufort_Sea	1070445	1070151	294	1069711	734
(2) Chukchi_Sea	855058	964292	-109234	966006	-110948
(3) East_Siberian_Sea	1087137	1087134	3	1087137	0
(4) Laptev_Sea	897845	897841	4	897845	0
(5) Kara_Sea	851317	865798	-14482	909296	-57979
(6) Barents_Sea	169752	380779	-211027	225917	-56165
(7) Greenland_Sea	487544	584147	-96603	564913	-77369
(8) Baffin_Bay_Gulf_of_St._Lawrence	952305	886295	66010	733978	218328
(9) Canadian_Archipelago	853214	853009	205	852767	447
(10) Hudson_Bay	1241127	1152343	88784	1029530	211596
(11) Central_Arctic	3148299	3215124	-66825	3206257	-57957
(12) Bering_Sea	118989	329544	-210555	246957	-127968
(13) Baltic_Sea	15543	19208	-3665	16	15528
(14) Sea_of_Okhotsk	477365	266483	210882	344508	132857
(15) Yellow_Sea	0	4017	-4017	0	0
(16) Cook_Inlet	5555	7180	-1624	12191	-6635

The main deficit to average is in Barents and Greenland Seas on the Atlantic side, and in Bering and Chukchi Seas on Pacific side.  The Canadian and Siberian sides are locked in ice, with sizable surpluses in Baffin and Hudson Bays, along with Okhotsk.  The differences with 2006 are similar, though resulting in a surplus.

Summary

There is no need to panic over Arctic ice this year, or any year.  It fluctuates according to its own ocean-ice-atmospheric processes and we can only watch and be surprised since we know so little about how it all works.  Judah Cohen at AER thinks the much greater snowfall in October will make for a very cold winter.  We shall see.

https://rclutz.wordpress.com/2016/07/06/warm-is-cold-and-down-is-up/

 

 

Two days ago I posted on Resurging Arctic Ice, and noted that most of the deficit to average was in Barents Sea and Hudson Bay.  Look what happened yesterday in Hudson Bay.

Now compare to the day before

2016 Arctic ice extent has now topped 12M km2, and continues to close in on the average for the last decade.

Sea Ice Index from NOAA is running more than 400k km2 behind. But the polar bears know the truth, and it has set them free.

 

The regrowth of Arctic ice extent was slower than usual until recently. After showing resilience in September, ending higher than 2007, ice growth lagged in October, but is now rapidly ramping up toward the averages.  The map above shows the lack of ice is mainly in Hudson Bay and Barents Sea.  Kara, Bering and Chukchi seas have all rebounded and everything else is locked in ice.

In the last 24 days, 2016 ice extent has grown by 123k km2 per day, compared to the 10-year average 79k km2 per day.  As of day 348, 2016 ice extent is ~3% less than average (2006 to 2015).  The chart also shows the variability of ice extent over the years during this season.  2015 was the highest ice recovery rate in the last decade, while 2007 was the slowest.  The chart also shows 2016 Sea Ice Index (SII) from NOAA has been lagging behind by more than 300k km2.

The table below shows this year compared to average and to 2007 for day 348.

Region	2016348	Day 348 Average	2016-Ave.	2007348	2016-2007
(0) Northern_Hemisphere	11762663	12168943	-406279	11501915	260748
(1) Beaufort_Sea	1070445	1069448	997	1062676	7769
(2) Chukchi_Sea	937738	937789	-51	725062	212676
(3) East_Siberian_Sea	1087137	1083777	3361	1053584	33553
(4) Laptev_Sea	897845	897832	12	897845	0
(5) Kara_Sea	742906	866952	-124046	800920	-58014
(6) Barents_Sea	271463	352281	-80818	236964	34499
(7) Greenland_Sea	494089	559862	-65773	488595	5494
(8) Baffin_Bay_Gulf_of_St._Lawrence	846064	813794	32270	793616	52448
(9) Canadian_Archipelago	853214	853007	207	852556	658
(10) Hudson_Bay	820629	1085867	-265238	1217263	-396634
(11) Central_Arctic	3200307	3211105	-10799	3186190	14116
(12) Bering_Sea	227611	250748	-23137	54836	172775
(13) Baltic_Sea	11782	10418	1364	2898	8884
(14) Sea_of_Okhotsk	297235	166142	131093	127576	169659
(15) Yellow_Sea	0	1278	-1278	0	0
(16) Cook_Inlet	3055	6601	-3546	48	3007

Most of the deficit to average is in Kara, Barents, Greenland Sea and Hudson Bay.  Central Arctic, CAA, Bering Sea and BCE are all average, while Okhotsk and Baffin Bay are showing surpluses.  Presently 2016 ice extent is 260k km2 greater than 2007 with large surpluses in BCE, Baffin, Bering and Okhotsk Seas.  Only Hudson Bay and Kara were larger in 2007.

There is no need to panic over Arctic ice this year, or any year.  It fluctuates according to its own ocean-ice-atmospheric processes and we can only watch and be surprised since we know so little about how it all works.  Judah Cohen at AER thinks the much greater snowfall in October will make for a very cold winter.  We shall see.

https://rclutz.wordpress.com/2016/07/06/warm-is-cold-and-down-is-up/

 

 

The growth of Arctic ice extent has been slower than usual this year.  After showing resilience in September, ending higher than 2007, ice growth lagged in October, and is only now ramping up toward the averages.  The map above shows the lack of ice is mainly in Hudson Bay, and the slow freezing of Kara and Barents Seas.  Everything else is locked in ice, except for some open water in Bering and Chukchi.

In the last five weeks, 2016 ice growth has surged twice, firstly from day 303 to 314, and then the current surge the last 10 days starting day 325.  The chart also shows the variability of ice extent over the years during this season.  2015 was the highest ice recovery rate in the last decade, while 2006 was the lowest.  The chart also shows Sea Ice Index (SII) from NOAA is lagging over 300k km2 behind.

There is no need to panic over Arctic ice this year, or any year.  It fluctuates according to its own ocean-ice-atmospheric processes and we can only watch and be surprised since we know so little about how it all works.  Judah Cohen at AER thinks the much greater snowfall in October will make for a very cold winter.  We shall see.

https://rclutz.wordpress.com/2016/07/06/warm-is-cold-and-down-is-up/

 

 

Researchers found that ice conditions in the 19th century were remarkably similar to today’s, observations falling within normal variability. The study is Accounts from 19th-century Canadian Arctic Explorers’ Logs Reflect Present Climate Conditions (here) by James E. Overland, Pacific Marine Environmental Laboratory/NOAA, Seattle,Wash., and Kevin Wood, Arctic Research Office/NOAA, Silver Spring, Md.   H/t GWPF

Overview

This article demonstrates the use of historical instrument and descriptive records to assess the hypothesis that environmental conditions observed by 19th-century explorers in the Canadian archipelago were consistent with a Little Ice Age as evident in proxy records.  We find little evidence for extreme cold conditions.

It is clear that the first-hand observations of 19th-century explorers are not consistent with the hypothesized severe conditions of a multi-decadal Little Ice Age. Explorers encountered both warm and cool seasons, and generally typical ice conditions, in comparison to 20th-century norms.

Analysis

There were more than seventy expeditions or scientific enterprises of various types dispatched to the Canadian Arctic in the period between 1818 and 1910. From this number, we analyzed 44 original scientific reports and related narratives; many from expeditions spanning several years. The majority of the data come from large naval expeditions that wintered over in the Arctic and had the capacity to support an intensive scientific effort. A table listing the expeditions and data types is located at http://www.pmel.noaa.gov/arctic/history. The data cover about one-third of the possible number of years depending on data type, and every decade is represented.

Our analysis focuses on four indicators of climatic change: summer sea ice extent, annual sea ice thickness, monthly mean temperature, and the onset of melt and freeze as estimated from daily mean temperature. Historical observations in these four categories were compared with modern reference data; the reference period varied, depending on data availability.  Both sea ice extent and the onset of melt and freeze were compared to the 30- year reference period 1971–2000; monthly means are compared to the 50-year period 1951–2000. Modern sea ice thickness records are less continuous, and some terminate in the 1980s; the reference period is therefore based on 19 to 26 years of homogeneous record.

Fig.1.

(a) Proxy record of standardized summer air temperature variation derived from ice cores taken on Devon Island. This proxy record suggests that a significantly colder climate prevailed in the 19th century. Shading indicates temperatures one standard deviation warmer or colder than average for the reference period 1901–1960 [Overpeck,1998].

(b) Historical monthly mean temperature observations compared to the 20th-century reference period 1951–2000. Sixty-three percent of 343 monthly mean temperatures recorded on 19th-century expeditions between 1819 and 1854 fall within one standard deviation of the reference mean at nearby stations (reference data from Meteorological Service of Canada,2002; and National Climatic Data Center,2002).

(c) Onset of melt observed by expeditions between 1820 and 1906 expressed as departures from the mean for the reference period 1971–2000. The period of melt transition observed by 19th century explorers is not inconsistent with modern values.

(d) Onset of freeze observed between 1819 and 1905 compared to the reference period 1971–2000. The onset of freeze transition is frequently consistent with modern values,but in some cases occurred earlier than usual. The incidence of an early onset of freeze represents the largest departure from present conditions evident in the historical records examined in this study. Melt and freeze transition dates for the reference period 1971–2000 were calculated from temperature data extracted from the Global Daily Climatology Network data base (National Climate Data Center, 2002).

Fig.2. The ship tracks and winter-over locations of Arctic discovery expeditions from 1818 to 1859 are surprisingly consistent with present sea ice climatology (contours represented by shades of blue). The climatology shown reflects percent frequency of sea ice presence on 10 September which is the usual date of annual ice minimum for the reference period 1971–2000 (Canadian Ice Service,2002). On a number of occasions,expeditions came within 150 km of completing the Northwest Passage, but even in years with unfavorable ice conditions, most ships were still able to reach comparatively advanced positions within the Canadian archipelago. By 1859, all possible routes comprising the Northwest Passage had been discovered.

Summary

As stated here before, Arctic ice is part of a self-oscillating system with extents expanding and retreating according to processes internal to the ocean-ice-atmosphere components. We don’t know exactly why 19th century ice extent was less than previously or less than the 1970s, but we can be sure it wasn’t due to fossil fuel emissions.

Explorers encountered both favorable and unfavorable ice conditions. This drawing from the vicinity of Beechey Island illustrates the situation of the H.M.S.Resolute and the steam-tender Pioneer on 5 September 1850 [from Facsimile of the Illustrated Arctic News,courtesy of Elmer E.Rasmuson Library,Univ.of AlaskaFairbanks].

After showing resilience in September, Arctic ice growth faltered in October. The graph shows MASIE ice extents in 2016 in comparison with other years and with SII.

Last year was above average recovery, while both SII and MASIE showed October 2016 well below average, and dropping down to 2007 levels.  Interestingly, MASIE and SII show virtually the same October average extent at 6.1M km2. This amount is 10% below the MASIE Oct. ten-year average, but that figure is 21% below SII ten-year average for Oct.

6.1M km2 is still higher than 2007 by 260k km2 despite the slower recovery this year. There is an anomaly in the 2007 record: a leap of almost 800k km2 on day 303, including E. Siberian Sea adding 460k km2 to more than double its ice extent on that day. Kara also jumped up 50% adding 100k km2. More about those two seas below.

The table shows by region the differences from ten-year MASIE averages for day 305, October 31.

Region	2016305	Day 305 Average	2016-Ave.
(0) Northern_Hemisphere	7283405	8865912	-1582506
(1) Beaufort_Sea	769445	998163	-228718
(2) Chukchi_Sea	276157	506472	-230315
(3) East_Siberian_Sea	600113	1001404	-401291
(4) Laptev_Sea	897773	897591	182
(5) Kara_Sea	162641	559477	-396837
(6) Barents_Sea	24522	120163	-95641
(7) Greenland_Sea	319983	446472	-126489
(8) Baffin_Bay_Gulf_of_St._Lawrence	262586	260072	2514
(9) Canadian_Archipelago	728911	782867	-53956
(10) Hudson_Bay	76561	90864	-14303
(11) Central_Arctic	3155443	3190140	-34696

Clearly in the table and in the image at the top, ice is missing mostly in the BCE region on the Asian side (Beaufort, Chukchi, E. Siberian) and in Kara and Barents on the European side.

What’s Up with Arctic Ice

For some answers let’s turn first to Dr. Judah Cohen of AER.  His analysis is thorough and interesting (here).  Excerpts:

Arctic sea ice reached its annual minimum on September 10th. Despite an initial spurt of ice growth, overall Arctic sea ice has been expanding at an anemic rate. One area to note on the North Atlantic side is in the Barents-Kara Seas, where negative sea ice anomalies have been growing (Figure 13). Recent research has shown that regional anomalies are important and the sea ice region most highly correlated with the winter AO is the Barents-Kara seas region where low Arctic sea ice favors a negative winter AO. Given that sea ice is running well below normal, this currently favors further expansion Eurasian snow cover in the coming weeks, followed by a strengthened Siberian high and a weakened polar vortex/negative AO this upcoming winter.

Sea ice anomalies also remain well below normal in the East Siberian, Chukchi and Beaufort seas. This could favor positive geopotential height anomalies near Alaska forcing downstream troughing and cold weather across eastern North America.

Additional context is provided by Ted Scambos in an article at Discover Magazine (here).  Excerpts below.

Arctic sea ice has been been growing much more slowly than average during October. Its geographic extent is now in a rough tie with 2007 and 2012 for the lowest in the satellite record for this point in the year.

According to Ted Scambos, a senior research scientist at the National Snow and Ice Data Center, two big Arctic storms in August contributed significantly to the very low minimum extent in September. They probably did this by “breaking up the ice and stirring a trace of warmth up from deeper levels (5-10 meters below the surface),” Scambos says.

Not too long after the storminess subsided, the surface water refroze fairly quickly, accounting for the rapid expansion of sea ice after the low point in September sea ice

That rapid freeze-up happened in the heart of the Arctic Ocean, where temperatures would normally be coldest. Meanwhile, more at the periphery — particularly in the Barents, Kara and Laptev seas — “the ice edge has barely changed,” says Julienne Stroeve, who also serves as a senior research scientist at NSIDC.

Sea ice has expanded somewhat in the Beaufort Sea, off Alaska’s north coast. But the edge is still far short of the long term average in the region.

Why has Arctic sea ice been growing so slowly in the past couple of weeks? Once again, Scambos thinks the warm sea surface temperatures that have been retarding ice growth also have been warming the air directly above the water. 

“It is not surprising that it is extremely warm in the weeks following near-record minimums [in sea ice extent], because open water beneath the atmosphere buffers the air temperature at near the freezing point for several weeks until several centimeters of ice are formed,” Scambos says. “In this case, I would say that the high air temperature anomaly is an effect, not a cause, of slow sea ice growth, and that the leading cause is ocean temperature.”

What’s the Winter Outlook

Dr. Judah Cohen provides his seasonal forecast here.

Snow cover advance across Eurasia continued consistently above normal for the entire month of October. Also because much of the advance has occurred at latitudes south of 60°N, the snow advance index is also well above normal. Above normal snow cover extent, especially south of 60°N, favors a strengthened Siberian high and a weakened polar vortex/negative AO this upcoming winter with cold temperatures across the continents of the NH.

Predicted winter surface temperature anomalies for the United States Dec-Jan-Feb 2016/17 in degrees Fahrenheit. The model is forecasting colder than normal temperatures for much of the Eastern United States, with warmer than normal temperatures for the Western United States. The model uses October Siberian snow cover, sea level pressure anomalies, predicted El Niño/Southern Oscillation anomalies, and observed September Arctic sea ice anomalies. October Siberian snow cover has so far this month advanced at an above normal rate. This is an indication of an increased probability of a weakened polar vortex or a sudden stratospheric warming, and a predominantly negative Arctic Oscillation during the winter and cold temperatures – especially east of the Mississippi. This is a preliminary forecast and not the official winter forecast as the model requires full monthly values for snow cover and sea level pressure anomalies. The forecast will be updated next month. Current forecast produced on October 19, 2016.

Summary

Both Cohen and Scambos explanations are consistent with the lack of ice both in BCE on the Asian side (Beaufort, Chukchi, E. Siberian), and in Kara and Barents on the European side.

Warm ocean water reduces ice extents, the air is warmed and moistened, and then snowfall increases. The conditions for winter ice formation are shaping up, though somewhat later than usual, perhaps because of August storm activity.

Lastly, it’s important to realize that two weeks is a short period of time, and Arctic sea ice could start forming up at a more normal rate for this time of year. In fact, that’s what Scambos predicts: “Now I expect that sea ice growth will take a more typical path (typical for the past 5-8 years, at least).”

One more time: Remember the 3 Ws when it comes to Arctic ice extents.  Firstly it’s the Water, and then the Wind circulation, and the Weather.  Air temperatures are an effect not the cause.

For more on this see:

Arctic Sea Ice: Self-Oscillating System

Arctic Not a Refrigerator

August 2016 was the final report of the Sea Ice Prediction Network (SIPN) before the actual September monthly extent is reported by NOAA Sea Ice Index (SII). The report (here) gave this overview:

This month the median pan-Arctic extent Outlook for September 2016 sea ice extent is 4.4 million square kilometers (km2) with quartiles of 4.2 and 4.7 million km2, which is slightly higher than July’s value (4.3 million km2) (See Figure 1 in the full report, below). If the median Outlook should agree with the observed estimate come September, this year would be the third lowest September in the satellite record. The spread in the Outlook contributions narrowed slightly from July to August, with an overall range this month of 3.7 to 5.2 million km2.

The chart shows the September monthly sea ice extents from NOAA and MASIE. Both datasets show 4.5 M km2, above what was predicted, about 200k km2 higher than 2007.

In MASIE, September 2016 finished as the fourth lowest, ahead of 2012, 2007 and 2008, slightly behind 2011.

Some had anticipated a late Arctic cyclone might again produce a lower number, as it did in 2012. But as the chart shows, 2012 is looking increasingly like an outlier, interrupting a steady recovery of ice extents since 2007.