Author Topic: Home brew AMSR2 extent & area calculation (Read 1991273 times)

Wipneus · « **Reply #1550 on:** May 14, 2015, 08:53:40 AM »

Update 20150513.
Changes from 20150506.

Extent: -492.2 (-187k vs 2014, NAk vs 2013, -589k vs 2012)
Area: -388.9 (-127k vs 2014, NAk vs 2013, -484k vs 2012)

Strong declines in total extent (average -80k/day) and area (average -55k/day) strengthen the lead that 2015 has over the previous 3 years (no data for 2013 due to a AMSR2 outage, but the distance is at least -400k). Regionally all the peripheral regions contribute, with Baffin Bay and the Bering region contributing most. The drop in the Hudson Bay may be regarded as "early".
More North there area's of open water developed on the Beaufort-Canadian Archipelago border, the Chukchi region and in the Kara Sea.
Warm temperatures are predicted over the Beaufort and Chukchi but that is not very visible. Actually extent and area are still above 2014. Regions of melting ice that popped up during last week in the Jaxa thickness map have disappeared quickly. Actually total melting area has dropped (again) this week and is now below 2013 and 2014 levels. Also not conductive to solar absorption are the high levels of compactness of the ice pack, these are still at values normal to mid-winter (see the compactness graph

(click the thickness map for an animation)

You will find the updated graphs in the top post

The details (in 1000 km²):

Extent: Central Arctic Basin East Siberian Sea Laptev Sea -25.7 -6.1 4.1 Kara Sea Barents Sea Greenland Sea -30.6 -41.4 18.2 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -130.5 -0.1 -58.0 Canadian Archipelago Beaufort Sea Chukchi Sea -32.3 -29.6 -16.4 Bering Sea Sea of Okhotsk Total Extent -85.6 -58.2 -492.2 Area: Central Arctic Basin East Siberian Sea Laptev Sea -9.3 -21.3 14.6 Kara Sea Barents Sea Greenland Sea -27.1 -29.8 22.2 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -97.8 0.2 -61.9 Canadian Archipelago Beaufort Sea Chukchi Sea -8.9 -41.5 -25.5 Bering Sea Sea of Okhotsk Total Area -55.3 -47.5 -388.9

Wipneus · « **Reply #1551 on:** May 14, 2015, 09:07:23 AM »

Hudson Bay has opened early, with help from NW winds.

Wipneus · « **Reply #1552 on:** May 18, 2015, 08:22:39 AM »

The Hudson keeps dropping, today with -34k (while Beaufort and Chukchi declined with -6k and -7k).

I have updated the animation in the post above.

Neven · « **Reply #1553 on:** May 18, 2015, 08:54:45 AM »

I'm very surprised at what's happening in Hudson Bay. I thought this winter's cold blast over NE America would have made the ice super strong.

plinius · « **Reply #1554 on:** May 18, 2015, 11:26:42 AM »

The models do not seem to have had very thick ice progged there. Would it be possible that we just did not have enough wind/too much early snow on the Bay?

Neven · « **Reply #1555 on:** May 18, 2015, 11:45:28 AM »

Yes, I think snow could be a factor here.

Gray-Wolf · « **Reply #1556 on:** May 18, 2015, 01:32:28 PM »

Maybe 'snow' will become a thing for all the peripheries with larger dumps effectively sealing in ocean heat that could have been shed into the atmosphere before freeze commenced? As the slush freezes it 'caps' the warmth below but that warmth then limits growth into the 'warmth' below?

plinius · « **Reply #1557 on:** May 18, 2015, 01:54:07 PM »

Slush isn't effective on that, but a simple snow cover on young ice is. You need the air as insulation.

Yuha · « **Reply #1558 on:** May 18, 2015, 02:24:18 PM »

I still think it's mostly compaction by winds. Below is a snapshot from May 12, 2014. There's a lot of room for compaction and with similar winds it might have looked much like this year.

plinius · « **Reply #1559 on:** May 18, 2015, 03:32:59 PM »

looks rather like real loss:
http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn_nowcast_anim30d.gif

otherwise this would show significant thickening.

Yuha · « **Reply #1560 on:** May 18, 2015, 06:52:17 PM »

Quote from: plinius on May 18, 2015, 03:32:59 PM

looks rather like real loss:
http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn_nowcast_anim30d.gif

otherwise this would show significant thickening.

There is undoubtedly some melting going on but without the help of winds it would not show very strongly as a drop in extent until later.

Yuha · « **Reply #1561 on:** May 18, 2015, 07:30:09 PM »

Quote from: Yuha on May 18, 2015, 06:52:17 PM

There is undoubtedly some melting going on but without the help of winds it would not show very strongly as a drop in extent until later.

Continuing the thought, I would be similarly cautious in interpreting too much to short term changes in extent elsewhere such as Beaufort and Kara. I think area is a more meaningful indicator of actual melt progress, not only because it is less sensitive to compaction by winds, but because it is more sensitive to melt pond formation.

Wipneus · « **Reply #1562 on:** May 21, 2015, 08:50:01 AM »

Update 20150520.
Changes from 20150513.

Extent: -456.4 (-363k vs 2014, -739k vs 2013, -610k vs 2012)
Area: -407.9 (-137k vs 2014, -548k vs 2013, -431k vs 2012)

Another week with strong declines: -65k/day fro extent and -60k/day for area. The lead with the other recent years is increasing, except perhaps for 2014-area. The latter is an indication that compactness (the ratio of area and extent) is high, actually it is the highest from 2012-2015 as can be seen on the compactness graph. That is despite the melt starting in the Beaufort and Chukchi regions, as can be seen on the ADS-NIPR AMSR2 ice thickness/melting map.

Regionally most of the melt occurred in the Hudson Bay, followed by the Bering, Baffin and Kara regions. Beaufort and Chukchi where the melting powers seem to be high dropped by -39k and -35k. One thing is certain: Bering region will next week not be in the top of this list, there just is not enough ice left there.
Regions that are not joining the melting crowds are the ESS and Laptev. It is too early to call that late, but something should be happening there soon or it will be.

You will find the updated graphs in the top post

The details (in 1000 km²):

Extent: Central Arctic Basin East Siberian Sea Laptev Sea 5.1 2.6 -14.1 Kara Sea Barents Sea Greenland Sea -55.8 7.4 -39.9 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -67.4 -0.2 -109.8 Canadian Archipelago Beaufort Sea Chukchi Sea -17.7 -38.5 -35.1 Bering Sea Sea of Okhotsk Total Extent -81.1 -11.8 -456.4 Area: Central Arctic Basin East Siberian Sea Laptev Sea 3.9 23.2 -11.9 Kara Sea Barents Sea Greenland Sea -47.9 2.9 -47.8 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -83.6 -0.1 -69.3 Canadian Archipelago Beaufort Sea Chukchi Sea -37.3 -39.3 -27.9 Bering Sea Sea of Okhotsk Total Area -65.9 -6.9 -407.9

Anne · « **Reply #1563 on:** May 21, 2015, 09:13:07 AM »

Quote from: Wipneus on May 21, 2015, 08:50:01 AM

It is too early to call that late, but something should be happening there soon or it will be.

Wipneus, you are one of my heroes on here, witty as well

Wipneus · « **Reply #1564 on:** May 21, 2015, 11:41:06 AM »

Quote from: Anne on May 21, 2015, 09:13:07 AM

Quote from: Wipneus on May 21, 2015, 08:50:01 AM
It is too early to call that late, but something should be happening there soon or it will be.
Wipneus, you are one of my heroes on here, witty as well

Thank you. That was struggling with the English language actually, but I am glad you understood me.

Wipneus · « **Reply #1565 on:** May 21, 2015, 11:46:09 AM »

Freezing conditions no more in the north of the Baffin Bay and in the Parry Channel. The remaining ice is swept away leaving lots of open water (causing a drop in extent locally).

Wipneus · « **Reply #1566 on:** May 22, 2015, 09:32:18 AM »

The open water in the Amundsen Gulf is not satisfied with the situation and breaking the remaining local ice pack loose. Several "toches" and interesting leads are visible.

jdallen · « **Reply #1567 on:** May 22, 2015, 05:25:42 PM »

Quote from: Wipneus on May 21, 2015, 11:46:09 AM

Freezing conditions no more in the north of the Baffin Bay and in the Parry Channel. The remaining ice is swept away leaving lots of open water (causing a drop in extent locally).

I was noticing that; it's definitely not waiting to melt in place.

The ice seems more mobile than in the past, and more consistently getting exported into the Labrador Sea. As spring heat and flow from the south heats that body, it will insure the ice won't last.

Bets on how soon the NW passage will open?

Jim Hunt · « **Reply #1568 on:** May 22, 2015, 06:29:59 PM »

Quote from: jdallen on May 22, 2015, 05:25:42 PM

Bets on how soon the NW passage will open?

Sooner than 2013!

Wipneus · « **Reply #1569 on:** May 24, 2015, 03:59:46 PM »

The dramatic drop of ice cover in the Bering Sea is just about complete (with almost nothing left). The last bit "escaped" to the Chukchi but won't survive there much longer.

Wipneus · « **Reply #1570 on:** May 28, 2015, 09:59:37 AM »

Update 20150527.
Changes from 20150520.

Extent: -362.1 (-442k vs 2014, -748k vs 2013, -564k vs 2012)
Area: -393.4 (-232k vs 2014, -544k vs 2013, -280k vs 2012)

Declines in Arctic sea ice continue at a steady pace, -50k/day for total extent and -55k/day for total area. Regionally (see also the attached delta map) the melting was greatest in the Barents and Kara regions, the Beaufort and Chukchi regions and the Hudson Bay region.
The latter contributes now nearly half of the difference in extent between 2014 and 2015, with the remainder mostly in the other peripheral regions.
Within the Arctic Basin (CAB+Beaufort+Chukchi+ESS+Laptev), 2015 and 2014 are about the same: -10k by extent and +15k by area.
Melting did start in the some area's in the Beaufort, Chukchi and ESS. Total melt area (as indicated by the ADS-NIPR AMSR2 maps) is less than in 2013 and 2014 though.
In summary, the increased melting in the Beaufort and Chukchi is still compensated by the stronger ice condition in the rest of the Basin, with a focus in the Laptev region. Are the differences more important or will 2015 show that in the end the similarities are counting?

You will find the updated graphs in the top post

The details (in 1000 km²):

Extent: Central Arctic Basin East Siberian Sea Laptev Sea -11.9 -1.4 -0.3 Kara Sea Barents Sea Greenland Sea -60.0 -61.4 -12.0 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -28.5 0.4 -40.3 Canadian Archipelago Beaufort Sea Chukchi Sea -3.0 -35.0 -49.0 Bering Sea Sea of Okhotsk Total Extent -25.5 -34.3 -362.1 Area: Central Arctic Basin East Siberian Sea Laptev Sea -15.8 -5.9 -8.4 Kara Sea Barents Sea Greenland Sea -56.1 -57.3 -15.0 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -23.3 0.3 -49.0 Canadian Archipelago Beaufort Sea Chukchi Sea -0.9 -36.8 -77.0 Bering Sea Sea of Okhotsk Total Area -19.5 -28.7 -393.4

Wipneus · « **Reply #1571 on:** May 28, 2015, 10:33:16 AM »

In the animation the torching of areas in the Beaufort and Chukchi seas can be seen.
It is interesting that an area of widespread cracking (in the lower half of the image) is seen as anomalous thick ice in the ADS-NIPR thickness map (seen in the previous post). Peculiar errors that can be seen in the Kara Sea as well.

Entropy101 · « **Reply #1572 on:** May 28, 2015, 11:08:41 AM »

Quote from: Wipneus on May 28, 2015, 10:33:16 AM

It is interesting that an area of widespread cracking (in the lower half of the image) is seen as anomalous thick ice in the ADS-NIPR thickness map (seen in the previous post). Peculiar errors that can be seen in the Kara Sea as well.

Couldn't that be due to a setting in how the sensor output is used when drawing a map? When doing a survey and choosing the data point for a grid, several settings can be used. Depending on grid size, you have a set of thickness data points in a grid to choose from. You can choose how these data points are used to get your end value used to plot the map. In the cracked area, there is probably a lot of ridging going on, which will show a high thickness. If the algorithm that analyses all data points in the set has a bias towards the use of the data points with a larger thickness, than it will show a thick ice sheet, where there is actually no ice or very limited ice.

Wipneus · « **Reply #1573 on:** May 28, 2015, 12:56:40 PM »

Quote from: Entropy101 on May 28, 2015, 11:08:41 AM

When doing a survey and choosing the data point for a grid, several settings can be used. Depending on grid size, you have a set of thickness data points in a grid to choose from. You can choose how these data points are used to get your end value used to plot the map. In the cracked area, there is probably a lot of ridging going on, which will show a high thickness. If the algorithm that analyses all data points in the set has a bias towards the use of the data points with a larger thickness, than it will show a thick ice sheet, where there is actually no ice or very limited ice.

E101,

The thickness algorithm uses microwave bands of 36 and 18GHz. The AMSR2 field of view of these bands are 7x12km and 14x22km. For the melting ratio the 7GHz band is used which has a FOV of 35x62km. The data is gridded using a polar stereographic projection with nominal cell size of 10x10km. So there is not much to choose from and signatures of of ice and open water will mix inevitably.
(the concentration map is on the 3.125x3.125km grid, as the ASI algorithm uses the 89GHz bands with FOV of 3x5km)

Note that ADS provides thickness maps composed from ascending and descending orbit positions. Descending is the default, but (interestingly again) the ascending map does not show such a "anomalous thick ice" area. Not to mention the much larger "melting" area.

Which might indicate that the combining of orbital swaths is another important factor.

Entropy101 · « **Reply #1574 on:** May 28, 2015, 01:41:41 PM »

@ Wipneus,

I understand what you are explaining and I like your point on orbit positions.

What I tried to discuss is that normally in a survey a multitude of measurements are done in one grid cell. All those measurements are than analyzed and processed to one single value for that grid cell. Depending on how you analyze (i.e - use the average of the 10% highest, average all, average 10% lowest or whatever algorithm you want) you get a different result for that cell. Depending on what you want to convey or expect, all can be plausible but can also give a bias in your representation of your map.

Andreas T · « **Reply #1575 on:** May 28, 2015, 02:05:06 PM »

are ascending / descending tracks always at the same time of the day? MODIS orbit tracks obviously are set to descend around midday and ascend on the night side of the globe IIRC. Day / night differences are not so large at this time in the arctic but could sun angle explain the differences in Satellite data?

Wipneus · « **Reply #1576 on:** May 28, 2015, 02:21:58 PM »

Quote from: Andreas T on May 28, 2015, 02:05:06 PM

are ascending / descending tracks always at the same time of the day? MODIS orbit tracks obviously are set to descend around midday and ascend on the night side of the globe IIRC. Day / night differences are not so large at this time in the arctic but could sun angle explain the differences in Satellite data?

GCOM-W1, the satellite that bears the AMSR2 instrument, is "parked" in the NASA satellite cloud called A-train. AQUA (that bears the defunct AMSR-E instrument) is there as well.
The A in A-train is afternoon, meaning that it passes the equator in the descending direction at local afternoon. The ascending node is 12 hours apart, during the night. The timing will be skewed in the (ant)arctic, but ascending and and descending must be hours apart yes.

nukefix · « **Reply #1577 on:** May 28, 2015, 05:41:46 PM »

Quote from: Andreas T on May 28, 2015, 02:05:06 PM

are ascending / descending tracks always at the same time of the day?

Yes if the orbit is sun-synchronous.

plinius · « **Reply #1578 on:** May 29, 2015, 01:05:29 AM »

but not for the same region. Such a satellite (depending on the precise altitude) has typically an around 99 minute orbit. Which means that the satellite will have 14.5 orbits per day, i.e. also an about 24 degree spacing in latitude that progresses about 12 degrees every day. Concentration will be biased by time differences (in particular if they have the descending data too and use a small viewing angle, which brings you to the night side, depending on algorithm), and by viewing angle of the surface.
In my naive picture that's weather and ice situation dependent (plus the geometric terms), so that you should expect a fluctuating oscillation with a period of around 2 days. Might be visible if one correlates the concentrations with the positions of the orbit swaths.

seaicesailor · « **Reply #1579 on:** May 29, 2015, 07:08:00 AM »

Will "Santa's backyard" thermometer (DMI 80+) hit 0C on June 1st?
Some cooling on American ~~Pacific~~ side (freezing temps at "night", maybe open skies under the huge High?)
But it is the turn of Laptev and ESS. Kara keeps being beaten
Very interesting. Nothing like '13 or '14, for the moment

PS: HYCOM+CICE thickness evolution, predicts compaction in next 7 days

Wipneus · « **Reply #1580 on:** May 29, 2015, 08:32:58 AM »

Quote from: plinius on May 29, 2015, 01:05:29 AM

but not for the same region. Such a satellite (depending on the precise altitude) has typically an around 99 minute orbit. Which means that the satellite will have 14.5 orbits per day, i.e. also an about 24 degree spacing in latitude that progresses about 12 degrees every day. Concentration will be biased by time differences (in particular if they have the descending data too and use a small viewing angle, which brings you to the night side, depending on algorithm), and by viewing angle of the surface.
In my naive picture that's weather and ice situation dependent (plus the geometric terms), so that you should expect a fluctuating oscillation with a period of around 2 days. Might be visible if one correlates the concentrations with the positions of the orbit swaths.

Yes plinius, I agree. Yesterday I could not find the orbital details so quickly, but the satellites in the A-train have a cycle of 16 days/233 orbits after which the tracks repeat. That works out to be 233/16 = 14.5625 orbits per day.

For completeness the F17 DMSP satellite, that carries the SSMIS instrument which NSIDC depends on has a cycle of 17 days/240 orbits. That gives 14.1176 orbits/day. There an "fluctuating oscillation with a period of around 2 days" is not to be expected, although it is observed.

nukefix · « **Reply #1581 on:** May 29, 2015, 09:35:57 AM »

Quote from: nukefix on May 28, 2015, 05:41:46 PM

Quote from: Andreas T on May 28, 2015, 02:05:06 PM
are ascending / descending tracks always at the same time of the day?
Yes if the orbit is sun-synchronous.

More specifically, the observations from the same track (identified with the track-number) always happens at the same time of the day. The exact time-of-day depends on the track number. Many EO satellites are on a dawn-dusk sun-synchronous orbit:

https://en.wikipedia.org/wiki/Sun-synchronous_orbit

Wipneus · « **Reply #1582 on:** May 30, 2015, 10:25:59 AM »

Melt in the Laptev may be two weeks behind 2014.

(click req'd)

Andreas T · « **Reply #1583 on:** May 31, 2015, 08:22:27 AM »

correct me if I am wrong but I believe there is something odd with the ice concentration shown by the AMSR2 sensor in this region. polarview shows this for the 30.5.

where worldview shows unbroken landfast ice.

the loose ice north of the open water in in a slightly different shape as well but that may be due to images taken at different times?

werther · « **Reply #1584 on:** May 31, 2015, 09:46:13 AM »

Andreas, hi,
It is a common ‘misrepresentation’ by ASMR-2 this time of the year. Especially on the Laptev shores. Spring has just arrived, according to today’s Wetteronline forecast:

SE winds bring in low, moist clouds and fog, affecting snow cover on the fast ice.
Climate Reanalyzer and Wetteronline project continuing temperature increase on the shores of East Sib and Laptev Seas. For 5 June, the Lena delta will get to 16-18 dC daily high, so the MODIS images will turn green/brown during next week Which is quite late out there, compared to earlier years.

Wipneus · « **Reply #1585 on:** May 31, 2015, 11:44:59 AM »

Andreas, what Werther said we have seen this before. Not sure what it is but it is associated with warming conditions around 0^oC and water vapor and clouds. Smoke was suggested, but I do not see it on MODIS yet.
And yes it is what you see on hi-res AMSR2 concentration maps using the ASI algorithm.

Wipneus · « **Reply #1586 on:** June 01, 2015, 08:45:10 AM »

Hudson Bay region is the region where ice is disappearing fastest, far earlier than usual.

Wipneus · « **Reply #1587 on:** June 02, 2015, 08:18:26 AM »

The "Grand Crack" is back!

(click you must)

seaicesailor · « **Reply #1588 on:** June 02, 2015, 08:31:24 AM »

I can see at the left that the Beaufort is getting more cracks.

Wipneus · « **Reply #1589 on:** June 04, 2015, 10:05:45 AM »

Update 20150603.
Changes from 20150527.

Extent: -356.9 (-346k vs 2014, -758k vs 2013, -571k vs 2012)
Area: -312.2 (+155k vs 2014, -491k vs 2013, -295k vs 2012)

Arctic Sea Ice continued to decline with total extent dropping at an average rate of about 50k/day. Total area was a bit slower at 44k/day. That extent is dropping faster than area can also be seen when comparing the numbers with previous years: the difference in extent is nearly twice as big compared with 2012 and 2014. This is exactly what is shown in the compactness graph with compactness highest for the date. The meaning of this is simple, the average concentration within the ice pack's extent is high, little open water. That makes the albedo higher, solar absorption lower and heat loss by thermal radiating higher when everything else is the same.
That seldom seems the case, and with temperatures now near zero or even above zero in the whole Arctic Basin, melting can occur everywhere and extensive meltponding can occur. The ADS-NIPR sea ice thickness and melting maps do shows this to be happening this week in the Hudson, Kara and Barents regions. Within the Arctic Basin the melting area is still small but occur very near the Pole (watch closely) and I suspect that the algorithms may show open water (and perhaps melt ponds) as very thick ice. For example look at the Beaufort regions. Compared with 2013 and 2014 the melting area in the early season is considerable less.

You will find the updated graphs in the top post

The details (in 1000 km²):

Extent: Central Arctic Basin East Siberian Sea Laptev Sea -14.3 -19.4 -42.6 Kara Sea Barents Sea Greenland Sea -58.3 -41.3 -35.5 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -33.1 -0.7 -11.6 Canadian Archipelago Beaufort Sea Chukchi Sea -5.7 -1.9 -72.4 Bering Sea Sea of Okhotsk Total Extent -6.3 -13.8 -356.9 Area: Central Arctic Basin East Siberian Sea Laptev Sea -47.1 -36.4 -37.5 Kara Sea Barents Sea Greenland Sea -52.3 -37.0 -22.7 Baffin/Newfoundland Bay St. Lawrence Hudson Bay -35.0 -0.6 0.1 Canadian Archipelago Beaufort Sea Chukchi Sea 14.7 2.9 -44.4 Bering Sea Sea of Okhotsk Total Area -5.9 -11.2 -312.2

Wipneus · « **Reply #1590 on:** June 04, 2015, 10:15:49 AM »

There was not much regional information in the weekly report, but Chukchi show the biggest decline of the lot. As the animation shows, there is a lot of compaction and some melt..

Wipneus · « **Reply #1591 on:** June 05, 2015, 08:44:53 AM »

Export through the Fram Strait is still active and will be for the next few days if the wind forecasts of nullscholl is to be believed. That is in contrast with 2014 when the ice could not get out through the Fram and during the summer it had to go to the next exit: the Victoria Channel (IIRC) ~~west~~ east of Svalbard.

(do click please)

Neven · « **Reply #1592 on:** June 05, 2015, 10:00:44 AM »

Quote from: Wipneus on June 05, 2015, 08:44:53 AM

That is in contrast with 2014 when the ice could not get out through the Fram and during the summer it had to go to the next exit: the Victoria Channel (IIRC) west of Svalbard.

You mean east of Svalbard?

Wipneus · « **Reply #1593 on:** June 05, 2015, 10:18:01 AM »

Quote from: Neven on June 05, 2015, 10:00:44 AM

You mean east of Svalbard?

How careless of me! Corrected and thanks.

Wipneus · « **Reply #1594 on:** June 07, 2015, 10:05:28 AM »

Kara shows the largest drop in ice cover during the past week.

Wipneus · « **Reply #1595 on:** June 08, 2015, 08:34:41 AM »

Area and extent keep dropping, but not as fast as in the last few years. Total area is now above 2012 and 2014, and the lead in extent is getting less:

Update 20150607.
Extent: -48.3 (-256k vs 2014, -572k vs 2013, -239k vs 2012)
Area: -73.1 (+25k vs 2014, -179k vs 2013, +120k vs 2012)

Regionally the declines are mostly within the Arctic Basin though:

Extent: Central Arctic Basin East Siberian Sea Laptev Sea -0.7 3.6 -24.7 Kara Sea Barents Sea Greenland Sea -2.8 -6.2 16.0 Baffin/Newfoundland Bay St. Lawrence Hudson Bay 0.9 0.7 4.0 Canadian Archipelago Beaufort Sea Chukchi Sea -0.8 -19.7 -9.2 Bering Sea Sea of Okhotsk Total Extent -0.5 -8.9 -48.3 Area: Central Arctic Basin East Siberian Sea Laptev Sea 8.6 3.4 -40.4 Kara Sea Barents Sea Greenland Sea -8.7 -10.9 8.5 Baffin/Newfoundland Bay St. Lawrence Hudson Bay 2.7 0.7 -4.6 Canadian Archipelago Beaufort Sea Chukchi Sea -2.0 -15.8 -9.4 Bering Sea Sea of Okhotsk Total Area -0.4 -4.9 -73.1

Wipneus · « **Reply #1596 on:** June 08, 2015, 08:47:18 AM »

Lots of open water in the Beaufort. Some of it is between a mix of ice floes but is resolved by the hi-res sea ice concentration product from Uni Hamburg, where NSIDC show only ice.
It is eating into the Canadian Archipelago as well.

(needs a click)

Wipneus · « **Reply #1597 on:** June 09, 2015, 09:27:59 AM »

The Laptev region awakens and that goes with some serious "torching" of the fast ice. So serious that the UH AMSR2 sea ice concentration dropped yesterday under 15% . Same is true for the Jaxa AMSR2 sea ice concentration, normally the product not so sensitive to these events.
Attached are today's images, when there appears some recovery.

Wipneus · « **Reply #1598 on:** June 09, 2015, 09:31:06 AM »

And an animation, comparing with the same date in 2014. As can be seen the event is not that unprecedented (but the the actual Laptev bite was already much bigger then).

Wipneus · « **Reply #1599 on:** June 10, 2015, 09:25:45 AM »

Interestingly NSIDC ice index and ADS-NIPR Jaxa Arctic sea ice extent melt has shown a "hiatus" if not a downright uptick. This is not seen in the Uni Hamburg 3.125 km AMSR2 product.

Regionally most of this is caused by the ice in the Greenland Sea, Jaxa sees the extent grow nearly +100k in a few days.
The animation shows what is happening: the ice in the southern part of the Greenland Sea is spreading and melting. Only the UH AMSR2 map has a good enough resolution not to see this as an increase in extent.

(click the AMSR2 map to start that animation)

News:

Author Topic: Home brew AMSR2 extent & area calculation (Read 1991273 times)

seaicesailor

seaicesailor