Support the Arctic Sea Ice Forum and Blog

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Steven

Pages: [1] 2 3 ... 13
The politics / Re: The Alt Right
« on: October 11, 2020, 01:01:41 PM »
Any attempt to shift the discussion to the "alt-left" is disingenuous and should be treated as such.

It's the usual trolling by Tom_Mazanec.  That guy has been spouting far right-wing propaganda all over this forum. 

I'm glad I live in Western Europe and not in the USA.  Here in Belgium we had an openly gay prime minister a few years ago and now we have a transgender minister, and nobody had a problem with that.  And we have very liberal laws on abortion, euthanasia etc.  I cannot imagine any of that happening in the USA: there would be nonstop protests from right wing religious zealots.

Unfortunately, the far right-wing party here in the rich northern part of Belgium has been steadily gaining support in recent elections and polls.  They're thriving on anti-immigration rhetorics.  They're pretty much the same people spouting conspiracy theories, climate denial and anti-science, and they are increasingly derailing facebook and twitter threads.  Not as bad as in the USA, but it's becoming increasingly worrysome.

Arctic sea ice / Re: PIOMAS vs CryoSat
« on: October 10, 2020, 07:48:49 PM »
That big red blob from Cryosat (+2m ice??) is not there in PIOMAS.
Cryosat issue?

I'm not sure.  That red blob looked quite strong in the 2020 melt season.  The ice there seemed compact and apparently it had a relatively late melt onset and weaker surface melting than other parts of the Arctic Ocean.  Compare e.g. with the average SMOS map for the peak insolation months June and July, here.

Or maybe some other factors are messing up the CryoSat results: snow depth, ice surface roughness and whatnot.

The CryoSat data file includes an uncertainty map, suggesting that the uncertainty of the sea ice thickness data is less than half a meter for most of the Arctic Ocean:

Arctic sea ice / Re: PIOMAS vs CryoSat
« on: October 10, 2020, 01:00:14 PM »
First results from CryoSat this Autumn:

(Image created by opening the nc file with panoply software.)

Arctic sea ice / Re: NSIDC 2020 Arctic SIE September average: August poll
« on: October 06, 2020, 07:13:20 PM »
There is a mistake on the total members that voted. Should be 62, not 59.

It's a bug in the forum software.  It means that 62 people have voted in this poll, but 3 of them have deleted their user account in the meantime.  Those votes are removed from the "Total members voted", but are still included in the individual bins.

Arctic sea ice / Re: Glossary ... for newbies and others
« on: October 03, 2020, 09:46:19 PM »
CAS = Central Arctic Sea.  Used by NSIDC.

Where does NSIDC call it "Central Arctic Sea"?  On their website they call it Central Arctic Basin (or briefly Central Arctic), just like everyone else. 

There are different ways to define the CAB (Wipneus uses a different definition than the NSIDC).  But I see no reason to add some non-existent "CAS" terminology that was made up by gerontocrat.  By that logic, you may as well rename Hudson Bay to Hudson Sea...

Arctic sea ice / Re: Accuracy of poll predictions
« on: October 01, 2020, 07:31:20 PM »
I've updated the graphics to include 2020.  The poll predictions were pretty accurate this year.

JAXA June polls:

NSIDC June polls:

NSIDC poll medians:

The forum / Re: PHPSESSID
« on: August 31, 2020, 08:58:40 PM »
Do you see PHPSESSID section in ASIF links?

It's probably because you disabled cookies.  If I disable third-party cookies in Chrome, then I also get the phpsessid query parameter.  It disappears when I enable cookies.

The rest / Re: Good music
« on: August 30, 2020, 04:43:37 PM »
Youssou N'Dour and Neneh Cherry - 7 Seconds (1994)

(Lyrics mostly in English, but some parts in French and Wolof)


Weezer - Island In The Sun (2001)

Arctic sea ice / Re: The 2020 melting season
« on: August 30, 2020, 04:32:13 PM »
SMOS remains lowest on record for the date, by a large margin.  This suggests there is still substantial surface melting going on.  Weather forecasts show colder air in the next few days, so presumably (hopefully) the curve will get back upward to more normal values for the time of year.

JAXA AMSR2 melt graphs show similar results: surface melting in August has been very strong this year.

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: August 30, 2020, 04:15:30 PM »
But I have a question.  In previous message (,2975.msg283986.html#msg283986 with remaining melt at the 10 year average, the 2020 minimum JAXA Extent table value is 3.70.  The chart in message above shows that by following the 10 year average melt, the 2020 minimum JAXA Extent bottoms out at about 3.775.  Such a small difference does not change the future of human civilization on the planet, but it just makes me wonder why the chart value does not match the table value.

This is to be expected.  The 10-year average trajectory is an unrealistically smooth curve.  Individual years fluctuate around that, with some peaks and troughs and with the minimum in one of the troughs.

The rest / Re: Good music
« on: August 23, 2020, 08:15:51 PM »
It seems that more than 90% of the content of this thread are oldies, which is overly repetitive in my opinion.  Anyway here are a couple more recent songs: 

Tame Impala - Let It Happen (2014)


The Raconteurs - Many Shades Of Black (2008)

Arctic sea ice / Re: "Smart" and "Stupid" Questions - Feel Free To Ask
« on: August 21, 2020, 11:21:09 PM »
I understand that the Sea Ice Extent is based on a there being 15% or less ice per square kilometer. This is a very conservative figure and I believe under estimates how low the Sea Ice Extents actually are.  Has anyone actually worked out the Extent using a higher figure, EG 30 or 40% ?

The graph below shows the NSIDC daily sea ice extent for yesterday (20th August) for the years 2005-2020 using 3 different threshold values to define extent  (blue: 15% concentration, red: 30%, green: 45%).

The forum / Re: Forum Decorum
« on: August 17, 2020, 10:23:59 PM »
Signal-to-noise ratio is quite high in the melting season thread.

I think it has improved recently, as usual at this time of year.  It's an annual pattern: in June and July the melt season thread is a mess, due to a very loud minority of folks who think every year that the September extent will be a new record low by a huge margin.  Next, at some point in July or August it becomes obvious that that isn't going to happen, and then those folks disappear or change their tone and the signal-to-noise ratio improves.

Arctic sea ice / Re: The 2020 melting season
« on: August 17, 2020, 10:08:39 PM »
I've long predicted this year would be exceptional, looks like it still will be.

(for me, that means being top 3 in at least 3 of the following 4 areas: extent, area, volume, eye test)

A top 3 rank for the September minimum is not exceptional at all.  In fact, most years were "exceptional" by your definition.  Five of the last 10 years had a top 3 minimum for both extent, area and volume at the time.  And 16 of the last 30 years.

The forum / Re: Forum Decorum
« on: August 16, 2020, 03:39:02 PM »
It's true that last year saw consistently more users, but the drop hasn't been what it seems

I agree that page views are an unreliable metric. 

But there are other indicators.  Participation in the sea ice minimum polls has plummeted this year.  The June and July polls had the lowest number of participants ever in the history of this forum, by far.  Participation in the August poll was the lowest since 2014.

Either it's because people are getting tired of those sea ice polls, or because there is general discontent with the forum.  Personally, I'm getting increasingly tired of the forum.  I think there are too many clowns on the forum and too many childish, nonsensical or hyperbolic comments.

Arctic sea ice / Re: The 2020 melting season
« on: August 15, 2020, 08:49:33 AM »
5-day pixel-wise median of the latest Bremen images:

3-day median:

Arctic sea ice / Re: The 2020 melting season
« on: August 09, 2020, 08:29:09 PM »
The image below shows the 5-day median (pixel-wise) of the latest Bremen images.  I think this kind of smoothing helps to reduce cloud interference and other artefacts in the daily Bremen images.   (Another member of this forum, petm, used to post similar images frequently last year, but I haven't seen him on the forum recently.)

And here is the 3-day (rather than 5-day) median:

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: August 09, 2020, 08:07:06 PM »
NSIDC daily area is currently 3.47 million km2, which is second lowest for the date. 

It is 0.33 million km2 above 2012, 0.10 million km2 below 2019, 0.21 below 2016, and 0.48 below the 2010s average. 

Area has been dropping slowly in the last few days, at an average rate of -0.02 million km2 per day during the last 5 days. 

The area anomaly (relative to the 1981-2010 average) is currently -1.90 million km2.

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: August 05, 2020, 09:49:29 PM »
NSIDC daily sea ice area: 2020 is lowest on record for the date, basically tied with 2012.

I created the graph using data from Wipneus' site.

Ranking of the NSIDC daily area values for 4th August:

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: August 02, 2020, 03:57:56 PM »

I made 2 analysis with the data from NSIDC provided by Wipneus. The first analysis looks to the changes from July 21st to July 30th. The true is that I choose the 21st just randomly, without a reason. The second analysis is from July 25th to July 30th. I think that July 25th was the beginning of the storm that had an impact on the ASI.

At first, after looking at the numbers, I did not know what to think. There is a bigger lost on area that it is on extent. I was expecting that. But the drops, in both analyses, are concentrated on the Central Arctic Basin, not on Chukchi and Beaufort.

On a 2nd thought, the drops on Chukchi and Beaufort were important when they are measure on percentage (2nd analysis). So, what I think is that the CAB had a hit north of Chukchi and Beaufort. I need to analyze the frontier of both seas versus the CAB, to see is that is true. But that is beyond the analysis that I am looking to do now.

I include the numbers of both analyses.

Maybe a picture could help to clarify this further:  here is a delta map, showing the change in NSIDC sea ice concentration over the past 7 days (i.e., 1 August minus 25 July 2020).  The strongest drops in concentration are in the area north of Chukchi and Western Beaufort where the cyclone was positioned.  As always, these could be real drops in sea ice concentration or transient weather effects affecting the surface wetness of the ice.

Arctic sea ice / Re: SMOS
« on: August 01, 2020, 11:07:59 PM »
Can you show maps after June 1? Thank you in advance.

Here is the average over the past two months, for 2020 (left) vs 2012 (right):


And here are the averages for the month of July only:

July 2020 average map: 
July 2012 average map:

Arctic sea ice / Re: SMOS
« on: August 01, 2020, 10:01:24 AM »
You were right, it was an interesting week.  Surface melting ramped up in the Central Arctic due to warm air advection.  And I guess the cyclone on the Pacific side brought a lot of rain.

Average for the past week for 2020 (left) vs 2012 (right):


Here is the difference map (2020 minus 2012):

And here is the anomaly map for 2020 relative to the 2010-2019 average:

To visualise this seasons BOE, i mean for the 2100 BOE when none of us will be around to see it according to the IPCC.

For your information, the latest IPCC assessment report (published 7 years ago) says

For high GHG emissions such as those corresponding to RCP8.5, a nearly ice-free Arctic Ocean (sea ice extent less than 1 × 106 km2 for at least 5 consecutive years) in September is likely before mid-century (medium confidence). This assessment is based on a subset of models that most closely reproduce the climatological mean state and 1979 to 2012 trend of Arctic sea ice cover.

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: July 30, 2020, 11:02:27 AM »
I have to check the numbers more carefully.
At a fast look, I get on July 28th.:  4.04M km2 for area, excluding lakes.
It is an interesting data set.

It should be 3.89 million km2 for 28 July 2020.  I'm not sure how you got 4.04 million km2?

Here are the single-day NSIDC area values for the last 10 days.  The table also shows the change from the previous day and the anomaly relative to the 1981-2010 average, in million km2.

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: July 29, 2020, 10:28:18 PM »
You can find the single-day NSIDC regional extent and area values on Wipneus' site.  I think this is the file you're looking for:
Thank you, Steven.
I will look at them.  :)

Glad it's useful. 

Maybe I should add that Wipneus uses a somewhat different definition of the regions than on the NSIDC website.  In particular, the CAB region as defined in Wipneus' data is about 38% larger than on the NSIDC website.

In addition, he has a spreadsheet with the daily regional NSIDC data for all previous days and years:

Arctic sea ice / Re: The 2020 melting season
« on: July 28, 2020, 10:10:06 PM »
I am going with the model produced by the Sanwa Elementary School...based on a wide 'U' profile through September and probably a new record or very near record minimum.

Sanwa Elementary School for the win!!! 

This is a perfect example of how difficult it is to predict the sea ice melting season. Those Elementary School kids are very likely to be closer to the final result than a lot of the “experts” when the final numbers are tallied.

That was the 2019 report, not 2020.  And Sanwa Elementary School actually did worse than everyone else last year (except Wang et al.) 

For 2020, they made a prediction of 4.4 million km2.

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: July 28, 2020, 07:35:04 PM »
Seems that they don't make public the single day value.  :(

You can find the single-day NSIDC regional extent and area values on Wipneus' site.  I think this is the file you're looking for:

Arctic sea ice / Re: 2020 Sea ice area and extent data
« on: July 26, 2020, 08:44:30 PM »
NSIDC daily sea ice area anomalies:

2020 is currently the lowest on record for the date.  We'll see whether it can stay below 2012 in the next week, as 2012 dropped very fast at the end of July and beginning of August.  The anomaly data are from Wipneus' site and have leap year corrections.

Arctic sea ice / Re: The 2020 melting season
« on: July 26, 2020, 10:37:10 AM »
Here is my current prediction for the sea ice min.

I'd love to see everyone's thoughts.

The region you indicated in that map is about 2.0 million km2 (based on pixel count).  I would be very surprised if the minimum extent is that low.

Arctic sea ice / Re: SMOS
« on: July 26, 2020, 10:27:56 AM »
The coming week is going to show interesting changes.

Maybe.  The Beaufort sea ice is starting to develop some weak spots, it could be interesting to see the impact of the forecasted cyclone in the next few days.  And the forecasts show warm air advection on the Atlantic side of the CAB and in the CAA.

But there doesn't seem to be a great area of ice on the brink of melting out completely right now.  In contrast, in 2012 there was a huge area of very vulnerable ice between the Beaufort and Laptev Seas (blue colors in the SMOS maps above).

Arctic sea ice / Re: SMOS
« on: July 25, 2020, 09:15:28 AM »
Another week has passed.  Here are the 7-day averages for 2020 (left) vs 2012 (right):


Here is the difference map, for 2020 minus 2012:

And here is the anomaly map for 2020 relative to the 2010-2019 average:

Take that graph with a grain of salt.  Thickness should not be calculated by dividing volume by NSIDC area.  It is known that NSIDC area substantially underestimates the real sea ice area, by about 10 to 25% in September.  So the "thickness" in that graph is an overestimate.

I have understood that area data was badly affected by melt ponds when insolation was high in early to mid-summer, and that produced the underestimates. But I also thought I had read that this effect diminished in the late Arctic summer, (starting around now) by which time melt ponds had drained, and as insolation quickly reduced in the high Arctic, melt ponds would no longer form to any great extent. So I always thought that by the minimum, the area data was less out of wack.
But I cannot find a study / science paper with the 10% to 25% figure. Can you point me to it?

So, as **Wipneus stated that the PIOMAS uses NSIDC Area data as input, for the time being for me  it's "Thickness = PIOMAS Volume divided by NSIDC Area".

If you had bothered to compare your thickness graph with the official PIOMAS thickness graph, you would have noticed that your numbers are systematically too high during Summer and Autumn, by about the percentages I posted.  Even as late as November there is still a discrepancy.

PIOMAS uses NSIDC concentration as some kind of replacement for melt ponds: PIOMAS doesn't model melt ponds explicitly, but only implicitly by assimilating the NSIDC concentration data.  But that does not mean that you can divide the volume by NSIDC area to get the average thickness.

NSIDC area is useful for making predictions about the minimum (Slater, Dekker, Tealight etc).  But if you want to know the real sea ice area, you're better off using the Hamburg AMSR2 sea ice area.  It's available on Wipneus' site.

NSIDC area underestimates the real sea ice area, not only in Summer but also in Autumn.  This was discussed several times in the Home Brew thread.  Last year I posted some graphs about this: see here.  Below is a similar graph, showing the ratio of NSIDC and AMSR2 area, averaged over the past 7 years.  Looking at individual years, the daily ratio in September ranges between 0.76 and 0.91.

average thickness chart posted by gerontocrat at,119.msg275579.html#msg275579  (A chart which I nominate for the ASI Graphical Hall of Fame).

Take that graph with a grain of salt.  Thickness should not be calculated by dividing volume by NSIDC area.  It is known that NSIDC area substantially underestimates the real sea ice area, by about 10 to 25% in September.  So the "thickness" in that graph is an overestimate.

The rest / Re: Good music
« on: July 21, 2020, 12:00:13 PM »
Somewhat related to the Arctic...

Within Temptation - Ice Queen (2000)  (symphonic metal song)


Pixies - Caribou (1987)  (alternative/rock)

Arctic sea ice / Re: Home brew AMSR2 extent & area calculation
« on: July 20, 2020, 06:22:33 PM »
I see that Wipneus has updated the graphs in the meantime (thanks, Wipneus).  The graphs seem to be in line with NSIDC area and SMOS, suggesting strong surface melting in the first week of July 2020, but a slowdown in the past two weeks:

Arctic sea ice / Re: SMOS
« on: July 18, 2020, 11:33:12 AM »
Average for the past week, for 2020 (left) vs 2012 (right):


The left image below shows the difference map: green = less melting in 2020 than in 2012, blue = more melting in 2020 than in 2012.  The right image shows the anomaly map for 2020 relative to the 2010-2019 average:


Arctic sea ice / Re: SMOS
« on: July 11, 2020, 10:55:43 AM »
Average for the first 10 days of July, for 2020 (left) vs. 2012 (right).  This suggests that recently there has been more surface melting than in 2012 on the Atlantic side of the CAB (including the north pole), but less on the Pacific side of the CAB.


Arctic sea ice / Re: Home brew AMSR2 extent & area calculation
« on: July 09, 2020, 08:15:45 PM »
Wipneus, If it's not too much trouble, could you update the JAXA AMSR2 melt graphs on your website?  Those graphs haven't been updated since August 2019.

Arctic sea ice / Re: The 2020 melting season
« on: July 09, 2020, 08:11:17 PM »
While extent continues to drop, area has actually stalled for the last 4 days. More draining melt ponds?

The JAXA AMSR2 melt graphics also show a reduction in surface melting in the last few days.  I guess the lack of warm air advection from lower latitudes is playing a role.

Arctic sea ice / Re: The 2020 melting season
« on: July 04, 2020, 10:15:31 PM »
It looks like this 'stirring' should result in a lot of of Fram export

I don't see much Fram export in that forecast.  Due to Coriolis force there's an angle of about 30 degrees between wind direction and sea ice drift, with the drift deviating toward the high pressure area, see the references here.  In the forecast posted above there won't be much ice crossing the boundary between CAB and Greenland Sea.

Arctic sea ice / Re: SMOS
« on: July 04, 2020, 09:32:18 PM »
I wrote some code to take weekly averages of the SMOS images

Can you average from the beginning of June?

Here is the average for the past 33 days for 2020 (left) vs. 2012 (right).  The large difference between those two years is mainly due to the fact that 2012 was way ahead in the first half of June.

Arctic sea ice / Re: SMOS
« on: July 04, 2020, 07:39:50 PM »
I wrote some code to take weekly averages of the SMOS images, to reduce the daily noise and fluctuations in those images. 

The left image below shows the average for the past week, 27 June to 3 July 2020.  For comparison, the right image shows the same period in 2012.  FWIW, this suggests that the Beaufort Sea and northern ESS had less surface melting than in 2012 during that period whereas the Canadian Archipelago and the Atlantic side of the CAB had more surface melting than in 2012.

Arctic sea ice / Re: SMOS
« on: July 01, 2020, 08:19:51 PM »
June monthly average:

The June monthly average has a good correlation with the September NSIDC sea ice extent.  The correlation coefficient is 0.86 over 2010-2019  (and even 0.95 if 2010 is excluded).  FWIW, this would give a prediction for the September 2020 NSIDC extent of 4.49 +/- 0.52 million km2 (95% prediction interval).

Here is a similar graph for the number of beige pixels in June, rather than the weighted pixel count.  In this case the correlation with the September NSIDC extent is 0.82 over the past 10 years, and the prediction for September 2020 extent would be 4.44 +/- 0.58 million km2.

June 2020 report from the Sea Ice Prediction Network:

For the Arctic, the median June Outlook for September 2020 sea-ice extent is 4.33 million square kilometers, with quartiles of 4.06 and 4.59 million square kilometers. To place this year's Outlook in context, the historical record September low was set in 2012 at 3.57 million square kilometers, and the second lowest record was 4.27 million square kilometers set in 2007. This year's projection is close to the 2019 June Outlook, which was 4.40 million square kilometers, and close to the observed September sea-ice extent of 4.32 million square kilometers. For September 2020, only two of the outlooks project sea-ice extent below the historical low record. Thus, there is a consensus judgement against a new record low September sea-ice extent.

Arctic sea ice / Re: SMOS
« on: June 20, 2020, 12:54:00 PM »
As the beige pixels are running out, it may be interesting to include information from the other pixel colors as well.  The graph below plots a weighted sum of all the sea ice pixels in the SMOS images, as discussed upthread last year:

There are only 65 votes in this poll.  That is the lowest number of participants ever, by far.

Number of people voting in the June JAXA polls on this forum:

2016:  142 people
2017:  123 people
2018:  121 people
2019:  112 people
2020:    65 people

Arctic sea ice / Re: SMOS
« on: May 16, 2020, 11:40:48 AM »
Just like last year, I will be running a pixel counting algorithm on the SMOS images.  The beige pixel counting graph will be updated daily:

Science / Re: Where are we now in CO2e , which pathway are we on?
« on: May 10, 2020, 12:33:20 PM »
Anyway, if there would be a formula (but as far as I can remember there is none) to add concentrations of greenhouse gases into a "global GHG CO2 equivalent" I would recalculate my spreadsheet and present it again. If not, I will keep it for myself from now on. The smart thing about this formula would be that the easily available concentrations of the different gases could be directly converted into one CO2 equivalent.
If there is anybody to help me out - you're welcome.

There are formulas in the NOAA link I posted:

see especially Table 1.  The CO2 equivalent concentration (in parts per million) can be calculated by using this formula:

CO2equivalent = 278 exp( RF / 5.35)   

where RF = radiative forcing.  For example, for the year 2018 that gives the following values for CO2 equivalent concentration:

• CO2 only:  407 ppm  (RF = 2.044)
• CO2 + CH4 + N2O:  465 ppm (RF = 2.755)
• CO2 + CH4 + N2O + halogenated gases:  496 ppm (RF = 3.101)

Consequences / Re: COVID-19
« on: May 09, 2020, 02:29:18 PM »
As of 5/6, according to NY DOH, NYC had 47 deaths. 47!!!! Down from 171 the week prior. My statement that NYC deaths are nearing zero is absolutely CORRECT, considering we are now at 47 a day, which is 72% down from the number a week prior.

From your link: "Due to delays in reporting, recent data are incomplete".

Using the history data of their github repository, this is how their data changed over the past 10 days:

Science / Re: Where are we now in CO2e , which pathway are we on?
« on: May 09, 2020, 11:23:10 AM »
I add the graph for the NOAA gases (20y and 100y CO2 equivalents) from 2000 to 2020.

That graph doesn't make sense.

I suspect you've been using GWP multipliers in your calculations.  But those can only be used in the context of emissions, not in the context of concentrations.  The GWP approach can not be used to calculate the CO2 equivalent for the entire atmospheric amount of CO2/methane/N2O etc.  The only meaningful way to do that is to calculate radiative forcing and use a logarithmic transform, e.g. see here.

There seems to be a lot of confusion about this topic on this forum.  I think the source of confusion is a poorly written wikipedia page that discusses two very different definitions of "CO2 equivalent": one definition for emissions and another one for concentrations.  Discussions on GWP and 20-year timescales etc are only relevant to the former definition but are not related to the latter definition.

Pages: [1] 2 3 ... 13