[DrCaleb]

N_Fiddledog N_Fiddledog:

Well, there's a curious thing about ice extent graphs. After the melt season is over and the lines have all been smoothed, you can look at the year in its entirety and get a different impression.

Another interesting thing about the graph is what's happening right now. Watch.

What's interesting is how you always seem to on about 'extents', but never 'volume' or 'multi year ice'. Here's what happens when attention is payed to that:



http://nsidc.org/arcticseaicenews/2015/ ... 5-maximum/


I also notice that 'Real' Climate only posts the Danish results for coastal ice. What do the result look like, for the same Danish study, but accounting for all the ice?



http://ocean.dmi.dk/arctic/icecover.uk.php

Oh, look at that average line, all up there all alone! Anyone out picking cherries lately?

N_Fiddledog N_Fiddledog:

Cool, eh? Highest in ten years right now. But it gets better. There's what's happening on Greenland.

$1:

Land ice has also grown at a record rate since September 1, with Greenland gaining about three billion tons of ice a day since September 1.

http://realclimatescience.com/2015/11/r ... continues/

But when compared to average, it really doesn't seem all that great, does it?

[N_Fiddledog]

Well, yeah, the sudden rise in ice is just a current thing, but I thought I made that point. It's only 'mildly interesting' is what I said.

You wanted something real to talk about concerning ice and I gave you something.

I do get the feeling both the Antarctic and the Arctic ice extents have been spending more and more time at intermittent times over the last few successive years in the grey zone though. Does that mean they're normalizing?

[N_Fiddledog]

Speaking of ice graphs they've always been a little different at each organization depending on methodology. My favorite used to be Nansen, but they don't seem to do it anymore.

Did you know Watts keeps a collection of all ice data put out from the different organizations here.

http://wattsupwiththat.com/reference-pa ... -ice-page/

[DrCaleb]

N_Fiddledog N_Fiddledog:

I do get the feeling both the Antarctic and the Arctic ice extents have been spending more and more time at intermittent times over the last few successive years in the grey zone though. Does that mean they're normalizing?

That's why temperatures are normally compared to 'pre-industrial' temperatures from 1850-1900, and ice extents are compared to the period 1900-2000. They have known averages, and accurate enough measurements to provide meaningful comparisons.

If we're still around in 2100, perhaps the new normal will be 2000-2100.

[N_Fiddledog]

DrCaleb DrCaleb:

N_Fiddledog N_Fiddledog:

ice extents are compared to the period 1900-2000. They have known averages, and accurate enough measurements to provide meaningful comparisons.

Hang on...say what?

Find somebody somewhere who agrees with that and can explain it.

If you're talking about the mean used to calculate anomaly graphs for temperature I'm pretty sure you're wrong about that one too.

[DrCaleb]

N_Fiddledog N_Fiddledog:

DrCaleb DrCaleb:

N_Fiddledog N_Fiddledog:

ice extents are compared to the period 1900-2000. They have known averages, and accurate enough measurements to provide meaningful comparisons.

Hang on...say what?

Find somebody somewhere who agrees with that and can explain it.

Why does somebody need to agree? The land temperature record (CRUTEM4) began in 1850, because that's when people started recording it. The data agrees, that's all that's needed.



http://www.metoffice.gov.uk/climate-gui ... mp-records

The same goes for ice extents. 1900 was when they started to be recorded.

$1:

Satellite data from the SMMR and SSM/I instruments have been combined with earlier observations from ice charts and other sources to yield a time series of Arctic ice extent from the early 1900s onward. While the pre-satellite records are not as reliable, their trends are in good general agreement with the satellite record and indicate that Arctic sea ice extent has been declining since at least the early 1950s.

https://nsidc.org/cryosphere/sotc/sea_ice.html

N_Fiddledog N_Fiddledog:

If you're talking about the mean used to calculate anomaly graphs for temperature I'm pretty sure you're wrong about that one too.

Nope. The data is the data. It begins when we started recording it.

As far as graphs go, anyone with an agenda can tell any story they like, if they choose the data shown.

[DrCaleb]

Another example:

$1:

While late 19th century temperatures are commonly taken to be indicative of pre-industrial, there is no fixed period that is used as standard and a variety of other periods have been used for observational and palaeo datasets. There are limitations in available data in the early instrumental record, making the average temperature in the reference period less certain. There is not a reliable indicator of global temperatures back to 1750, which is the era widely assumed to represent pre-industrial conditions. Therefore 1850-1900 is chosen here as the most reliable reference period, which also corresponds to the period chosen by IPCC to represent a suitable earlier reference period.

http://www.metoffice.gov.uk/news/releas ... one-degree

[N_Fiddledog]

As to the ice - NSIDC is not talking about any current anomaly graph you are likely to see.

There is only scattered data before the satellite era. All he's saying there is if you wanted to you could select some of those bits and make an extrapolation that would agree with the idea this current melt is anomalous.

The confusion I have over the land surface temperatures thing is I'm not a math guy, but I always see them arguing over what 30 year trend to use for the anomalies.

Like this -

$1:

3. Choosing an anomaly period.

The choice of the anomaly period is particularly important for reconstructions using CAM, as it will determine the amount of usable records. The anomaly period can also result in odd behavior of anomalies if it is too short, but in general the choice makes little difference to the results. In the figure that follows Mosher shows the difference between picking an anomaly period like CRU does, 1961-1990, and picking an anomaly period that maximizes the number monthly reports in a 30 year period. The period that maximizes the number of monthly reports over a 30 year period turns out to be 1952-1983. 1953-82 (Mosher). No other 30 year period in GHCN has more station reports. This refinement, however, has no appreciable impact.

http://wattsupwiththat.com/2010/07/13/c ... mperature/

Specifically when he says this:

$1:

In the figure that follows Mosher shows the difference between picking an anomaly period like CRU does, 1961-1990

What's he talking about? What is CRU doing? CRU would be the Climatic Research Unit at the University of East Anglia, wouldn't it?

As I say I don't actually understand the totality of how they calculate land surface temps for anomaly graphs I just know I hear them arguing about what 30 year trend to use.

(I used to hear it a lot in reference to whatever Hansen was doing at NASA GISS.)

With ice though the graphs you see are using the satellite period of thirty years or so.

[N_Fiddledog]

In fact I shouldn't have even allowed myself to be drawn off topic there to talk about land surface temps.

That was dumb of me.

Consider yourself right for now on that if you like.

Scratch those questions above concerning calculating land surface temps.

We're talking about ice for this one.

[DrCaleb]

N_Fiddledog N_Fiddledog:

As to the ice NSIDC is not talking about any current anomaly graph you are likely to see.

There is only scattered data before the satellite era. All he's saying there is if you wanted to you could gather up those bits and make an extrapolation that would agree with the idea this current melt is anomalous.

Yes, the data mostly comes from unreliable sources before satellites, but there are other sources that are taken into account. For example, when a local says the ice used to cover an area all year long, for his lifetime and his grandfather's; but now there is no ice - then it can be as reliable as the old ship captains who recorded extents in the early 1900's.

And the older data agrees with the newer, so it's considered 'good enough'.

N_Fiddledog N_Fiddledog:

The confusion I have over the temp thing is I'm not a math guy, but I always see them arguing over what 30 year trend to use for the anomalies.

Like this -

$1:

3. Choosing an anomaly period.

The choice of the anomaly period is particularly important for reconstructions using CAM, as it will determine the amount of usable records. The anomaly period can also result in odd behavior of anomalies if it is too short, but in general the choice makes little difference to the results. In the figure that follows Mosher shows the difference between picking an anomaly period like CRU does, 1961-1990, and picking an anomaly period that maximizes the number monthly reports in a 30 year period. The period that maximizes the number of monthly reports over a 30 year period turns out to be 1952-1983. 1953-82 (Mosher). No other 30 year period in GHCN has more station reports. This refinement, however, has no appreciable impact.

http://wattsupwiththat.com/2010/07/13/c ... mperature/

See, now I am a maths guy, so I see what they are trying every time! 'Picking a period' is just another way to get results you are looking for to fit a certain agenda. In Politics, it's called 'gerrymandering'. In Science, it's 'shilling'.

They use 'noisy' data to try to hide current trends in a statistical model. For example, if a tobacco company didn't want a trend in lung cancer to seem unusual, they'd pick the period 1915 - 1930 for their 'baseline' for comparison. As it happens, after WWI many men took up the smoking habit because cigarettes were included in standard 'care' packages, so all of a sudden the cases of lung cancer skyrocketed after WWI. But compare 1915-1930 to 1970-1985, and the surge of cases in the 70's disappears in the noise. Compare 1970-1985 to 1900-1915, and you'd see the surge.

The same as picking a period for temperatures. 1850 - 1900 is the time when we have the most accurate, and least 'noisy' data because urbanization and industrialization had not yet begun to affect temperatures. The years 1961-1990 would be much 'noisier' than 1850-1900, so any variations will be lost in the statistics. You'd use 1961-1990 to try to hide averages and means in data you don't want to seem as inaccurate, or to show weak trends that you don't want to call attention to.

N_Fiddledog N_Fiddledog:

Specifically when he says this:

$1:

In the figure that follows Mosher shows the difference between picking an anomaly period like CRU does, 1961-1990

A little further on he says:

$1:

This refinement, however, has no appreciable impact.

Which is a bold faced lie!

Another way to 'pick' the data is to have more data points than another. If I flip a coin 50 times, I have 50 data points and can get a rough average for heads v tails. If I flip it a million times, I have more data and I get a more precise and also a different number for heads v tails. Choosing a 30 year period with more readings most certainly affects the outcome!

Other ways to affect statistics and to 'pick' is to compare data that have different levels of error. Yes, I mean satellite vs land based temperature readings, or new vs old ocean surface temp readings. Comparing two sets with different rates of error means that the two results my overlap given the error rate. One set of average heights of third graders may be in metric, with a error rate of +/- 1cm, and another measured in British Imperial with an error rate of 3/8 of an inch. The two may overlap, just from the error rates! Comparing them may be meaningless.

N_Fiddledog N_Fiddledog:

What's he talking about? What is CRU doing? CRU would be the Climatic Research Unit at the University of East Anglia, wouldn't it?

As I say I don't actually understand the totality of how they calculate land surface temps for anomaly graphs I just know I hear them arguing about what 30 year trend to use.

With ice though the graphs you see are using the satellite period of thirty years or so.

I'm not sure what 'CRU' thy are referring to, but you may be right.

I wouldn't pretend to know how they calculate it all as well. I do know the task of getting averages for like data, converting them so the error rates all align and can then be compared in a meaningful way is daunting.

But what I do know is when comparing apples to hen's teeth, someone is shitting on my head and expecting to be thanked for the new hat.

N_Fiddledog N_Fiddledog:

In fact I shouldn't have even allowed myself to be drawn off topic there to talk about land surface temps.

That was dumb of me.

Consider yourself right for now on that if you like.

Scratch those questions above concerning calculating land surface temps.

We're talking about ice for this one.

It's fine, because everything I wrote applies to ice as well. The accurate data for ice extents is pretty recent. Comparing them gives a little perspective, but the bigger concern is volume. The data for volume is pretty sparse, until the satellites went up.

The lack of long term (100 year+) data doesn't really give much time for trends to be seen, but it does give us some very worrying short term trends. So what if Antarctic ice extent is increasing, if it's only an inch thick? It's all going to melt next summer anyhow. It's the multi-year ice disappearing, the ice shelves breaking off, and the possibility that the undersea currents that regulate climate that may already be changing that is the worry.

[N_Fiddledog]

DrCaleb DrCaleb:

It's fine, because everything I wrote applies to ice as well.

I don't know about that, but I'm pretty confident about this.

All the data you will see from different organizations on ice now when discussing this topic uses the 36 year satellite data and that's all.

Here's a collection. Show me one that doesn't.

http://wattsupwiththat.com/reference-pa ... -ice-page/

(As I understand it there are a lot of credibility problems and not much actual data when you get into what's actually available pre-satellite.)

[DrCaleb]

N_Fiddledog N_Fiddledog:

DrCaleb DrCaleb:

It's fine, because everything I wrote applies to ice as well.

I don't know about that, but I'm pretty confident about this.

Maths are maths, and statistical analysis is always the same whether you are discussing temperatures or manufacturing of widgets.

The same rules apply.

N_Fiddledog N_Fiddledog:

All the data you will see from different organizations on ice now when discussing this topic uses the 36 year satellite data and that's all.

Here's a collection. Show me one that doesn't.

http://wattsupwiththat.com/reference-pa ... -ice-page/

(As I understand it there are a lot of credibility problems and not much actual data when you get into what's actually available pre-satellite.)

Exactly. And 36 years when we are discussing the climate is a pretty low number. That's why I put more stock in people who have lived in the area for generations who say things are not good.

The data we have shows some disturbing trends, but we don't know what the long term consequences will be. Are we seeing a long term trend, or a repeating cycle? Our computer models contain the best knowledge we can come up with, but are they accurate enough? Those are the questions we grapple with. And going back to 'the grandfather test'; are we willing to subject our grandchildren to the low probability outcome that everything will be just fine? Or, is our knowledge woefully accurate, and predictions are more dire than they appear?

Like the study says, if rising oceans aren't coming from Antarctica, then they are coming from somewhere we don't yet understand. And when the rising glaciers of Antarctica do melt (as they have in the past) then we are really in a pickle.

[N_Fiddledog]

This article from the Washington Post is from 1922 not last week.



And here's one on scientific expeditions into the arctic during the 19th century when they found unusual large melts.

http://wattsupwiththat.com/2009/06/20/h ... rctic-ice/

Stuff like that is easy to find once you know it's there. There's much more.

But there's another current study on the Antarctic.

$1:

Missing factors in climate models suggest Antarctic melt may not be so bad after all


From MCGILL UNIVERSITY

Geophysics could slow Antarctic ice retreat
Gravitational effects, variations in Earth structure could damp rise in global sea levels

The anticipated melting of the massive West Antarctic Ice Sheet could be slowed by two big factors that are largely overlooked in current computer models, according to a new study.

The findings, published online in Nature Communications, suggest that the impact on global sea levels from the retreating ice sheet could be less drastic – or at least more gradual — than recent computer simulations have indicated.

Over the past year, numerous studies have warned that parts of the West Antarctic Ice Sheet are on the verge of a runaway retreat. Just last week a high-profile research paper forecast that this could lead eventually to a rise in global sea levels of as much as three metres.

The authors of the new Nature Communications paper, however, focus on two geophysical elements that they say aren’t adequately reflected in computer simulations for this region: the surprisingly powerful gravitational pull of the immense ice sheet on surrounding water, and the unusually fluid nature of the mantle beneath the bedrock that the ice sits on.

“The fate of the polar ice sheets in a warming world is a major concern for policy makers — and attention is rightly focused on the importance of restraining CO2 emissions and preparing for rising sea levels,” says lead author Natalya Gomez, an assistant professor of Earth and Planetary Sciences at McGill University in Montreal. “But our study shows that for Antarctica, in particular, computer models also need to take into account how gravitational effects and variations in Earth structure could affect the pace of future ice-sheet loss.”...

http://wattsupwiththat.com/2015/11/10/m ... after-all/

[BartSimpson]

DrCaleb DrCaleb:

Like the study says, if rising oceans aren't coming from Antarctica, then they are coming from somewhere we don't yet understand.

Or, Occam's Razor; they're not rising.

[DrCaleb]

N_Fiddledog N_Fiddledog:

This article from the Washington Post is from 1922 not last week.

And here's one on scientific expeditions into the arctic during the 19th century when they found unusual large melts.

http://wattsupwiththat.com/2009/06/20/h ... rctic-ice/

Stuff like that is easy to find once you know it's there. There's much more.

Indeed!



http://www.huffingtonpost.ca/2015/08/10 ... 68024.html

N_Fiddledog N_Fiddledog:

But there's another current study on the Antarctic.

$1:

Missing factors in climate models suggest Antarctic melt may not be so bad after all


From MCGILL UNIVERSITY

Geophysics could slow Antarctic ice retreat
Gravitational effects, variations in Earth structure could damp rise in global sea levels

The anticipated melting of the massive West Antarctic Ice Sheet could be slowed by two big factors that are largely overlooked in current computer models, according to a new study.

The findings, published online in Nature Communications, suggest that the impact on global sea levels from the retreating ice sheet could be less drastic – or at least more gradual — than recent computer simulations have indicated.

Over the past year, numerous studies have warned that parts of the West Antarctic Ice Sheet are on the verge of a runaway retreat. Just last week a high-profile research paper forecast that this could lead eventually to a rise in global sea levels of as much as three metres.

The authors of the new Nature Communications paper, however, focus on two geophysical elements that they say aren’t adequately reflected in computer simulations for this region: the surprisingly powerful gravitational pull of the immense ice sheet on surrounding water, and the unusually fluid nature of the mantle beneath the bedrock that the ice sits on.

“The fate of the polar ice sheets in a warming world is a major concern for policy makers — and attention is rightly focused on the importance of restraining CO2 emissions and preparing for rising sea levels,” says lead author Natalya Gomez, an assistant professor of Earth and Planetary Sciences at McGill University in Montreal. “But our study shows that for Antarctica, in particular, computer models also need to take into account how gravitational effects and variations in Earth structure could affect the pace of future ice-sheet loss.”...

http://wattsupwiththat.com/2015/11/10/m ... after-all/

Or they may be as bad. Or worse. That the problem, not knowing. But it's going to be cheaper to prevent the water rising now - or at least mitigate it, than to try to move all the people in all the cities that will be flooded when it does melt. Or rather, it'll be cheaper for your grandchildren for us to start now.