Arctic sea ice minimum early prediction

[Sciguy]

I am obliged to remind you that anthropogenic aerosols (air pollution/SO2) is currently cooling the arctic between 2 and 4C and preserving over 30% of total annual sea ice loss.

What is the source of this information?  It appears to be a gross overestimate of the amount of cooling from aerosols.

This paper from 2015 shows that the cooling effects of aerosols to be between 1.3 and 2.2 degrees C of cooling: http://www.nature.com/articles/518140e

Aerosols have a cooling effect by reflecting sunlight back into space. Mohammad Reza Najafi at the University of Victoria in Canada and his colleagues analysed nine climate models running from 1913 to 2012, comparing simulations with and without greenhouse gases, aerosols and other climate drivers. Their results show that aerosols have offset 1.3–2.2 °C of Arctic warming from greenhouse gases, limiting the observed warming to 1.2 °C. With aerosol emissions projected to drop in the coming decades, the rate of the warming is likely to increase.

And this paper from 2017 states that the projected reductions in anthropogenic aerosols will lead to a warming in the Arctic of 0.7 degrees C: https://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-17-0287.1

Observations show that the Arctic sea ice cover has been shrinking at an unprecedented rate since the 1970s. Even though the accumulation of greenhouse gases in the atmosphere has been closely linked with the loss of Arctic sea ice, the role of atmospheric aerosols in past and future Arctic climate change remains elusive. Using a state-of-the-art fully coupled climate model, the authors assess the equilibrium responses of the Arctic sea ice to the different aerosol emission scenarios and investigate the pathways by which aerosols impose their influence in the Arctic. These sensitivity experiments show that the impacts of aerosol perturbations on the pace of sea ice melt effectively modulate the ocean circulation and atmospheric feedbacks. Because of the contrasting evolutions of particulate pollution in the developed and developing countries since the 1970s, the opposite aerosol forcings from different midlatitude regions are nearly canceled out in the Arctic during the boreal summer, resulting in a muted aerosol effect on the recent sea ice changes. Consequently, the greenhouse forcing alone can largely explain the observed Arctic sea ice loss up to the present. In the next few decades, the projected alleviation of particulate pollution in the Northern Hemisphere can contribute up to 20% of the total Arctic sea ice loss and 0.7°C surface warming over the Arctic. The authors’ model simulations further show that aerosol microphysical effects on the Arctic clouds are the major component in the total aerosol radiative forcing over the Arctic. Compared to the aerosol-induced energy imbalance in lower latitudes outside the Arctic, the local radiative forcing by aerosol variations within the Arctic, due to either local emissions or long-range transports, is more efficient in determining the sea ice changes and Arctic climate change.

[Steven]

Here's a graph showing (for 2018) the standard error of regression, when relating each day's extent to the ensuing minimum, based on data from 2007-2017:

Here is a similar graph, using sea ice area on a given day as a predictor of the extent minimum.  As above, the graph shows the standard error of the regression based on data for 2007-2017:



Between about day 140 and day 220 (i.e., from late May to mid-August), the values in this graph are substantially lower than in your graph above.  So sea ice area is a better predictor than extent during that period (based on 2007-2017 data).

Source for the sea ice area data: https://sites.google.com/site/arctischepinguin/home/sea-ice-extent-area/data/nsidc_arc_nt_main.txt.

[RoxTheGeologist]

i use a careful approach to my statistical analysis. I look at where the majority of the forum have voted, and then choose one bin less pessimistic.

[magnamentis]

i use a careful approach to my statistical analysis. I look at where the majority of the forum have voted, and then choose one bin less pessimistic.

but that means your purpose is not to asses sea-ice development but where you stand compared to others, means where is the biggest chance to be right against others and not about ice?

are you aware that the this way of thinking is part of the reason why mankind made it to current state, in terms of the good and the bad?

i think the only real purpose of such a poll in this kind of forum should be to build ourselves an opinion based on information and explain it eventually so that we can eventually find out more about how things play together and understand the system better, so that we humans can ultimately and hopefully take the most efficient measures to fight the problem which is not only a problem of global warming and sea-ice but is related to the problem how we humans live and especially what our priorities are, sustainability or one of the other things like profit, fun, victory,righteousness, comfort etc.

[RoxTheGeologist]

I forget that a British sense of humor can be lost internationally. I was poking gentle fun at the trend on this forum to choose a more pessimistic outcome.

I actually vote what I think. Typically what would be the bin which is the average of the last 5 years. That tends to be the bin slightly more pessimistic than the rest of the forum.

On the current situation, There is no means from the data except to look at the general trend, and hope we aren't in 2012. Now if this was an El Nino year, I might have gone for the top bin. 

[Richard Rathbone]

that would only be true if the likelihood of achieving large (>2.5km2) melt in August at the beginning of the time period was the same as today.  Since we are in a dynamic system currently experiencing great changes this is not accurate.

Do you have any evidence for this? The analyses I've seen for melt rates show its pretty much the same across the last 40 years in August. (and every other month except May/June)

[DavidR]

that would only be true if the likelihood of achieving large (>2.5km2) melt in August at the beginning of the time period was the same as today.  Since we are in a dynamic system currently experiencing great changes this is not accurate.

Do you have any evidence for this? The analyses I've seen for melt rates show its pretty much the same across the last 40 years in August. (and every other month except May/June)

The attached image shows August extent  loss over the past 30 years indicating an average increase from < 1.5 to  > 2.0  million square kilometers. The maximum and minimum variations from the trend appear to  be consistent ranging  from -300 K to  almost  +800 K.  All of the years setting records have done it with big losses in August.  The size of those big losses does seem to  be increasing.  There can be little doubt that the late July  / August  weather is critical in determining the loss for the year.

[Ned W]

Here's a graph showing (for 2018) the standard error of regression, when relating each day's extent to the ensuing minimum, based on data from 2007-2017:

Here is a similar graph, using sea ice area on a given day as a predictor of the extent minimum.  As above, the graph shows the standard error of the regression based on data for 2007-2017:



Between about day 140 and day 220 (i.e., from late May to mid-August), the values in this graph are substantially lower than in your graph above.  So sea ice area is a better predictor than extent during that period (based on 2007-2017 data).

Source for the sea ice area data: https://sites.google.com/site/arctischepinguin/home/sea-ice-extent-area/data/nsidc_arc_nt_main.txt.

Hey, that is very cool.  Here are both (area and extent) on the same graph:



This is the standard error for estimating the end-of-season JAXA daily extent minimum, based on each day's JAXA extent or NSIDC area.

I'm not sure what to make of those dips and spikes in June/July in the area data.

[Steven]

I'm not sure what to make of those dips and spikes in June/July in the area data.

Upthread I had smoothed the sea ice area data somewhat, by using 5-day running averages.  But even then, the fluctuations are fairly large.

The NSIDC sea ice area is quite sensitive to melt ponds.  So it's more variable than extent.  If weather conditions change, then the melt pond fraction could change relatively quickly.

It's probably the main reason why area is a better predictor than extent: melt ponding in June and July preconditions the ice for extent losses later in the melt season.

Another factor is that only 11 years of data were used in the regression analysis (2007-2017).  It's a small sample size.  Doing the same analysis for 1979-2017, then the curve decreases more smoothly, and area is still a better predictor than extent during early summer, although by a smaller margin.

[magnamentis]

the gaps between more and more smaller ice floes are often too small to be detected by sensors and wind and ice drift can change that in the way that one they the gaps are larger and then the ice is more compact the other day. that could at least be one possible explanation for frequent ups and downs in area.

i think that one of the higher resolutions is around 3-4 kilometers, just imagene gapps smaller than that, they go un-detected. and this BTW is why i expect a very sudden drop in one of the coming yers when the state of the ice will see to some kind of more rapidly developing larger open water gapps within the still more than 15% ice covered areas.

[Stephan]

I processed the arctic sea ice extent data by calculating the decrease from monthly averages for all years between 1979 and present. As the monthly averages are probably mid month values, these differences represent what is going on between the 15th of the previous month and the 15th of the actual month. These differences (on x-axis) are now plotted against the yearly JAXA minimum extent (y-axis) for all months between April and September for all years since 1979. The results are interesting:
April shows random values (see April.jpg)
May isn't better (see Mai.jpg)
June has a slight correlation (see Juni.jpg)
July correlates quite well (see Juli.jpg)
August and September follow soon.

[Stephan]

August has a weak correlation (see below)
September also correlates quite a little bit - even now it is definitively too late for an early prediction.
https://forum.arctic-sea-ice.net/Smileys/default/wink.gif

Therefore in my opinion the "most decisive time" for the level of the yearly minimum is the second half of June and July.
If anyone has the data by hand of beginning and end of each month a comparable month-by-month evaluation may be easily possible.

[jai mitchell]

I am obliged to remind you that anthropogenic aerosols (air pollution/SO2) is currently cooling the arctic between 2 and 4C and preserving over 30% of total annual sea ice loss.

What is the source of this information?  It appears to be a gross overestimate of the amount of cooling from aerosols.

This paper from 2015 shows that the cooling effects of aerosols to be between 1.3 and 2.2 degrees C of cooling: http://www.nature.com/articles/518140e

Aerosols have a cooling effect by reflecting sunlight back into space. Mohammad Reza Najafi at the University of Victoria in Canada and his colleagues analysed nine climate models running from 1913 to 2012, comparing simulations with and without greenhouse gases, aerosols and other climate drivers. Their results show that aerosols have offset 1.3–2.2 °C of Arctic warming from greenhouse gases, limiting the observed warming to 1.2 °C. With aerosol emissions projected to drop in the coming decades, the rate of the warming is likely to increase.

And this paper from 2017 states that the projected reductions in anthropogenic aerosols will lead to a warming in the Arctic of 0.7 degrees C: https://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-17-0287.1

Observations show that the Arctic sea ice cover has been shrinking at an unprecedented rate since the 1970s. Even though the accumulation of greenhouse gases in the atmosphere has been closely linked with the loss of Arctic sea ice, the role of atmospheric aerosols in past and future Arctic climate change remains elusive. Using a state-of-the-art fully coupled climate model, the authors assess the equilibrium responses of the Arctic sea ice to the different aerosol emission scenarios and investigate the pathways by which aerosols impose their influence in the Arctic. These sensitivity experiments show that the impacts of aerosol perturbations on the pace of sea ice melt effectively modulate the ocean circulation and atmospheric feedbacks. Because of the contrasting evolutions of particulate pollution in the developed and developing countries since the 1970s, the opposite aerosol forcings from different midlatitude regions are nearly canceled out in the Arctic during the boreal summer, resulting in a muted aerosol effect on the recent sea ice changes. Consequently, the greenhouse forcing alone can largely explain the observed Arctic sea ice loss up to the present. In the next few decades, the projected alleviation of particulate pollution in the Northern Hemisphere can contribute up to 20% of the total Arctic sea ice loss and 0.7°C surface warming over the Arctic. The authors’ model simulations further show that aerosol microphysical effects on the Arctic clouds are the major component in the total aerosol radiative forcing over the Arctic. Compared to the aerosol-induced energy imbalance in lower latitudes outside the Arctic, the local radiative forcing by aerosol variations within the Arctic, due to either local emissions or long-range transports, is more efficient in determining the sea ice changes and Arctic climate change.

Ken,
See my response here
https://forum.arctic-sea-ice.net/index.php/topic,1384.msg157759.html#msg157759

[Feeltheburn]

Had I voted I would have guessed around 4th to 7th lowest. It is not likely to be a record this summer, and doubtful second lowest either.