The big Arctic Sea-Ice shift of 2007: Ice refuses to melt

by Javier

I have maintained since 2015 that in the 2006-2007 season the Arctic underwent a cyclical phase shift, and the rapid sea-ice melting observed over the previous decades ended. A few scientists predicted or explained this shift based on their study of multi-decadal oscillations (see bibliography). They were ignored by mainstream climatology and the press because the “anthropogenic” melting of the Arctic is one of the main selling points of the climate scare. See for example:

A devastating Arctic temperature rise that could submerge coastal cities and trigger species extinction is now locked in. Business Insider March 15, 2019

Year after year the data supports my view over the desperate scaremongers like Tamino. With the passing of time it is more and more difficult to defend the idea that Arctic melting is continuing, so alarmists keep changing the metric. First it was September sea-ice extent (SIE), then September sea-ice volume, and now annual average SIE. However, the reference measurements are September minimum SIE and March maximum SIE.

This article is more than a biannual update on the Arctic ice situation, as I will focus specifically on showing evidence for the trend change that took place in 2007. As 12 years have passed since the shift, the best way is to compare the 2007-2019 period with the previous 1994-2006 period of equal length to display the striking differences between both periods.

Figure 1

Figure 1. Changes in September SIE for both periods as a percentage change over the first year of the period.

The supposed dangers of an ice-free Arctic appear to be decreasing. While the first period showed a September (minimum) SIE loss of 20%, the second has seen a gain of 10%.

Figure 2

Figure 2. Changes in March SIE for both periods as a percentage change over the first year of the period.

March (maximum) SIE still shows a decreasing trend, although it is so small as to be negligible. While the first period showed a March SIE loss of 8%, the second period displays the same March SIE in 2019 as 12 years before.

Figure 3

Figure 3. Seasonal SIE melting from March to September

Nowhere is the 2007 Arctic shift better seen than in SIE melting. This is the sea-ice surface melted every year from the March maximum to the September minimum. The 2007 melting season saw a jump from ~ 9 million km^2 to ~ 10.5 million km^2 melted, but accompanying this huge jump in melted surface came a trend inversion, so the surface melted has been decreasing since then. It will be interesting to see what happens to the annual melted surface over the next few years

Figure 4

Figure 4. Changes in atmospheric CO₂ for both periods as a percentage change over the first year of the period.

We have been told repeatedly that our emissions are responsible for the melting of the Arctic. There is a problem with this hypothesis. Despite completely different melting profiles, both periods display the same percentage increase in CO₂. Changes in CO₂ levels do not explain the differences in sea-ice behavior for the two halves of the last 26 years. And this is also a problem because we have been told repeatedly that by reducing our CO₂ emissions we can save the Arctic. Yet Arctic sea-ice is unlikely to respond to changes in our CO₂ emissions given its lack of response to consistently increasing CO₂ levels.

Figure 5

Figure 5. Changes in 80-90°N annual average temperature from ERA interim reanalysis.

The most common scientific explanation for Arctic sea-ice melting is the extraordinary warming taking place at high latitudes due to Arctic amplification. It is described as a positive feedback where the decrease in ice and snow cover reduces planetary albedo, thus increasing radiative warming. This, together with increased heat transport from lower latitudes drives further reductions in ice cover. This explanation is problematic, as Arctic amplification has continued unabated in the absence of SIE reduction and without producing further SIE reduction. The decrease in ice albedo from losing 40% of the September sea-ice cover between 1994 and 2007 has been unable to drive further loses since then, refuting the sea-ice “death spiral” proposed by Mark Serreze, director of the National Snow and Ice Data Center.

The climatic factor that explains sea-ice behavior and was the basis for the only correct prediction of the 2006-2007 Arctic shift is internal variability: The existence in the climate system of multi-decadal oceanic-atmospheric oscillations. The problem for the climate alarmists with this explanation is that if it explains why the ice is not melting now, it also explains in great part why it was melting before, greatly reducing the possible anthropogenic contribution. Another problem for them is that these oscillations are not part of the general circulation models, because their origin is unknown. Thus, making the general circulation models essentially useless to project changes. And these are all strong talking points for climate skeptics.


Årthun, M., et al. 2017. “Skillful prediction of northern climate provided by the ocean.” Nature Communications, 8, ncomms15875.

Divine, D.V. and Dick, C., 2006. Historical variability of sea ice edge position in the Nordic Seas. Journal of Geophysical Research: Oceans, 111 (C1).

Miles, M.W., et al. 2014. “A signal of persistent Atlantic multidecadal variability in Arctic sea ice.” Geophys. Res. Lett., 41, 463–469.

Wyatt, M.G. and Curry, J.A., 2014. Role for Eurasian Arctic shelf sea ice in a secularly varying hemispheric climate signal during the 20th century. Climate dynamics, 42 (9-10), pp.2763-2782.


Arctic Sea Ice extent

Temperature reanalysis



Evidence that multidecadal Arctic sea ice has turned the corner

Arctic melt season changes and the Arctic regime shift