Germans Prof. Fritz Vahrenholt and Frank Bosse took a look at the July 2014 Arctic sea ice data. They found astonishing results. What follows is the Arctic sea ice part of the story I reported on here.
The Sun In July 2014 And Arctic Sea Ice In Mid Summer
By Frank Bosse and Prof. Fritz Vahrenholt
(Translated/edited by P Gosselin)
Every summer over the past years Arctic sea ice has been the subject of much public interest. Especially after 2006 there have been years (recall 2007 and 2012) when a large part of the Arctic ice cap melted away and all the talk was about the “Arctic death spiral“.
Since 1979 we have been collecting really good, validated and homogenized data from satellite observations that look at two important magnitudes: 1) the sea ice cover area in square kilometers, and 2) the so-called ice extent in square kilometers², i.e. that is at least 15% ice covered. The linear trend of this is indeed downward since observations began:
Figure 4: July sea ice extent since 1979.
The excitement at the start of the melt season in May/June is always big. How many millions of square kilometers will still be left in September? Estimating this value is no easy task because upward deviations from the trend line have occurred time and again (like in the 1990s and early 2000s), and downward deviations (2007, 2011, 2012).
But last year and the year before the Arctic sea ice ended up close to the trend line. Every year the Sea Ice Prediction Network conducts a survey of educated guesses of sea ice extent expected by mid-September. This is what it looks like this year:
Figure 5: Forecast for September 2014 sea ice extent, from various sources and methods. Source: arcus.com.
For comparison, last year’s (2013) figure of 5.4 million km² is added in the chart. The mean value of this year’s estimates made in June is 4.8 million km², i.e. far above 2012’s record low value of 3.6 million km².
Also a co-author of this article participated; his forecast made in early July was 5.0 million km². He used the heat content of the Arctic basin and the existing ice volume at the end of the winter. There are many decisive factors that are impossible to know so early on, e.g. the weather (foremost the currents from warmer land and wind), and other natural factors.
We already know that every prediction is difficult, especially those dealing with the future. Having observed the development so far until the end of July, it is highly probable that we will see a repeat of 2013, which would mean that the mean of all those submitting a forecast will be approx. 0.6 million km² too low.
Only 5100 km³ melted in July 2014
The sea ice could turn out to be more robust than previously assumed. That has to do with a magnitude that is not visible in satellite images: the thickness of the ice – or its volume. Here we are left to rely on models. One that is often used is PIOMAS. It provides daily data on ice volume and the latest July value yields the following chart over the years:
Figure 6: Ice volume (1979 = 100%) as to PIOMAS for July 31 of each year, eight-year smoothed non-linear trend.
July is a very important month during the melt season: In the Arctic the sun shines 24 hours per day, irradiance there exceeds that of the tropics as a result. The sun is also high above the horizon and so only about 15% of the sun’s radiation in the water is reflected. The open water absorbs large quantities of energy, which can melt the ice from the side and from underneath. And this year something completely unusual happened: while since 1979 approx. 6500 km³ melted in July, this year only 5100 km³ melted.
Figure 7: July ice volume loss plot since 1979. Shown are also the limits depicting 2 standard deviations (bold black lines). Source: psc.apl.washington.edu/arctic-sea-ice-volume-anomaly/data/.
The very low melt in 2014 is a 3-sigma event!
A negative record – since 1979 – that is unparalleled. The relatively steady decline shown in Figures 4 and 6 is deceptive – the ice of the current year is decisively determined by the previous year. To a great extent the series is auto-correlated. One sees the true development much better with the Figure 7 loss diagram.
Something seems to have happened during the peak of summer to lead to a stopping of the downward trend. It was internal natural variability within our climate – it certainly was not the weather. All experts are currently baffled. This year the weather resembles more that of the year 2012, and back then there was a low point in ice volume development.
Over the last years natural variability has been increasingly used in climate science to explain away over-hasty predictions of greenhouse effects. Also the observations of the Arctic ice are adding to the scramble for explanations. If you are interested how the Arctic sea ice will finish in September – then stay tuned!