The extent of Arctic ice in the Barents Sea has shown a declining trend since the 1860s. A continuation of this trend would eventually result in an ice-free Arctic Ocean in summer. An ice-free Arctic Ocean would, in effect, mean the opening of a new territory, a region allowing for shipping through the Northwest Passage and the exploitation of new seabed resources. At the same time, the Arctic has attracted increasing political interest.
Projecting past trends into the future yield’s uncertain forecasts. It is like driving a car while looking in the rear-view mirror. The question is whether the extent of Arctic ice will continue its downward trend in the years ahead. The answer lies in understanding how Arctic ice is influenced by its surroundings.
Prevailing view
Arctic Ice East covers the area of sea ice in the Barents Sea, Norwegian Sea, and Kara Sea, bounded by 10°E, 80°N, and 70°E. Measurements of ice extent are based on April observations from 1864 to 1998 (Vinje, 2001). The extent of Arctic ice has been in decline since the mid-1860s.
The causes of Arctic ice reduction have been discussed for more than 100 years. Even in Nansen’s time, it was known that warm Atlantic water influences the extent of Arctic ice. Another explanation involves changes in the salt balance. Melting ice and freshwater runoff alter salinity, affecting how cold surface water sinks to the deep ocean. Some researchers suggest that pollution on the ice surface reduces sunlight reflection, leading to more melting. Over time, the prevailing view became that the reduction of Arctic ice is primarily due to human-induced global warming. This led to the assumption that the downward trend would continue and that the Arctic would eventually become ice-free in summer.
The signature of Arctic Ice East
Norwegian researcher Torgny Vinje conducted extensive work documenting Arctic ice extent. His studies were based on measurements, hunting logs, and observations from Arctic expeditions. I recall that, while on a research mission in the Arctic, he sent me two time series of Arctic ice extent by a simple e-mail, the result of many years of work. The first series covered the extent of Arctic Ice East of Svalbard from 1864–1998, while the second covered Arctic Ice West over the same period.
My own study of Arctic Ice East showed that the ice extent had a lunar-driven signature with periods of [1/2, 1, 4] × 18.6 years. This matched the NAW (North Atlantic Water) signature for the inflow of warm Atlantic water into the Norwegian Sea, confirming that the extent of Arctic Ice East can be traced to changes in Earth’s rotation. The difference was that the periodic changes in Arctic Ice East had a phase difference of π/2 radians.
To understand how this connection works, I analyzed the signature’s phase shift. The study showed that the phase is delayed by π/2 radians relative to Norwegian Sea temperature (NAW), indicating that the extent of Arctic Ice East is governed by the inflow rate of warm Atlantic water. Further analysis showed that the lunar-driven 4 × 18.6-year period reached an estimated maximum around 1955.
Figure 1 illustrates the relationship between the 74-year NAW period (blue) and the 74-year Arctic Ice East (AIE) period in the Barents Sea (green) for 1850–2050.
The ~74-year climate period can be traced from Earth’s axial nutation to tidal cycles and further to Norwegian Sea inflow (NAW) and ultimately to Arctic ice extent. Figure 1 shows normalized NAW (74 years, blue) and Arctic Ice East extent (AIE) in the Barents Sea (74 years, green) for 1850–2050. The 74-year period for AIE is estimated from measurements dating back to 1822.
The figure highlights the role of inertia in nature. The AIE extent lags NAW by π/2 radians, showing that AIE is current driven. The growth rate of AIE peaks in 1979 when NAW is at a minimum. The maximum rate of reduction in AIE occurs around 2015 when temperatures peak. The minimum AIE extent is expected around 2034, by which time NAW will be heading toward a colder period.
The ice edge position in the Barents Sea
In 2021, researchers at the Norwegian Polar Institute continued Vinje’s work and developed a time series for the April ice edge position in the Barents Sea from 1579–2021. Although much of the data before 1800 is missing, the long-term series can still be analyzed. The results confirmed that the ice edge position has a lunar-driven signature with periods of [1/2, 1, 4, 16] × 18.6 years. This means there is a ~297.76-year period (16 × 18.6 years) with maximum extent around 1867 and minimum extent around 2015. There is thus a resonance between the 4 × 18.6-year and 16 × 18.6-year periods. The 4 × 18.6-year period coincides with deep-water circulation in the Arctic Ocean, while the 16 × 18.6-year period corresponds to bottom-water circulation.
Figure 2 illustrates the relationship between the 74-year period (blue) and the 298-year period (green) for Arctic Ice East development from 1850–2050. The 298-year period represents the long-term trend in Arctic ice development, while the 74-year period drives positive and negative deviations within this long trend.
We can also see the outline of global climate change that began in the early 20th century. There was a temporary warming until the 1940s, reducing Arctic ice extent. Then followed cooling until 1980, increasing ice extent. After a short cold spell, a new warming of NAW occurred until 2015. Both cycles reached a minimum in 2015, resulting in the lowest extent of Arctic Ice East within the ~300-year cycle.
Scope of the study
This study was published in 2006 and has since been cited in about 100 scientific publications. It concluded that the extent of Arctic Ice East is driven by the inflow of warm Atlantic water, with a lunar-driven spectral signature. The implication of the factual measurements is that the extent of Arctic Ice East is likely about to reverse its trend. This means we can expect a gradual increase in Arctic Ice East extent, eventually reaching a maximum similar to that of the 1800s. Such a period of increased ice extent would have ripple effects on shipping, fisheries, the oil industry, and political assessments of Arctic regions.
Reference
- Yndestad H. (2006). The influence of the lunar nodal cycle on Arctic climate. ICES Journal. ICES Journal of Marine Science, 63: 401-420 (2006), doi:10.1016/j.icesjms.2005.07.015.
It has always baffled me why people think sea-ice in summer is an essential part of Nature’s plan for us.
In 1970, there was a prediction by a well known climatologist in Canada ( where I was living at the time) that by 2000 the temperature in the Arctic would have risen a full 20 F.* It caused quite a stir. We thought it was wonderful. All that worthless real estate in the North would become valuable! The trade of the world would pass though our waters. Never happened of course.
*I do not remember why. It was a regional forecast. Nothing to do with global warming.
Nigel,
I was going to defend my dr. philos. degree in the fall of 2004. At the same time, the Intergovernmental Panel on Climate Change (IPCC) published a new report on the development of the global and Arctic climate. A UN committee, I had never heard of, claimed that global temperature will rise by 5–7 degrees this century. The Arctic Ocean will be ice-free in the summer.
At the public defense, I was given the task of explaining how the information from the IPCC could affect the ecosystem in the Barents Sea in the years to come. The University of Copenhagen gave me access to a series of data showing the temperature development in Greenland in the period 550–1970. It turned out that the temperature in Greenland followed a distinct cycle of about 180 years. This pattern persisted throughout the 1500-year period. A continuation of this cycle suggested that we can expect a new cold climate period with a minimum around the year 2070. I presented this in my lecture during the public defense. I remember the examiner asking: “What about CO₂?”
Harald
Harald,
“What about carbon dioxide?” was put to you.
“As the concentration of this substance in the environment was stable for most of the 1,500 years, it did not seem worthwhile to consider the variations of a constant to be causal of…anything.” would be one way of answering.
Of course, the examiner was waiting for you to say, “My conclusions are, naturally, of little importance, compared to our imminent heat-death from carbon dioxide.”
There is a technical point. Although the passage of time can be an explanatory variable it can never be a causal variable.* Time series analysis therefore is only suggestive of, “something going on.” So, “time” is a proxy for hidden variables.** Another proper answer to the question (although probably not one appreciated by the examiner) would be that carbon dioxide is automatically “there” (latent) in a time series analysis if carbon dioxide itself has some temporal patterns. And, obversely, carbon dioxide may not be “there” in a time series analysis even though it DOES have a cause and effect relationship with something measured in the analysis.
*Every Christmas I get fat; the date EXPLAINS it, and could be used to forecast it, but greed CAUSES it. How ridiculous would it be to say, “Here comes December 25, it will cause Nigel to become fat even though he is not eating this year.”
**In a sense, ALL variables, and therefore, equally, in a sense, NO variables
Nigel