Image: The Motion of the Ocean
Guest Post by Wim Röst
Five million years ago, average temperatures were higher than they are now. During the Pliocene, the era just before the period of the Quaternary Ice Ages, ‘glacials’ did not yet exist because temperatures were too high. As cooling of the deep seas continued, temperatures became that low that large surfaces of the Northern Hemisphere became covered with snow. The earth’s albedo grew fast and large ice sheets started to develop. Only short warm interglacials separated the glacials. The emergence of the interglacials first showed a 41,000-year period (as proposed by Milankovitch) and in the last part of the Quaternary they showed a 100,000-year pattern. A difference that so far is not well understood. Here it is suggested that the continued cooling of deep sea temperatures is the cause of that diminished frequency of interglacials. Colder deep-sea temperatures resulted in lower sea surface temperatures that lowered the atmospheric temperatures. The general background temperature of the Earth became lower and lower, changing climate processes like the glacial – interglacial rhythm. As oceans cooled, atmospheric temperatures lowered and more energy was needed to get out of the glacial state. The extra energy came from combined favourable orbital circumstances, which only happens roughly once in 100,000 years.
5 Million years of ever lower temperatures
As figure 1 shows, during the last 5 million years, deep sea temperatures are falling. This cooling does not seem to be spectacular, deep sea temperatures are going down from an average of plus 2 degrees Celsius to minus 0.25 degrees Celsius, but – as will be argued – this lowering is of the utmost importance for the development of the Earth’s climate. At certain times the lowering of deep sea temperatures is important, even when the lowering is only fractional.
Figure 1: Falling deep ocean temperatures from Pliocene (to the left) into the Pleistocene (to the right in the figure). Time from left to right, in millions of years.
As shown in previous posts, the deep sea is directly connected with the sea surface by a process called ‘ocean upwelling’ sometimes shortened to simply ‘upwelling’. The ever colder deep ocean waters are welling up into the ocean surface layer in large quantities (more than a million cubic kilometres every year). This is a relatively slow process where the cold upwelling waters are warmed by the sun.
But, the lower the starting temperature of the upwelling waters, the colder the surface layer will be. The deep sea cooled more than two degrees Celsius during this period and therefore the sea surface has also cooled.
The world ocean surface comprises 71% of the Earth’s surface and it is generally accepted that the surface (air) temperatures at sea adapt to the temperature of the underlying sea surface water. Therefore, as sea surface waters cool, the atmosphere above 71% of the Earth is cooling. Colder currents will flow to the poles and colder air will be transported to poles and continents, diminishing the warming of those surfaces too. Convection will transport less and colder air upwards. In this way, the colder deep-sea temperatures end up not only in lower sea surface temperatures but also in a colder atmosphere – all other things remaining the same.
Figure 2: Estimate of global surface temperatures from the Pliocene into the Pleistocene, in degrees Celsius. In this figure, we see the same trend in figure 1.
A Deep Sea / Surface temperature Amplifier
It is interesting to see that a two degree C drop in deep sea temperature (figure 1) ends up as a 5 degree C lower surface temperature as shown in figure 3. This is a drop from 17 to 12 degrees Celsius. In this period, we see a large ‘amplification factor’ of around 2.5. A deep-sea temperature that is 0.2°C lower/higher, corresponds with a 0.5°C lower/higher surface temperature. As we shall see, the existence of this ´deep sea / surface temperature amplifier´ is important.
The ‘Earth’s General Background Temperature’
All climate processes on earth are taking place in a setting of a certain background temperature. As argued here, that general background temperature is set by the deep oceans connected with the surface layer that is connected with the atmosphere. The colder deep ocean is the cause of a colder atmosphere. Fluctuations (seasonal, annual, decadal, multidecadal, centennial, millennial) all occur against this ‘background temperature’ of the deep ocean.
The warm deep oceans fifty million years ago had an average temperature of more than 12°C (see figure 3). Those warm oceans created a completely different background temperature than our present deep oceans do. The present average temperature of all our ocean water (inclusive the warm surface layer) is only 3.9°C, the deep oceans themselves are several degrees colder. A difference of around 10°C. Therefore, our present ‘general background temperature’ is very low. Our cold oceans are even allowing glacial periods – that wouldn’t have occurred when the oceans were warmer. Our cold oceans brought us, or perhaps allowed us to have our very cold Pleistocene era. Figure 3.
(Remaining question: what made sea temperatures ending that many degrees lower after 55 million years? More about a possible / probable answer: in future posts)
Figure 3: Estimated deep ocean temperature in the last 65 million years by James Hansen et. al. 2013 Deep sea temperatures were highest 55 million years ago. In the last fifteen million years there is a nearly continuous downward trend.
From here, it is but a small step to find the solution for the 41,000 – 100,000-year problem.
The 41,000 – 100,000-year interglacial problem
During the first period of the Pleistocene interglacials, there was a 41,000-year glacial/interglacial cycle but during the last million years there was only a warmer period once every 100,000 years. See figure 4.
Figure 4: Temperature development in the last five million years according to the Antarctic Vostok Ice Core. The green lines show the 41,000 and the 100,000-year periods in the Pleistocene. The left side of the graph is the warmer Pliocene, the period that was still too warm to permit ice ages.
Milankovitch’ cycles played the dominant role in taking the Earth out of the glacial state. The glacial state is the normal state in the Pleistocene. Eight or nine of every 10 years in the Pleistocene were ‘glacial years’. Very cold. With rough and very changeable weather and climates, as is shown by the high variance in temperature – reflecting frequent and turbulent climate changes.
Javier explains the change in the frequency of interglacials as follows: “The 100 kyr problem is solved because there is no 100 kyr cycle, just a 41 kyr cycle that skips one or two beats.” Italics added.
The question remains: And what causes the skipping of one or two beats?
The answer is: it is the ever lower deep ocean temperature that is translated into ever lower atmospheric temperatures that makes it more difficult to come out of that ever more dominating glacial state. Renee Hannon recently: “The past four glacial cycles are increasing in duration from 89 kyrs to 119 kyrs.”
In the end of the period, because of the extreme cold of the deep sea, only the most favourable (combined) orbital conditions allow a glacial to enter the warmer interglacial state, which has more stable temperatures.
The difference between ‘snow’ and ‘water’ might be only one or two tenths of a degree Celsius. A temperature of + 0.1 °C means ‘melt’ and ‘rain’. A temperature of – 0.1 °C means ‘snow’ and ‘ice’.
The above-mentioned amplification factor comes into mind. Deep sea temperatures relate to (surface) air temperatures but with an amplification factor of around 2.5 for surface air. A 0.2 °C lower deep sea temperature is translated into a half degree Celsius lower atmospheric temperature. Therefore, even a difference of less than one tenth of a degree of the temperature of the deep sea can make a substantial difference in the presence of ice and snow over large Northern Hemisphere land areas. Areas that are covered with ice and snow have a much higher albedo. A rising albedo will further cool the Earth.
In this way, at a certain point, a fractional lowering of deep sea temperatures results in enhanced lowering of the Earth surface temperatures. First, because of the deep sea / surface amplification factor, and second, because of the additional ice and snow albedo amplification.
Once more the development of deep sea temperatures: figure 5.
Figure 5: Glacials and falling deep ocean temperatures from Pliocene into the Pleistocene. Glacials developed from a certain low deep ocean temperature. As cooling continued, interglacials switched their cycle from once per 41,000 years to once per 100,000 years. Added in the figure: squares and the corresponding periods below in the figure.
At the start of the Pleistocene, every obliquity cycle resulted in an interglacial. But later in the period the warming effects by obliquity alone were not enough to compensate the effect of the further cooling deep sea. Some help from other factors (eccentricity, precession and possibly non-orbital factors) was needed to reach that warmer and more stable ‘interglacial state’. As Renee Hannon concludes: “During the last 450 kyrs, the five major warm onsets with rapidly increasing temperatures are triggered by increases in the eccentricity, obliquity, and precession of Earth’s orbit. The nearly concurrent increase in these three astronomical forces appears a necessary component for a major warm onset”.Italics added.
The ‘Earth’s General Background Temperature’ continuously went down. The oceans cooled and processes changed.
The oceans gained heat content during the Holocene: deep sea temperatures rose. But since the Holocene Optimum the ocean heat uptake showed a diminishing trend. During the Little Ice Age, the oceans even experienced a net loss in heat content. Important, because now we know at what average temperatures the Earth starts cooling her oceans. Figure 6.
Figure 6: Holocene reconstructions of intermediate water temperatures. (C) Changes in Intermediate Water Temperatures (IWT) at 500 m, and (D) changes in IWT at 600 to 900 m. All anomalies are calculated relative to the temperature at 1850 to 1880 CE. Shaded bands represent T1 SD. Note the different temperature scales. Source: Rosenthal et al.
Only when the trend of continuously falling deep sea temperatures ends, the Earth will continue to be able to get out of a next glacial state. But, if this ever lower deep-sea temperature trend is not changed into a stable or a rising temperature, a ‘constant glacial state’ is what we can expect somewhere in the future.
Then glacials could continue without being interrupted by an interglacial and they could keep the Earth cold for a very long time – millions of years – in a barren glacial state. Which probably will be more severe than our Last Glacial Maximum, because the strong cooling during the glacial trend will not be ended by a warming climate state that could raise the deep-sea temperature to warm the Earth for a longer period.
Perhaps our Pleistocene glacial – interglacial rhythm was just a transition period to a more constant glacial state. The 41,000 → 100,000 trend might indicate such a future.
During the last 15 million years deep sea temperatures were continuously falling. Because of the process of oceanic upwelling the falling deep sea temperatures made sea surfacetemperatures fall as well. In turn, sea surface temperatures lowered atmospheric temperatures. A small decrease in deep sea temperatures resulted in an amplified surface temperature response. Surface temperatures responded 2.5 times the deep-sea response, such that a 0.2°C cooler deep sea resulted in a 0.5°C cooler surface temperature.
At a certain point, the falling deep sea temperatures resulted in atmospheric temperatures that enabled the development of large scale Northern Hemisphere snow and ice surfaces that increased the albedo of the Earth. That albedo caused a further cooling and resulted in even more snow and ice; another amplifier. Continental ice sheets developed. That was the moment the warm Pliocene terminated and the colder Pleistocene started.
Within the Pleistocene, short warmer and more stable periods – the interglacials – were alternating with glacial periods. During an interglacial the Earth reaches the more favourable ‘normal’ pre-Pleistocene state and is warmer and much more stable. Those interglacials first happened every 41,000 years, but as the deep-sea temperatures (and so atmospheric temperatures) decreased, more favourable orbital circumstances, rather than only increasing obliquity, were needed to get out of the cold glacial climate state. Because of the colder deep oceans during the last part of the Pleistocene the Earth only succeeded every 100,000 years in creating an interglacial.
If the 15-million-year trend of ever decreasing deep sea temperatures is going to continue, somewhere in the future the Earth will not be able to create a next interglacial. Millions of years of a deep and continuing glacial state might be in the future.
Earth’s general background temperature is also influenced by where the continental plates are located, and what kind of ocean currents, strength and direction it allows, if any. More oceans at the poles means lower general background temperature on earth and more ocean in the temperate zones would mean higher general background temperature as more water is being heated. Milankovitch’ cycles works on top of earth’s latent (general) background temperature.