Antarctic Refrigerator Effect & Climate Sensitivity

Antarctic Refrigerator Effect, Climate Sensitivity & Deformation professionnelle: 

Lessons from Past Climate Change - Part 2

For the past 55 million years the global surface temperature has declined by more than 10°C from a “hot house” condition into an “ice house” with increasing temperature variability as depicted in Figure 1 (Mya = millions of years ago). During the Cretaceous and Early Cenozoic, glaciers and ice caps were absent from both Antarctica and Greenland. Antarctica was covered in para-tropical vegetation and Greenland was home to crocodiles. More importantly for millions of years the oceans had been storing enormous amounts of heat. In contrast to near freezing temperatures today, Antarctic bottom waters averaged about 11°C, suggesting Antarctic coastal temperatures never dropped below 11°C even during the long polar nights.  Amazingly the equator to pole surface temperature difference averaged just 10°C compared to the 30°C gradient measured today. Of particular interest, changes in carbon dioxide fail to explain the greatest proportion of these ancient temperatures.

Figure 1  Cenozoic Evolution of Global Temperatures

For decades the consensus had been that ocean heat transport had ultimately maintained the polar regions’ ancient tropical conditions. Models had demonstrated without heat transport from the tropics, the poles would be 110°C colder than the tropics (Gill 1982, Lozier 2012). It was commonly believed, and is still believed by most, that as plate tectonics rearranged the continents, the Antarctic Circumpolar Current (ACC) formed and strengthened. Models now simulate that as drifting continents opened “gateways” and allowed for uninterrupted circumpolar flow, surface temperatures began cooling significantly (Bijl 2013). A strengthening ACC created a barrier inhibiting intrusions of warm tropical waters and minimizing both oceanic and atmospheric heat transport resulting in the Refrigerator Effect. The Refrigeration Effect radically cooled the southern ocean and altered the vertical temperature structure of all the earth’s oceans. (As discussed here, the ACC barrier to ocean heat transport is a major reason why Antarctic sea ice has currently increased in contrast to decreasing Arctic sea ice.)

However a few climate modelers began arguing CO2, not heat transport, was the ultimate climate control knob. They argued that high CO2 concentrations explained the polar warmth and the decline in CO2 explained the advent of polar ice caps and the 55 million year trend towards our icehouse climate. This debate between heat transport and greenhouse effects not only reveals a lack of climate consensus; it also reveals the subjectivity that influences how climate sensitivity is estimated. Proxy evidence increasingly suggests that ancient CO2 levels were far lower than what climate models require to simulate ancient warmth. In stark contrast to current research that is increasingly suggesting lower climate sensitivity to CO2 (i.e. Lewis & Curry 2014, and a growing list here and here), paleoclimate researchers who argue CO2 controls climate change, are forced to suggest climate sensitivity must have been much, much greater than anyone currently believes.

In contrast, researchers examining the Paleocene‑Eocene maximum temperatures concluded, “At accepted values for the climate sensitivity to a doubling of the atmospheric CO2 concentration, this rise in CO2 can explain only between 1 and 3.5°C of the [5-9°C] warming inferred from proxy records. We conclude that in addition to direct CO2 forcing, other processes and/or feedbacks that are hitherto unknown must have caused a substantial portion of the warming during the Paleocene–Eocene Thermal Maximum. Once these processes have been identified, their potential effect on future climate change needs to be taken into account.” [Emphasis added]  Zeebe (2009). 

However if “unknown feedbacks” and other forcings can explain an even greater proportion of past temperature changes, then researchers would be forced to suggest climate sensitivity to CO2 is much lower. The Antarctic Refrigerator Effect is such an effect and parsimoniously explains Cenozoic global cooling without invoking a CO2 contribution.

The Case Against a CO2 Climate “Control Knob

By creating a well‑mixed global “blanket”, the carbon dioxide greenhouse effect should act on a global scale. However as illustrated in Figure 1, initiation of Antarctic glaciation happened 35 million years ago, more than 30 millions years before Arctic glaciation ever began. Clearly Antarctic glaciation was not part of a global event, but a regional one. Although this gross time difference does not rule out a limited contribution from diminishing CO2 concentrations, the evidence most assuredly demands a different and more regional explanation for the drivers of Antarctica’s observed climate change.

Furthermore, in order for a CO2 greenhouse effect to have created the near‑tropical conditions observed in Antarctica’s fossil evidence, it requires CO2 concentrations far greater than what the growing number of paleo‑proxies are suggesting. Huber (2011) argued that their models could simulate tropical warmth in polar regions  if  CO2 reached 4480 ppm, an 11‑fold increase above today’s concentrations. However Huber 2011 also admitted their estimate of CO2 concentrations should not be taken literally. Instead it was his approach “equivalent to “tuning” climate sensitivity to a higher value, but is much simpler in practice.” They argued that the  “4480 ppm CO2 concentration used here should not be construed literally: it is merely a means to increase global mean warmth.”

Huber 2011 was wise to admit CO2 concentrations of 4480 are unrealistic. Based on growing proxy evidence, CO2 concentrations during the past 350 million years have not exceeded 1000 ppm (Franks 2014). However Huber 2011’s suggestion of greater CO2 climate sensitivity proves to be equally inappropriate and most likely a case of déformation professionnelle.

Deformation professionnelle is a French term referencing how one’s profession narrows and distorts one’s viewpoint and thus biases conclusions. For example if researchers whose funding and status has been driven by a paradigm that CO2 drives all climate change, any contrary evidence will be reinterpreted to maintain that viewpoint.

One major avenue of research strives to determine climate sensitivity by comparing varying CO2 concentrations with past climate change. Although Franks 2014 determined past CO2 variations only accounted for 20% of what Huber’s models required, they too felt obligated to suggest there must be a much greater climate sensitivity to the smaller changes in CO2. But they obviously ignore much more parsimonious inferences. Very simply, there are other dynamics that drive climate change, and current models driven by CO2 have failed to incorporate additional and alternative explanations. Similarly CO2 variations are insufficient to explain the Dansgaard‑Oeschger extreme warming events of the last Ice Age. But as discussed here, changes in ocean heat storage and ventilation offer a superior explanation. Likewise Antarctica’s Refrigerator Effect completely altered ocean heat storage and ventilation and can parsimoniously accounts for Cenozoic global cooling.

Unfortunately evaluations of CO2 climate sensitivity typically only compare varying CO2 concentrations with other estimated radiative effects to explain fluctuations in global mean surface temperatures (GMST).  However there are other powerful non-radiative effects that also contribute to a varying GMST as well as the increasing equator to pole temperature gradient.  For example examining changes in Cenozoic climate  Thomas (2014) concluded, “Stronger vertical mixing within the oceans potentially reconciles several long-standing greenhouse paleoclimate problems. Stronger vertical mixing invigorates the MOC [Meridonal Overturning Circulation] by an order of magnitude, increases ocean heat transport by 50–100%, reduces the zonal mean equator-to-pole temperature gradients by up to 6°C, lowers tropical peak terrestrial temperatures by up to 6°C, and warms high-latitude oceans by up to 10°C.”

Given that just the upper 3.5 meters of the oceans hold more heat than our entire atmosphere. And that average depth of the oceans is an order of 3 magnitudes greater, about 3600 meters; changes in ocean heat storage and ventilation have humongous impacts on global climate. Research that ignores contributions to GMST from ocean heat storage, ventilation and vertical mixing, CO2, will greatly exaggerate climate sensitivity to CO2. Today we witness global warming from heat ventilation during an El Nino and global cooling due to increased upwelling of cooler waters during La Ninas.  On time scales varying from a few years to millions of years, storage and ventilation of ocean heat has been the earth’s true climate control knob.

The Antarctic Refrigeration Effect

Our modern freezer and refrigerator appliances are all based on 2 simple principles. 1) A compressor‑refrigerant apparatus pumps heat out of the refrigerator’s interior. 2) The refrigerator is insulated to minimize any heat transfer into the refrigerator from the outside.

The Antarctic analogy to a refrigerant/compressor apparatus has been ever present. Due to the earth’s spherical shape and orbital effects, annual incoming solar radiation at the poles is so low, polar regions always radiate more heat back to space than is ever absorbed locally. Without a constant flow of heat from the tropics, polar regions would naturally descend into permanent ice house climates. Forcing by CO2 is not a significant factor, if a factor at all. Thus variations in Antarctica’s climate are governed by changes in heat transport versus the steady radiation of heat back to space.  Although Antarctica sat over the South Pole for hundreds of millions of years, it remained ice free for most of the Mesozoic and early Cenozoic because the “refrigerator door” was left open. However as continents began to shift and opened “ocean gateways”, the Antarctic Circumpolar Current (ACC) formed and intensified. The ACC closed the refrigerator door and resisted poleward heat transport from the tropics.  The ACC also generated more intense westerly winds and invigorated upwelling that increased vertical mixing. Most importantly as the ACC shut the refrigerator door, sea ice began forming in the southern seas. That initiated deep ocean cooling and a total reconfiguration of the global ocean’s vertical heat structure.

Before the ACC formed, Antarctic bottom waters were about 11°C. Bottom waters formed from competing regions. In shallow seas that dominated subtropical regions, warm salty water became dense enough to sink to the bottom. Elsewhere warm salty subtropical waters that were transported poleward cooled and sank.  In contrast, once the Antarctic refrigerator was established, cold salty brine was now extruded during sea ice formation. The sinking of cold brine either penetrated to the abyss forming near freezing bottom water, or slowly cooled the subsurface waters as the brine was turbulently mixed with its surroundings. Thus global oceans began a 35 million year cooling trend starting from the ocean abyss and working its way to the surface.

In Figure 13 below (from Kennett 1990), the bottom frame labeled Proteus, illustrates a simplified vertical structure of the Atlantic Ocean around 60 million years ago. Warm Salty Deep Water (WSDW) dominated the ocean depths. Much of that warm bottom water is believed to have been generated in shallow equatorial seas, like the Tethys, where evaporation exceeded precipitation.  Our modern Mediterranean Sea is a remnant of the Tethys, and still contributes warm salty water to the Atlantic.

The surface waters around Antarctica were much fresher because cooler polar regions experience greater precipitation relative to evaporation. Antarctic Intermediate Water (AAIW) forms as upwelling bottom waters mixed with fresher surface waters. Subsequently, climate change has been greatly affected as Antarctic Intermediate Water have cooled and exerted a tremendous effect on tropical sea surface temperatures for millions of years via “ocean tunneling”.

The middle frame of Figure 13, labeled Proto-Oceanus, illustrates how the ocean’s vertical structure evolved over the next 10+ million years after the formation of a strong Antarctic Circumpolar Current. Due to the refrigerator effect, cold saline Antarctic Bottom Water (proto‑AABW) began to dominate the ocean floor. Contributions of Warm Saline Deep Water (WSDW) diminished, and the influential Antarctic Intermediate Water (AAIW) was increasingly cooled by much colder Antarctic Bottom Water. As the colder AAIW flows back towards the equator it modulates the global temperature by cooling subsurface waters that would potentially reach tropical surfaces via upwelling.

 

65 million year Evolution of Ocean Vertical Heat Structure

The upper frame labeled Oceanus, represents a simplified illustration of the Atlantic’s modern vertical structure. Due to the Antarctic Refrigerator Effect, the deep oceans continued to cool, and the thermocline that separates warm surface water from cooler deep waters became increasingly more shallow.

Between 2 and 3 million years ago the cooling of the deep oceans reached a tipping point, and modern upwelling regions ogf cold deep water off the coast of Peru, California and the west coast of Africa were established. There had always been upwelling along those coasts along with the associated increases in marine productivity. But now upwelled subsurface waters were cooler by 4 to 9°C. (Dekens 2007), corresponding to the cooling by Antarctic Bottom waters and its effect on subsurface waters. Analogous to the drop in global temperatures during La Nina events caused by upwelling of colder waters, upwelling of colder waters 2 to 3 million years ago also cooled global temperature to the point it initiated Arctic ice cap and glacier formation. The cooler Arctic then promoted formation of North Atlantic Deep Water (NADW in the upper frame of Figure 13) as salty Atlantic waters transported poleward cooled and brine rejection increased as more Arctic sea ice formed.

Declining CO2: A Result Not A Cause.

The Cretaceous Period (145 to 65 million years ago) was named for huge widespread chalk deposits that developed during that time period, especially in the Tethys Sea. Those chalk deposits were the result of sinking plankton that produced calcium carbonate shells like foraminifera and coccolithophorids, As discussed in Natural Cycles of Ocean Acidification, the creation of calcium carbonate shells pumps alkalinity to depth but produces CO2 at the surface thus adding to higher concentrations of atmospheric CO2.  More enlightening and contrary to catastrophic CO2 assertions that rising CO2 will decimate calcium carbonate shell producers, the greatest proliferation of calcium carbonate shell producers occurred during this period with the high temperatures and high concentrations of atmospheric CO2. Quite likely, high CO2 concentrations did not produce detrimental acidification, and were the result of coccolithophorids and foraminifera pumping CO2 to the surface.

The development of the Antarctic Circumpolar Current forever altered the carbon biological pump by increasing upwelling in the southern oceans, and later along continental west coasts by cooling upwelled waters. When the ACC caused upwelling in southern oceans to intensify, a more reliable supply of nutrients supported the evolution and proliferation of diatoms. As discussed in Natural Cycles of Ocean Acidification, diatoms are large, produce siliceous shells, and more rapidly shuttle CO2 from the surface to ocean depths. As evolving diatom populations expanded, a more efficient biological pump buried more CO2 at depth that is now detected as siliceous ooze or as biogenic opal deposits. In contrast CO2 emitting coccolithophorid populations and their chalk deposits dwindled. Changes in the biological pump contributed to observed declines in atmospheric CO2. Diatoms are also associated with explosive increases in ocean productivity, so it should be no surprise that the earliest appearance and evolution of whales also coincides with increased ACC upwelling and the evolution of diatoms.

In summary, due to continental drift, the formation of the Antarctic Circumpolar Current blocked intrusions of warm tropical waters that warmed Antarctic and initiated the Antarctic Refrigerator Effect. Cold polar regions are a natural result of inadequate solar radiation. Reduced forcing from diminished levels of CO2 is not required to explain global cooling. The resulting formation of Antarctic sea ice expelled colder, salty waters that filled the abyss and began cooling the deep oceans. After 30+ million years of cooling, 2 to 3 million years ago, colder ocean waters eventually upwelled in the mid latitudes along the west coasts of major continents as well as along the equator. The resulting global cooling, allowed the growth of Arctic ice caps, glaciers and sea ice. The Antarctic Circumpolar Current also increased global upwelling and the efficiency of the biological pump. Decreases in atmospheric CO2 are associated with reductions in populations of CO2 producing coccolithophorids along with increasing populations of diatoms that pumped CO2 to depth. If the Antarctic Refrigeration Effect can account for the changes in global temperatures, it suggests the global sensitivity to varying levels of CO2 is relatively insignificant.