CO2 & Temperature During The Middle Eocene Climatic Optimum

Though Earth and its climate are billions of years old, climate science is still very young. So young that surprising new discoveries are constantly being made. One such discovery in the field of paleoclimatology—the study of Earth's climate in the distant past—was the uncovering of a period of great warming around 40 million years ago, in the middle of the Eocene Epoch. In the midst of a general cooling trend beginning at the end of the preceding Paleocene Epoch (~55 mya) there were a number of dramatic, sudden bursts of global warming. The most celebrate of these is the Paleocene-Eocene Thermal Maximum or PETM, when surface temperatures rose by 5-7°C. Recently, science has discovered another hot interval 15 million years later during the Middle Eocene. Named the Middle Eocene Climatic Optimum (MECO), it marked a time when deep sea temperatures rose about 4-5°C and atmospheric CO2 levels peaked. As new information is uncovered, climate scientists are scrambling to interpret what caused this second, more sustained period of warming and what it may mean for current climate conditions.

The Eocene world was notably different from the modern one. Though the land masses that would become the modern continents are discernible, they differ from their modern counterparts in both shape and position. As described in The Resilient Earth, this was partly due to higher sea-levels and the ongoing breakup of Gondwanaland. Both Australia and South America were much closer to Antarctica, blocking the free circulation of water around the southern polar region. India was slowly colliding with the underbelly of Asia, a collision that would raise the Himalayan Mountains.

Earth during the early Eocene ca. 50 mya. Source Ron Blakey.

The name Eocene comes from the Greek and translates as “new dawn,” a reference to the first appearance of modern mammals. The Eocene climate is thought to have been quite homogeneous, with the temperature gradient from equator to poles only half what it is today. Rainy tropical climates extended as far north as 45° though the climate in the tropics was similar to today's. Most notably, the polar regions were much warmer than today and temperate forests extended right to the poles.

The first researchers to identify a significant warming event in the middle Eocene were Steven M. Bohaty and James C. Zachos, both from the Earth Sciences Department, University of California, Santa Cruz. In a 2003 article in Geology, entitled “Significant Southern Ocean warming event in the late middle Eocene,” they identified a prominent warming event in Southern Ocean deep-sea cores, indicating that long-term cooling through the middle and late Eocene was not monotonic as previously thought. Here is how the authors stated the significance of their find:

Stable isotope data from Southern Ocean sites reveal anomalous temperature variability during the middle and late Eocene. The ca. 41.5 Ma middle Eocene climatic optimum event is interpreted to represent an important climatic reversal in the midst of long-term cooling in the middle to late Eocene, indicating that this trend was not entirely monotonic. If global in nature, the middle Eocene climatic optimum would represent one of the more rapid global warming events of the Cenozoic. Regardless of its full global extent, the middle Eocene climatic optimum is clearly an important event in the regional climatic history of the Indian-Atlantic region of the Southern Ocean (Fig. 2). The rapidity and magnitude of warming (4 °C) imply that this climatic event dramatically affected both Southern Ocean biological communities and the coastal environments of Antarctica and Australia.

Now, a new study by Peter K. Bijl et al., “Transient Middle Eocene Atmospheric CO2 and Temperature Variations ,” claims to have refined estimates of CO2 levels during the MECO warming. Writing in the November 5, 2010, issue of Science, Bijl et al. report on a sedimentary succession spanning the MECO recovered from the East Tasman Plateau at Ocean Drilling Program (ODP) Site 1172 (the location of the site is shown in the figure below).

Paleogeographic configuration of the southern high latitudes during the middle Eocene.
Though the researchers warn that their results are based on a single core sample, marked by the orange star on the map, they nonetheless claim to have obtained good proxy measurements for both sea surface temperature (SST) and atmospheric partial pressure of carbon dioxide (pCO2). Here are the authors' descriptions of how these measurements were made:

To fully capture the magnitude of the sea surface temperature (SST) change associated with the MECO at this site, we applied two independent temperature proxies: the alkenone unsaturation index (UK37) and the index of tetraethers consisting of 86 carbon atoms (TEX86). At the onset of the MECO, UK37 and TEX86 indicate a rise in SST of 3°C and 6°C, respectively, which, also at this location, stands out as an interruption of long-term middle Eocene cooling. Bulk carbonate oxygen isotope values (δ18O) decrease by 1.0 to 1.2 per mil (‰), which, if controlled by SST only, also indicate a SST rise of ~4° to 5°C.


We assessed pCO2 changes by determining the stable carbon isotopic composition (δ13C) of alkenones, long-chained ketones exclusively synthesized by specific haptophyte algae. Carbon isotopic fractionation during carbon fixation (εp) by haptophyte algae varies as a function of dissolved CO2 [CO2(aq)], specific cell physiological parameters (which show good correspondence to the surface-water concentrations of soluble phosphate), and other environmental parameters, primarily light intensity. The carbon isotopic composition of diunsaturated alkenones (δ13CC37:2) ranges between –32.5 and –35.5‰. We used bulk carbonate δ13C to estimate the δ13C value of the dissolved inorganic carbon (DIC) pool in seawater to determine εp.

There are a number of cautions to take note of. First, the relationship between εp and pCO2 is exponential, which results in a relatively large uncertainty in reconstructed pCO2 levels with high εp values. In particular, “the soluble phosphate concentration exerts a strong influence on the relation between εp and pCO2, particularly if εp values are high.”

Second, chemical concentrations present at the time could have biased today's proxy readings and there is no way to correct for them. “With a realistic range of phosphate concentrations, pCO2 values were between 600 and 1600 ppmv just before the MECO, which is in line with previous estimates of middle Eocene pCO2 values using the same proxy, and rose to between 6400 and 15,000 ppmv during the MECO,” the authors state.

This means the calculated baseline pCO2 readings vary by ±45%, not the most accurate of measurements. Variation for the maximum pCO2 value is even more stunning. This means that the amount of CO2 in the atmosphere increased between 4 fold on the low end and 25 fold on the high end of these estimated figures. The error bars are indeed quite wide on these results.

Sea surface temperatures from two different proxies (red and purple). Atmospheric CO2 levels from algae (light gray band) and phosphate estimates (dark gray band). Yellow shaded area indicates the MECO warming interval. After Bijl et al./Science.

As for temperatures, Bijl et al. state that “absolute SSTs as indicated by UK37 and TEX86 are consistent, with 26°C or 24°C just below the onset of the MECO for the two proxies, respectively, and peak MECO SSTs exceeding 28°C.” In other words about 10°C higher than today. This is in good agreement with earlier work by Fredrik P. Andreasson and Birger Schmitz, reported in the April 2000 GSA Bulletin:

We have established intrashell stable isotope profiles of shallow-water gastropods from the warm early middle Eocene Epoch in order to determine the seasonality of coastal sea surface temperature (SST). Oxygen isotope profiles of shells from Texas and Mississippi suggest a seasonality of 8–9 °C along the early middle Eocene U.S. Gulf Coast, with a winter temperature of 19 °C and a summer temperature of 27–28 °C. Relative to the present temperatures in the area, the Eocene summer temperature was similar, whereas the winter temperature was 7–8 °C higher. A probable reason is a smaller impact than today by cold continental air from the north because of higher continental winter temperatures. Isotope profiles of shells from southern England indicate early middle Eocene seasonality similar to present day, about 10–12 °C, whereas the mean temperature was 8–10 °C higher in Eocene time. These data confirm previous temperature estimates of the early middle Eocene Epoch in France.

It looks like the estimate of a 4-5°C rise from the Eocene baseline to MECO maximum is consistent with work by other researchers. The strange thing is, Bijl et al. end up estimating a mid Eocene climate sensitivity, the amount of temperature rise for a doubling of atmospheric CO2, at ~2° to 5°C. This could only be derived by selecting CO2 increases at the bottom of their calculated range. If the range of increases given above are used the sensitivity would fall in the range of 0.2-1.25°C. Of course, their value is “a tentative assessment of high-latitude climate sensitivity to CO2 forcing on ~100,000-year time scales, assuming that all MECO warming was caused by pCO2 and associated feedbacks.”

The complexity of those feedbacks is touched upon in the paper and uncertainty regarding the ocean environment 40 mya is quite large. Still, it seems that to arrive at an estimated sensitivity that is in the same range as IPCC estimates for current conditions is too coincidental. Given that Eocene atmospheric CO2 started at 2 to 4 times modern levels one would expect a lower sensitivity. Add to that the lack of ice sheets, the melting of which could provide a strong warming feedback by changing albedo, and the authors' sensitivity numbers make little sense.

Finally, what caused this precipitous rise in CO2 levels? According to an accompanying Science perspective article by Paul N. Pearson, the original hypothesis for the MECO involved the disappearance of an ocean between India and Asia as the Himalayas were formed. “Did something unusual happen in this area about 40 million years ago that gradually released a huge amount of CO2?” Pearson asks.

About 225 million years ago, India was a large island still situated off the Australian coast, and a vast ocean (called Tethys Sea) lay between India the Asian continent. When Pangaea broke apart about 200 million years ago, India began to drift northward. About 80 million years ago, India was located roughly 6,400 km south of the Asian continent, moving northward at a rate of about 9 m a century. When India rammed into Asia about 40 to 50 million years ago, its northward advance slowed by about half. The collision and associated decrease in the rate of plate movement are interpreted to mark the beginning of the rapid uplift of the Himalayas. There are indications that the shrinking remnant of the Tethys Sea drained and flooded many times during this collision of continents.

One hypothesis is that the mantle plume that caused the massive Deccan Traps volcanism, which peaked around 65 mya, is to blame. Renewed volcanic eruptions could have released the carbon from subducted, carbon-rich sediment left by the draining of the Tethys Sea. It is clear from the article that the authors do not agree with this hypothesis but have no good explanation of their own for where all the carbon came from, just that it somehow got into the atmosphere and ocean. Here is their discussion of possible causes:

One outstanding issue is the source of carbon responsible for the increase in middle Eocene atmospheric CO2. The rise in pCO2 by 2000 to 3000 ppmv emerging from our data requires a carbon source capable of injecting vast amounts of carbon into the atmosphere. Moreover, the absence of a prominent negative carbon isotope excursion excludes reservoirs with δ13C signatures below that of marine DIC. One mechanism capable of emanating carbon with such a geochemical signature is the metamorphic alteration of carbonates (decarbonation). Massive decarbonation occurred until the late Eocene, with the subduction of vast amounts of Tethyan Ocean pelagic carbonates under Asia as India drifted northward. However, the flux of carbon required to increase pCO2 by 2000 to 3000 ppmv within ~400,000 years appears too high to invoke metamorphic (volcanic) outgassing as the sole mechanism.

Note that the level of CO2 didn't just rise by 3000 ppm, those levels were maintained for a prolonged period of time. The pulse of CO2 (or CH4) that caused the PETM was removed from the atmosphere quite rapidly, taking on the order of 20,000-40,000 years. In fact, a recent paper states that the ecosystem accelerated its rate of carbon sequestration in response to the PETM release (see “Rapid carbon sequestration at the termination of the Palaeocene–Eocene Thermal Maximum”). The implication here is that, though a massive PETM like release of CO2/CH4 can cause a rise in global temperatures, such a rise is only a perturbation of the system. The system will correct itself over time by removing any “excess” carbon from the atmosphere.

But the MECO rose to high levels over a period of half a million years, implying a continuous large infusion of carbon into Earth's climate system. Such levels of carbon flux dwarf those of possible human activity, as explained in “Could Human CO2 Emissions Cause Another PETM?” Still, with scientists at a loss for the source of the MECO carbon dioxide, it is a good thing that humans were not around at the time—we would inevitably be blamed.

Although the global climate remained comparatively warm throughout the rest of the Eocene the end of the epoch signaled the start of a global cooling trend that would eventually lead to permanent polar icecaps and the Pleistocene Ice Age. Scientists speculate that this global cooling was caused by the formation of the Antarctic circumpolar current which thermally isolated the southernmost continent. But it was not the changing climate that marked the end of the Eocene. As with most geologic time periods, the end of the Eocene is marked by a significant change in the fossil record—in other words an extinction event.

The Grande Coupure, or “great break,” around 33.5 mya affected mostly European fauna. While some paleontologists argue that the event was caused by climate change associated with the formation of polar glaciers and the accompanying fall in sea levels, others blame heightened competition from invading Asian species. A few even propose a more cosmic cause in the form of one or more asteroid impacts, including one in the vicinity of Chesapeake Bay on the east coast of North America. Most probably, it was a combination of multiple factors.

Earth could be headed for a new climatic optimum.

Despite this new paper's obvious flaws—most significant among them that data were gathered from a single sit in high southern latitudes—support for the discovery of the MECO is quite important. For one thing, it highlighting the fact that nature has provided both short, medium and long-term periods of global warming all on its own—no SUVs were needed to trigger the MECO. But the most significant contribution of Bijl et al. is in emphasizing just how large the natural buildup of atmospheric CO2 may have been—as much as 50 times modern levels. Yet even at these levels Earth was still quite livable.

It is worth noting that scientists call this very warm period a “climatic optimum,” much as the warmest period during the Holocene epoch is called the Holocene Climatic Optimum. Optimum is defined as the point at which the condition, degree, or amount of something is the most favorable. The Holocene Climatic Optimum is dated from about 5,000 to 3,000 BC and was a time when average global temperatures were 1 to 2° Celsius warmer than they are today. During that climatic optimum, many of humanity's great ancient civilizations began and flourished. What if Earth's climate is headed for another climatic optimum and we foolish humans are cajoled into doing everything we can to stop it? Acting out of fear of global warming could be the worst thing humanity has ever done.

Be safe, enjoy the interglacial and stay skeptical.


Is data precise enough to determine the simultaneity (or not) of the rise in temperatures and CO2 levels? After all, Vostok data shows a delay in CO2 rise.

Anyway, by the graph is is hard to tell, but in all the period it seems that the CO2 levels were on a general rising trend, while the temperatures were on a descending trend.

Source of carbon

I have often wondered what would happen should a large coal or oil deposit either subuct or see volcanism erupt through it. For example, imagine an oil field the size of that which exists in Saudi Arabia or a coal field such as that which exists in the Allegheny region. Now imagine a deposit that large subducting or some other plate subducting under it causing a series of volcanic eruptions through the deposit. I would imagine that the amount of CO2 released could be tremendous. Considering that as the surface is "recycled" over the millions of years, all that oil and coal is destined to erupt through a volcano anyway, I suppose.

Naturally Recycled Carbon

You are correct in your observation that the occurrence of volcanism in an area rich in hydrocarbon deposits will release copious amounts of CO2. The two major gases released from volcanoes are H2O and CO2. But it doesn't take oil or coal to release CO2, any carbonaceous rock will do. Since such rock is formed from sea floor deposits, and since plate material is constantly being subducted all over the world, the opportunity for carbon release from sedimentary rock is greater than that of oil and coal deposits. Even so, oil and coal do make their contribution to recycling geologically stored carbon into the active biosphere.

As people should have learned from the hoopla surrounding the Gulf oil spill, it doesn't take a volcano to release oil into the environment. Natural seepage occurs on land and under the seas in many parts of the world. Every year, nature seeps more oil into the ecosystem than humanity spills from wells, tankers and pipelines. This naturally released oil is accompanied by methane gas that eventually gets broken down into CO2. See “Crude Facts About Offshore Drilling” for details.

As for coal, it also doesn't need a volcanic eruption to burn in situ. There are thousands of coal seam files burning around the world, particularly in India and China. Some of these are manmade by most were ignited naturally by lightning strikes or wildfire. The amount of coal consumed each year has been estimated as being equal to the production of Ohio, Illinois and West Virginia. Wikipedia has a good article on coal seam fires. If so much CO2 can escape into the atmosphere from coal without digging it up and burning it, think about the folly of pumping CO2 back underground as called for in various “clean coal” schemes (see "Serious Black: The Quest for Clean Coal").