E-OG

Marine extinction intensity during Phanerozoic

%

Millions of years ago

(H)

K–Pg

Tr–J

P–Tr

Cap

Late D

O–S

Eocene–Oligocene extinction is labeled E– OG.

The Eocene–Oligocene extinction event, also called the Eocene-Oligocene transition (EOT) or Grande Coupure (French for "great cut"), is the transition between the end of the Eocene and the beginning of the Oligocene, an extinction event and faunal turnover occurring between 33.9 and 33.4 million years ago.^[1] It was marked by large-scale extinction and floral and faunal turnover, although it was relatively minor in comparison to the largest mass extinctions.^[2]

The boundary between the Eocene and Oligocene epochs is marked by the glaciation of Antarctica and the consequent beginning of the Late Cenozoic Ice Age.^[3] This enormous shift in climatic regime is the leading candidate for the extinction event's cause. Though ephemeral ice sheets may have existed on the Antarctic continent during parts of the Middle and Late Eocene,^[4] this interval of severe global cooling marked the beginning of permanent ice sheet coverage of Antarctica,^[5][6] and thus the end of the greenhouse climate of the Early Palaeogene.^[7] In central North America, the mean annual temperature (MAT) fell by about 8.2  ±  3.1 °C over a span of 400,000 years.^[8] Near-freezing conditions existed in central Tibet.^[9] The global cooling also correlated with marked drying conditions in low-latitudes Asia,^[10] though a causal relationship between the two has been contradicted by some research.^[11] The equatorial seas were marked by exceptionally low palaeoproductivity in the EOT's aftermath.^[12] Deep ocean temperatures plummeted in the eastern equatorial Pacific during the EOT.^[13]

A leading model of climate cooling at this time predicts a decrease in atmospheric carbon dioxide, which slowly declined over the course of the Middle to Late Eocene.^[14][15][16] Significant cooling took place in the final hundreds of thousands of years prior to the start of major Antarctic glaciation.^[17] This cooling reached some threshold approximately 34 million years ago,^[18][19][4] precipitating the formation of a large ice sheet in East Antarctica in response to falling carbon dioxide levels.^[20][21] The cause of the drop in pCO₂ was the drift of the Indian Subcontinent into equatorial latitudes, supercharging the silicate weathering of the Deccan Traps.^[22] Another factor was the opening of the Drake Passage and the creation of the Antarctic Circumpolar Current (ACC), which had the effect of creating ocean gyres that promote upwelling of cold bottom waters and diminishing heat transport to Antarctica by isolating the water around it.^[23] Likewise, the Tasmanian Gateway also opened up around the time of the EOT.^[24] Ocean circulation changes were, however, not as significant in engendering cooling as the decline in pCO₂.^[25] On top of that, the timing of the creation of the ACC is uncertain.^[26] The deepening of the calcite compensation depth increased carbonate ion storage in the ocean shortly before the onset of the Antarctic glaciation, suggesting the events may have been coupled.^[27]

Evidence points to the glaciation of Antarctica occurring in two steps, with the first step, the less pronounced and more modest step of the two, taking place at the Eocene-Oligocene boundary itself. This first step is referred to as EOT-1,^[4] which occurred approximately 34.1-33.9 Ma.^[28] Carbon dioxide concentrations dropped from about 885 ppm to about 560 ppm.^[29] The Oligocene Oi-1 event, an oxygen isotope excursion that occurred around 33.55 Ma,^[30] was the second major pulse of Antarctic ice sheet formation.^[4]

These large climate changes have been linked to biotic turnovers. Even before the Eocene-Oligocene boundary itself, during the early Priabonian, extinction rates went up in connection with falling global temperatures.^[29] Radiolarians suffered major losses thanks to a decrease in nutrient availability in deep and intermediate waters.^[31] In the Gulf of Mexico, marine turnover is associated with climatic change, though the ultimate cause according to the study was not the drop in average temperatures themselves but colder winters and increased seasonality.^[2]

On land, the increased seasonality brought on by this abrupt cooling caused the Grande Coupure faunal turnover in Europe.^[32] In the Ebro Basin, major aridification occurred amidst the Grande Coupure, suggesting causality.^[1] The remarkable cooling period in the ocean is correlated with pronounced mammalian faunal replacement within continental Asia as well. The Asian biotic reorganization events are comparable to the Grande Coupure in Europe and the Mongolian Remodeling of mammalian communities.^[33]

Another speculation points to several large meteorite impacts near this time, including those of the 40-kilometre (25 mi) diameter Chesapeake Bay crater^[34] and the 100-kilometre (62 mi) diameter Popigai impact structure of central Siberia,^[35] which scattered debris perhaps as far as Europe. New dating of the Popigai meteor strengthens its association with the extinction.^[36] However, other studies have failed to find any association between the extinction event and any impact event.^[37][38]

Imprints of sunspot cycles from the Bohai Bay Basin (BBB) show no evidence that any significant change in solar activity occurred across the EOT.^[39]

In central North America, reptiles, amphibians, and gastropods underwent drastic faunal turnover likely spurred on by a precipitous drop in mean annual temperature (MAT) over approximately 400,000 years.^[8] Malagasy lemurs experienced a significant extinction during the EOT.^[40]

The Grande Coupure, or 'great break' in French,^[41] with a major European turnover in mammalian fauna about 33.5 Ma, marks the end of the last phase of Eocene assemblages, the Priabonian, and the arrival in Europe of Asian species. The Grande Coupure is characterized by widespread extinctions and allopatric speciation in small isolated relict populations.^[42] It was given its name in 1910 by the Swiss palaeontologist Hans Georg Stehlin, to characterise the dramatic turnover of European mammalian fauna, which he placed at the Eocene–Oligocene boundary.^[43] A comparable turnover in Asian fauna has since been called the "Mongolian Remodelling".

The Grande Coupure marks a break between endemic European faunas before the break and mixed faunas with a strong Asian component afterwards. Before the Grande Coupure, European faunas were dominated by anoplotheriid, xiphodontid, choeropotamid, cebochoerid, dichobunid, and amphimerycid artiodactyls, palaeotheriid perissodactyls, pseudosciurid rodents, adapid and omomyid primates, and nyctitheriids. Post-Grande Coupure artiodactyl faunas in Europe are dominated by gelocids, anthracotheriids, and entelodontids, with true rhinoceroses representing the perissodactyl fauna, eomyids, hamsters, and beavers representing the rodent fauna, and hedgehogs representing the eulipothyphlan fauna. The speciose genus Palaeotherium plus Anoplotherium and the families Xiphodontidae and Amphimerycidae were observed to disappear completely during the Grande Coupure.^[44] An element of the paradigm of the Grande Coupure was the apparent extinction of all European primates at the Grande Coupure. However, the 1999 discovery of a mouse-sized early Oligocene omomyid, reflecting the better survival chances of small mammals, undercut the Grande Coupure paradigm.^[45] Herpetotheriids, cainotheriids, dormice, and theridomyids survived the Grande Coupure undiminished.^[44] Balkanatolia acted as a staging ground for Asian taxa that immigrated into Europe following the extinction of its own mammal fauna during the Grande Coupure.^[46]

It has been suggested that this was caused by climate change associated with the earliest polar glaciations and a major fall in sea levels, or by competition with taxa dispersing from Asia. However, few argue for an isolated single cause. Other possible causes are related to the impact of one or more large bolides in the Northern Hemisphere at Popigai, Toms Canyon, and Chesapeake Bay.^[47] Improved correlation of northwest European successions to global events confirms the Grande Coupure as occurring in the earliest Oligocene, with a hiatus of about 350 millennia prior to the first record of post-Grande Coupure Asian immigrant taxa.^[44] Research suggests that in the Ebro Basin of Spain, the turnover lagged the Eocene-Oligocene boundary by at most 500,000 years (500 kyr).^[1]

Additionally, a second dispersal event of Asian taxa into Europe, known as the Bachitherium Dispersal Event (named after the ruminant Bachitherium), occurred later, around 31 Ma. Unlike the Grande Coupure, which took place via Central and Northern Asia, this later dispersal occurred via a southern corridor.^[48]

In the Caribbean, mangroves dominated by Pelliciera rapidly disappeared, becoming replaced by mangroves that were dominated by Rhizophora, which remains the main constituent of Caribbean mangroves in the present day. This turnover has been named the Caribbean Mangrove Revolution.^[49]

In the marine realm, the frequency of drilling in recovery faunas, especially among bivalves, was drastically higher than in assemblages before the extinction event, a phenomenon attributed to a high extinction rate among escalated prey taxa with highly evolved defences against predators.^[50] Veneroid bivalves experienced a short-term size increase during the biotic recovery.^[51] Orthophragminid foraminifera (late Paleocene–early Eocene larger benthic foraminifera of two families, Discocyclinidae and Orbitoclypeidae) disappeared in the extinction event; in Alpine carbonates, bryozoan facies show an expansion in response to the loss of orthophragminids.^[52] The EOT is often considered to be a critical turning point in the rise of diatoms to their present-day evolutionary prominence, though this paradigm has been criticised for being based on incomplete evidence.^[53]

Some sites contain evidence that the Eocene–Oligocene extinction was not a sudden event but a prolonged biotic transition drawn out over as much as 6 million years. Localities near Eugene, Oregon, record a plant extinction 33.4 Ma (million years ago) and a marine invertebrate turnover 33.2 Ma; both of these turnovers post-date the supposed extinction event by hundreds of thousands of years.^[54]

• Extinction an essay by Richard Cowen