What was the Earth like when it was as hot on it as promised in 2100?

Today’s children will have their grandchildren when they live to the point where all climate forecasts end. Are there any tips in the past about our future?

Map of the modern Antarctic, which shows the speed of retreat (2010-2016) of the "clutch line", where glaciers lose contact with the bottom, as well as ocean temperatures. The lonely red arrow in the eastern Antarctic is the Totten Glacier, which contains so much water that it is enough to raise the level of the world ocean by 3 meters.

Everything that happened to us is just a prologue.

- William Shakespeare, "The Tempest"

The 2100th year looks like a line of restrictive flags at the finish line of climate change - as if all of our goals end exactly then. But, paraphrasing the warning on the rearview mirror, it is closer to us than it seems. Today’s children will have their grandchildren when they live to the point where all climate forecasts end.



However, in the 2100th year, the climate will not cease to change. Even if we successfully limit warming in this century to 2 ºC, the CO ₂ content in the air will be 500 ppm. Our planet has not seen such a level since the mid Miocene , 16 million years ago , when our ancestors were still great apes. Then the temperature was higher by 5–8 ºC , not by 2 ºC, and the sea level was higher by 40 meters , or even more — not by half a meter, which are expected by the end of this century, according to the report of the intergovernmental group on climate change (IPCC ) from 2013.



Where did the gaping gap between predictions at the end of the century and what happened in the past of the Earth come from? Does the planet’s climate past tell us that we have missed something?

Time

One big reason for breaking up is simple: time.

')

Earth needs time to react to changes in greenhouse gas content. Some changes last for years , others require a whole generation to reach a new balance. The melting of ice and permafrost, heating of the depths of the oceans, the formation of peat layers, the reorganization of vegetation cover — these processes take centuries and millennia .



Slow response of this type is not taken into account in climate models. This is partly due to the lack of computer capacity to calculate them, partly - because we concentrate only on what will happen in the next few decades, partly - because these processes are not 100% predictable. But, despite the fact that while climate models successfully predict the observed changes, uncertainties exist even for reactions that occur fairly quickly, such as cloud formation or increased warming at the poles.



The past of the Earth, on the other hand, shows us how climate change actually took place, summing up the entire spectrum of fast and slow responses of the planet. During past climate changes, during which the Earth had ice caps (like today), it usually warmed to 5 ºC - 6 ºC with each doubling of CO ₂ , while the whole process took about a thousand years . This is about twice the value of the “equilibrium climate sensitivity” ( Equilibrium Climate Sensitivity , ECS) used in climate prediction models up to the year 2100, which are calculated mainly based on historical observations.





“Everything that happened to us is just a prologue” - an engraving on the building of the national archives in Washington, DC



“We really expect that the system sensitivity of the Earth (change the CO ₂ and all systems - ice caps, plants, methane levels, aerosols, etc.) will react to this will be higher than ECS. Our study of the Pliocene says that it is about 50% higher, although this is not the limit, ”Gavin Schmidt, director of the Goddard Institute for Space Research at NASA in New York, told me.



Or, as Dana Royer of Wesleyan University said: “Simply put, climate models usually underestimate the extent of climate change relative to geological evidence.”



Partly for a higher level of change is responsible for the Earth’s slow-reacting systems responsible for global warming. Even if absolutely all greenhouse gas emissions stopped tomorrow, the sea level will rise for many centuries due to thermal expansion and melting of glaciers ; the ice caps of the Antarctic and Greenland will also continue to melt due to the temperature already accumulated by the climate over several decades. And since CO ₂ remains in the atmosphere for a long time , in the absence of geoengineering solutions for its removal, the world will overcome any temperature limit set for the end of the century, and it will remain high for several hundred more years.



But this does not fully explain the gap, which means that we do not take into account yet any reinforcing feedback. As stated in the US National Climate Assessment 2017 : “the discrepancy between the models and the data on past warming suggests that climate models miss at least one, and maybe more, process that is critical for future warming, especially in the polar regions.”

Can the Miocene tell us the future?

The climatic optimum of the mid-Miocene (Mid-Miocene Climate Optimum, MMCO) was an ancient climate warming, during which CO ₂ levels jumped from less than 400 ppm to 500 . The antiquity content of CO ₂ is measured by various indirect methods , such as the content of boron and carbon isotopes in fossil and ancient soils, or in pores on fossil leaves. The jump was caused by a rare volcanic phenomenon , a “large pyrogenic province, ” during which huge amounts of basalt were erupted to the surface in the west of the modern US 16.6 million years ago . Yvette Eley and Michael Hren [Yvette Eley and Michael Hren] from the University of Connecticut studied how this affected the climate.



They used tools such as fat molecules left in sediments from plants and microbes that lived at the time. Eley and Horseradish extracted Miocene microbial chemical residues from the mud of that period in Maryland, and then converted the percentage of various fatty molecules to soil temperature using calibrations based on more than a decade study of microbial fats in modern soils from all over the planet. “Certainly, the time of occurrence of these basalt streams and the time of climate change are very closely related,” said Eley. - Our biomarkers definitely track the behavior of CO ₂ . Whatever caused changes in the planet's ecology system, it definitely followed pCO _2. ”



But among the various examples of climatic fluctuations, MMCO was very mild compared to the end of the Permian period, the Triassic period, and other events associated with mass extinctions . CO ₂ emissions in the Miocene were slow enough to avoid significant ocean acidification , unlike today or extreme examples from the past.



In a similar way, they calculated the sea temperature using chemical remains of marine microbes: “We obtained a relative change in sea-surface temperature during the MMCO by 4-5 degrees - and then the sea temperature was 6 degrees more than today,” said Eley.

Warmer, wetter, drier?

They measured the atmospheric humidity in the Miocene , analyzing the chemical residues of the wax coating of the leaves of plants, calibrating them according to modern values ​​in various places on the planet. “If we use wax from the foliage, our biomarker, as an indicator of atmospheric humidity, then we conclude that in the middle of the Miocene, the atmosphere became more humid,” said Eley. - It is quite interesting to consider our work in the context of other reconstructions. The west of the United States has become drier, South America is wetter, parts of Europe are wetter, and other parts are drier. ”



Remote locations such as the east coast of the United States , the Pacific Northwest , western China , Patagonia , Central Asia, and Atacama in South America have become much more humid, leading to increased global erosion . As a result, there was an expansion of the forest area and their compaction . Interestingly, in North Africa or Asia there were no signs of deserts , and now we have the Sahara and Gobi deserts.



The widespread humidification of the atmosphere and the landscaping of the surface do not coincide with the predictions of the future that are being made for the current situation - according to these predictions , those parts that are now wet will become more wet, and the dry ones will be even drier. The difference may be that our climate change occurs very sharply compared to a much slower change in the Miocene.



Although there were a lot of forests on the planet before the middle of the Miocene (unlike today, reflecting the deforestation process, which was facilitated by people who lived between the ice ages for several millennia ), the warming in the Miocene led to clearly observed changes in vegetation throughout the world, fossil species, especially in the form of petrified pollen.



In most parts of Europe, subtropical plants have replaced plants that have adapted to the cold, and dense forests with an abundance of marshes filled the shores and river deltas on the territories of modern Denmark and Germany (then the coastline of Europe was 190 km deeper towards the ground than today). These swamps accumulated brown coal , today providing a quarter of Germany’s electricity generation. Spain resisted the tendency to humidify with a hot and dry climate in the south and a warm and humid climate in the north, just like today, and experienced long dry seasons .





Life in the middle of the Miocene on the territory of modern Spain in the artist's view



Judging by the European plants, between seasons the temperature difference was less .



In Siberia, it rained 3-5 times more often than today, and swamps in the east of Russia also accumulated coal. In Arctic Canada , where today there is a tundra with permafrost and without trees, in the middle of the Miocene low-temperature forests of birch, elm, holly and umbrella pine were replaced by high-temperature forests, where beech, hazel, amber, walnut and linden grew.



Near the equator, early elephants and antelopes walked along the grassy and humid Arabian Peninsula , and northern Africa was covered with forests where today sand dunes are moving. Great apes have spread across a wooded planet , and it was then that our ancestors, the hominids, separated from other apes.



But the Antarctic has changed the most.

Rising sea level by 40 meters

Thirds to one-third to three-quarters of Antarctic ice melted . Tundra and forests consisting of beech and coniferous trees appeared on the ground free from ice, which could not have been had the Antarctic summer been warmer than 10 ºC (it is much warmer than today's -5 ºC ). It is not clear what Greenland did, but in its north there could be a small ice sheet , which melted quite heavily.



As a result, sea levels rose by as much as 40 meters . Today, it would significantly move the coastline inland to the continents, and flood the densely populated regions in which a fourth of all people on the planet live.



40 meters is just a little more than the latest predictions of sea level rise for the near future: up to a meter by 2100 and up to 1.6 meters (when 5% of the world's population will be under water) by 2300, provided that we stabilize the warming by level around 2 ºC. The difference is only in time scales. According to the US National Climate Assessment of 2017, 2 ºC of warming will result in the loss of 3/5 parts of Greenland ice and one third of Antarctic ice, which will lead to a 25 m rise in sea level - however, over 10,000 years.



And yet, information from the Miocene suggests that the current sea level rise may be stronger and faster.



The coastal sediments of the eastern Antarctic demonstrate that its ice was extremely sensitive to even small changes in the level of CO ₂ and fluctuations of the orbit in the Miocene, and could melt rather quickly . How fast? Edward Gasson of the University of Sheffield in Britain calculated that Antarctica could initially raise sea level by about 2.5 m every hundred years, and then this process slowed down, and over 10,000 years the level was 30-36 meters higher. This speed coincides with the estimates of Robert Deconto of the University of Pennsylvania and David Pollard of Amherst College, made on the basis of the Pliocene, whose climate was cooler than in the middle of the Miocene, and sea level "only" 20 m higher than today. Deconto and Pollard suggested that modern warming of 2.5 ºC by 2100 would raise sea level by 5.7 m by 2500, a year by about 1.2 m per century. This rapid change may seem radical, but we know that periodically over the last 500,000 years the sea level has risen by 4–5.7 m every hundred years .



If the modern rise of the sea level turns out to be similar to the one that was in the Pliocene, 1.2 m in a hundred years, or in the Miocene, 2.4 m in a hundred years, and not like the IPCC - half a meter per century, then our future will be completely different. Rising sea levels, exacerbated by tidal flooding and storms , will make a huge amount of coastal infrastructure and possessions useless in a couple of generations.



And until now, computer models did not support such a high melting rate of ice.



Melting of ice under the influence of the ocean, which destabilizes and flushes glaciers, was critical for the Miocene , and seems critical today . This process can trigger self-sustaining instability of sea ice sheets, and glaciers will begin to retreat into the earth because of the bowl shape of Antarctica. The deeper the ice will climb, the faster it will melt due to pressure, and thinner glaciers will emerge, so they will retreat even faster until they form high peaks that will break under their own weight, which will further deteriorate the situation. And this process in Antarctica has probably already begun .



Another melting accelerator is water melting on the surface , which requires reaching temperatures above the freezing point. It penetrates into cracks, freezes, and splits ice like a wood splitter - this phenomenon was observed when the Jacobshavn glacier disappeared in Greenland. And today, surface melting occurs in parts of the Antarctic. Such processes that increase melting have only recently been added to new computer models, and now they show that our descendants will probably see the rates of sea level rise observed in antiquity.



The retreat of ice intensifies warming, as a bright, reflective surface is replaced by a dark, heat-absorbing water and earth. As a result, temperatures will slowly rise further.





How could the Antarctic ice sheet look like in the Miocene, from 14 to 23 million years ago

Hope for uncertainty?

Can the gap between the Miocene climate and our expected future exist simply due to the lack and inaccuracy of the ancient climate data?



“Changes in the level of CO ₂ in the average Miocene may exceed the calculated median value. About other factors nothing is known at all. Levels of methane or N ₂ O are not defined. “The amount of ozone or soot (appearing after fires or as a result of plant life) is also little known,” Gavin told me. “Therefore, even if we had ideal indicators of global temperature (and there are none), the sensitivity estimates obtained by simply dividing the temperature by the CO ₂ level cannot be compared with today's ECS estimates.”



And yet, despite the variation in the level values, they tend to accumulate around the value of 500 ppm for the Middle Miocene. Some studies even suggest the possibility of a lower level of CO ₂ , which has led, nevertheless, to higher temperatures. The picture of the warm climate is supported by geological evidence of high sea levels and fossils found throughout the world, including the seabed near the shores of the Antarctic.



Has the climatic optimum been increased due to cyclical orbit changes? Although the individual Miocene glacial cycles depended on orbital oscillations, as was the case during the last glacial period, warm weather and maximum ice retreat persisted over several orbital and glacial cycles, along with higher levels of atmospheric CO ₂ . So we can not hang the increase in optimum only on the orbit of the Earth around the sun.



Even more confusing is the fact that the beginning of the Miocene was different from today. The climate of the early Miocene was warmer than our pre-industrial times, then there were fewer grass-covered areas, and the oceans communicated with each other in a different way . The current from the Pacific to the Atlantic went where Panama is now located, and the Bering Strait was closed. However, scientists believe that these currents, perhaps, not so much influenced the climate , and in many ways the planet was very similar to today.



So, there are big uncertainties in how well the situation in the Miocene describes the future of our descendants. And, of course, at least in the last 66 million years there have been no processes similar to such a high rate of emissions into the atmosphere. On these grounds one can justifiably refuse to compare the situation with any ancient counterparts. It is only necessary to remember that uncertainty is a double-edged sword: it can work not only in a more favorable direction for the evaluator.



If all this seems too depressing to you, then know that there is hope! It lies in the low response rate of the Earth, which opens up a small window of opportunity for us.

Hand on fire

If you quickly hold your hand through the flame of a candle, you will not burn yourself. The same principle works with the Earth - if we minimize the time that the planet spends under the influence of temperatures exceeding pre-industrial temperatures, then perhaps we will be able to avoid rising ocean levels comparable to those in the Miocene.



Although the ice of Greenland and western Antarctica is already melting with acceleration , the eastern Antarctic — so far — remains in a relatively stable state (with the exception of Totten Glacier ). So if we keep warming well below 2 ºC, the Deconto and Pollard models suggest that East Antarctica will not make a significant contribution to sea level rise in the future.



But for this we need to reduce the concentration of greenhouse gases, and exceed the plans of the program Net Zero .



" Negative emissions " (active absorption of CO ₂ from the air) can slowly reduce global temperatures and stabilize many sea level rise factors in the 22nd century. According to Mathias Mengel from the Potsdam Research Institute for Climate Impact, a drop in CO ₂ levels will eventually allow Antarctica to begin to slowly accumulate ice, so sea levels will begin to fall again after about three hundred years .



But this assumption will be true only if negative emission technologies can be deployed on a large scale already by 2030 — this is a scenario with " limited realistic potential ." Every five years, implementation delays doom our descendants to an additional meter of sea level by 2300. Also, such a scenario implies that in the process of combating warming we will not trigger a large-scale collapse of ice sheets . Otherwise, this process will become irreversible on a scale of several millennia, even if we manage to remove CO ₂ from the atmosphere .



Our current window of opportunity will not remain open for a long time - scientists are trying to understand whether the collapse of ice sheets has already begun at one of the largest glaciers in western Antarctica. “Everything changes very, very quickly compared to everything we found in geological records,” says Eley. “I would really like to believe that we will not have one of the worst scenarios in our hands, but it seems to me that we are already on our way to these levels [CO ₂ ].”



“In the middle of the Miocene, the level of CO ₂ rose to 100-200 ppm. Since the beginning of the industrial age, we have already reached a 127 ppm increase. So we have already half gone this way, - said Fuck. “The uncertainty lies not only in what levels of CO ₂ we end up with, but also in how the system responds to such rapid changes.”