An early start to the Anthropocene

The large-scale burning of fossil fuels is not the only way to affect global climate; land use change can also be very important. Clearing forests releases the carbon locked up inside the plants and land use changes can potentially release large amounts of carbon from organic matter in the soils. In addition, altering vegetation changes the land's albedo (the amount of solar energy reflected back into space) which affects climate, as does the extent of wetlands as this is correlated with the amount of the greenhouse gas methane released into the atmosphere. As all of these things are affected by agriculture, it raises the possibility that as agriculture spread around the world it may have started to affect the atmosphere and climate on a global scale, long before the industrial revolution. While it sounds plausible that land use change associated with agriculture may have had global effects, what is needed to really establish this is a rigorous—preferably quantitative—approach to the question. Recent increases in our knowledge of past atmospheric chemistry (especially from gases trapped in ice cores), modern computer models of climate, and our increasing knowledge of environmental archaeology start to make this a realistic prospect, and William Ruddiman has recently attempted such a quantitative approach in a series of provocative and controversial publications.82-84

Most of us were taught at school that a very powerful approach to science is the use of replicated experiments—in which treatments (usually involving the change of one variable) are repeatedly compared with 'controls', that is, other experiments where (typically) no changes had been made to the system. Ideally, then we would rerun the past 13,000 years of Earth history multiple times, in some cases with, and in others without the development of agriculture and study the average effect on the atmosphere and climate. Clearly, this is impossible—indeed the inability to follow this experimental approach is a major problem whenever we try to gain a scientific understanding of past events. Ruddiman has attempted to address these problems by using previous interglacials as a control for our current post-glacial period and by running experiments with computerized climate models.

The recent geological history of the Earth has been marked by repeated swings between cold and warmer conditions—the glacial/interglacial cycles. The number of such swings depends on your exact definition of 'glacial' and 'interglacial'; however, it is now clear that there were at least 40-50 such transitions over the course of the geologically recent past83 (about 2.5 million years). This geological period is usually referred to as the Quaternary (although the correct name for this time period is currently a matter of controversy amongst geologists85). The Quaternary is best considered as starting 2.6 million years ago, associated with evidence for the occurrence of glacial ('ice age') conditions, and running up to the present day—although some scientists argue for a start date of 1.8 million years ago. The most recent part of the Quaternary, since the end of the last glacial and marked by the rise of agriculture, is called the Holocene which started around 11,600 calendar years ago (technically this is what geologists refer to as a series or epoch; the Pleistocene is the other series which collectively comprise the Quaternary).85

Before we continue with examining the effects of land use changes, it is necessary to give a little more background on glacial events. The idea that there had been 'ice ages' in the past gained currency during the first half on the nineteenth century, based on observations of the effects of former ice sheets and glaciers etched on the landscape of Europe. The obvious question was what causes these repeated glaciations? In 1842, Joseph-Alphonse Adhemer suggested that the cause might be astronomical, with ice ages driven by changes in the Earth's orbit affecting the amount of solar energy arriving on Earth.86 These ideas were further developed by James Croll during the 1860s. At this time, Croll was working as a janitor in a Scottish college—an injury had forced him to give up his work as a carpenter.86 He must have been far and away the most mathematically gifted janitor in Scotland, if not the world, and by 1870 he had published several papers on his work and convinced the eminent and influential Quaternary geologist James Geikie that the astronomical explanation was correct. Geikie championed Croll's ideas in his seminal book The Great Ice Age.87 These ideas were further developed by Milutin Milankovich at Belgrade University in the early twentieth century, a task that involved spending years on detailed calculations which could today be done in at most a few days with a computer. Indeed the changes in aspects of the Earth's orbit that are thought to drive glacial/interglacial cycles are now often referred to as 'Milankovich cycles' in his honour86 (see also Chapter 5 on the role of Milankovich cycles in mediating speciation rates). By the mid-1970s, these ideas were put on a much firmer foundation as it was shown that the predictions from astronomical cycles broadly matched changes in temperature reconstructed from ocean sediment cores.88

The repeated glacial/interglacial cycles provide one potential way to sidestep the impossibility of replicated experiments on the global effects of agriculture. In 2003, Ruddiman82 pointed out that the trends for carbon dioxide (CO2) in the Holocene were different from those reconstructed for the previous three interglacials, in that previously CO2 levels had fallen throughout the interglacial while in the Holocene they had started to rise over the past 8,000 years. Previously, working with one of his undergraduate students, he had pointed out that the methane (CH4) trends for the Holocene also appeared to behave oddly, unexpectedly starting to rise around 5,000 years ago.89 CO2 and CH4 are key greenhouse gases and Ruddiman even suggested that their rise had prevented the onset of the next glaciation.82,90 In brief, his suggestion is that the increase in CO2 has come from land clearance for agriculture while the CH4 has come from the expansion of wetlands associated with rice cultivation—we will discuss these mechanisms in more detail later.

One problem with comparing current glaciations with past glaciations is that not all glacial/interglacial cycles are the same: there are subtle differences in the variations in the Earth's orbit which makes the details of the climate different from cycle to cycle. A criticism of Ruddiman's approach has been that in using the previous three interglacials as a 'control' for the Holocene he had not been comparing like with like. An earlier interglacial—known as marine isotope stage (MIS) 11—is supposed to be the closest analogue to the Holocene in various orbital parameters.91 This interglacial lasted much longer than the current length of the Holocene, thus apparently undermining Ruddiman's suggestions that human changes in the levels of greenhouse gases over the past 8,000 years had prevented the onset of glaciation in parts of Canada. However, data from pollen grains preserved in sediments from a very long series of cores of lake mud from France show that MIS 11 was not the best match for the Holocene for European vegetation patterns,92 so illustrating the difficulty in identifying the best interglacial to use as a control. Despite this, as Ruddiman93 and others92 have pointed out MIS 11 provides some support for Ruddiman's ideas because it also fails to show the increases in greenhouse gases during the interglacial that appear anomalous in the Holocene.

Ruddiman82,83 outlines a range of mechanisms by which humans could have caused an early start to the Anthropocene. He suggests that the most plausible explanation for the rise in CH4 is an expansion of rice growing in irrigated fields. Rice was domesticated in China between 10,000 and 9,000 years ago94 but growing rice in flooded fields appears to be a later phenomenon becoming common from around 5,000 years ago.83 The low-oxygen conditions in wetland sediments—including the artificial wetlands of flooded rice fields—provide good habitats for methane-producing bacteria. Indeed, the predicted increase in rice cultivation over the next 40 years is seen as a potential problem in the context of global warming.95 Rice cultivation looks a particularly promising source of this methane as there is evidence from comparing Arctic and Antarctic ice cores that the main source of this extra methane was in the tropics—if, for example, it had come from the extensive high-latitude peatlands of the northern hemisphere, then it should have been more prominent in Greenland ice cores compared to ones from the Antarctic.83 In addition to rice cultivation, Ruddiman suggests that increased numbers of livestock—which harbour methane-producing microbes in the low oxygen parts of their guts—and burning of forests will also have tended to increase atmospheric methane levels. His main mechanism for increases in CO2 is the destruction of forests as agriculture expands. For example, as agriculture spread across Europe the amount of tree cover started to decrease quite dramatically96 (Fig. 9.4).

Ruddiman's calculations suggested that human-derived changes before the industrial revolution have added around 40 ppm (parts per million) of CO2 to our atmosphere. However, the magnitude of this change appears problematic to many other climate modellers. On a millennial timescale, around 85% of any atmospheric CO2 increase is absorbed by the oceans—this means that for an atmospheric increase of 40 ppm around 566 ppm CO2 needs to be emitted through landscape and vegetation changes. This seems implausible based on ice core records and computer climate models, which suggest that the human impact was more likely to be on the scale of 4-6 ppm CO2 (after equilibrium with the ocean had been obtained).97 Ruddiman83 now accepts that his initial estimates of the amount of CO2 added to the atmosphere may have been too high. He now views the role of humans as a two-stage process with a more modest direct effect of humans on CO2 level causing a warming, which affects factors such as sea ice preventing 'natural' drops in CO2 levels that have happened in previous interglacials. Initial attempts to model the effects of land use changes associated with the rise of

Figure 9.4 Langdale, in the Lake District National Park in northwest England. The 'wild' looking scenery is deceptive. The treeless mountain slopes are not natural; before forest clearance by early prehistoric farmers much of this landscape was forested.118 During the early Neolithic, approximately 4000-3000 BC in Britain, the Pike O'Stickle (the highest summit seen in the photograph) was a source of stone for making axes which were transported all over Britain. Photo: DMW.

Figure 9.4 Langdale, in the Lake District National Park in northwest England. The 'wild' looking scenery is deceptive. The treeless mountain slopes are not natural; before forest clearance by early prehistoric farmers much of this landscape was forested.118 During the early Neolithic, approximately 4000-3000 BC in Britain, the Pike O'Stickle (the highest summit seen in the photograph) was a source of stone for making axes which were transported all over Britain. Photo: DMW.

agriculture over the past 6,000 years and their effects on the carbon cycle also produce values lower than Ruddiman's initial estimates.98

Ruddiman's basic idea appears plausible—although for the reasons given earlier, the effects of humans may be less than he originally estimated, and it is currently an open question if they were large enough to have had a substantial impact on the Earth's climate before the industrial revolution. However, there are a number of potential issues that have not yet been addressed in a quantitative manner which are highly relevant to Ruddiman's ideas. First, as several people have recently pointed out99,100 the role of carbon in soils needs to be considered. In general, agriculture reduces the amount of organic carbon stored in soils101; this suggests that Ruddiman's estimates of the amount of CO2 produced by agricultural change may be an underestimate. That said, in some cases, clearance of forest can increase the amount of carbon stored in soils, as in parts of the uplands of Britain where forest clearance led to the formation of peat 'soils', which are almost 100% organic matter.102 A further complication is that increased amounts of organic carbon in soils can lead to a rise in microbially produced nitrous oxide, which is itself a greenhouse gas.103 However, on balance, incorporating the effects of agriculture on soils probably makes Ruddiman's hypothesis more likely— although better quantitative models addressing the question are needed before firm conclusions can be drawn.

Certainly, soils have been important in the context of more recent agricultural change. In a study considering only the period between 1860 and 1980, Houghton and colleagues104 estimated that recent land use changes had led to a significant addition of CO2 to the atmosphere from soils—for example, they estimated that in 1980 as much carbon was lost from soils worldwide as from the burning of forest trees.

At least two other problems require consideration for a full test of Ruddiman's hypothesis, that humans made an early start to altering the atmosphere as a by-product of their actions. The first is the effects of land use change on albedo. Clearing forest adds CO2 to the atmosphere and is a central mechanism in Ruddiman's hypothesis. However, in many cases, tree cover absorbs more solar radiation than grassland and so increases the Earth's temperature—for example, Gibbard and colleagues105 calculated that albedo changes caused by the replacement of the world's vegetation by trees would lead to a global warming of 1.3°C, while replacement by grasslands would give a cooling of 0.4°C. Because of the large amounts of water they transpire, tropical forests tend to lead to a net cooling, but at higher latitudes this transpiration does not offset the albedo effect in this way. These albedo effects have implications for Ruddiman's suggested feedbacks between forest clearance and sea ice. Clearing forest will release CO2 and so cause warming (and ice melting) but unless it was tropical forest that was cleared then Ruddiman's current calculations probably overestimate the warming effects as he does not factor in vegetation albedo.

The final complication we will mention applies to methane—a potent 'greenhouse' gas. In 2006 Keppler and colleagues106 caused considerable surprise with a paper in the prestigious journal Nature suggesting that plants produce large amounts of methane— the main 'textbook' source of methane is from microbes in oxygen-free conditions (such as the rice fields and animal guts described earlier). The great surprise was that this new methane source appeared to be produced in the presence of oxygen. If this is correct then it provides an added complication to Ruddiman's hypothesis as forest clearance could reduce methane production; yet, as Ruddiman points out, it shows a rise during the Holocene. It is currently unclear how the work of Keppler et al.106 should be viewed; however, the first published independent test of their ideas has failed to find plant-produced methane,107 so such methane may not be an important complication in discussions about early starts to the Anthropocene.

In summary, it seems likely that for thousands of years before the industrial revolution humans have been having an effect on the concentration of greenhouse gases in the atmosphere, but the magnitude of this effect (e.g. 4 ppm or 40 ppm increase in CO2) is currently unclear. As agriculture started to spread around the world, the impact of humans started to become global but it has accelerated tremendously in the past few hundred years. Ruddiman has also suggested that some of the 'wiggles' in the CO2 curve can also be explained through the effects of disease on human population size in the past. We now consider this and the related question of early modification of tropical forests—including some which have, until very recently, been considered relatively pristine ecosystems by many conservationists.

Solar Power

Solar Power

Start Saving On Your Electricity Bills Using The Power of the Sun And Other Natural Resources!

Get My Free Ebook

Post a comment