RealClimate logo

The Early Anthropocene Hypothesis: An Update

Filed under: — mike @ 15 March 2016

Guest post from Bill Ruddiman, University of Virginia

For over a decade, paleoclimate scientists have argued whether the warmth of the last several thousand years was natural or anthropogenic. This brief comment updates that debate, also discussed earlier at RC: Debate over the Early Anthropogenic Hypothesis (2005) and An Emerging View on Early Land Use (2011). The graph below outlines the evolution of that debate through 4 phases.


In phase 1 (the 1900’s), scientists viewed Holocene climate change as driven only by natural causes until the industrial era began. But by the late 1990’s, ice core data revealed late Holocene GHG rises unlike trends in previous interglaciations. Two hypotheses proposed natural causes for the CO2 increase: carbonate compensation (Broecker et al., 1999, 2001) and coral-reef construction (Ridgewell et al., 2003).

In phase 2 (2001-2003), the early anthropogenic hypothesis (EAH) challenged natural explanations for the anomalous late Holocene CO2 (and CH4) rises, attributing them to the spread of early agriculture thousands of years ago.

In phase 3 (2004-2008), several arguments were advanced against the EAH:
* too few people lived millennia ago to have had a significant influence on land clearance, GHG emissions and climate;
* a (proposed) interglacial stage 11 analog for the Holocene suggested that thousands of years of natural warmth still remain in the current interglaciation;
* the weak decrease in ice core δ13CO2 during the last 7000 years did not permit extensive deforestation which would have released abundant 12C -rich carbon.
Papers by myself, my co-authors at Wisconsin, and others during phase 3 rebutted some of these criticisms, but community opinion remained divided.

Phase 4 (2009-2016) has seen a major shift in viewpoint of published papers: 30 papers favor aspects of the EAH, 6 papers oppose it, and 5 are in the middle. Most of the phase 4 papers that oppose the hypothesis or are ‘in the middle’ are based on modeling studies. Many of the 30 supporting papers are broad-scale compilations of archaeological and paleoecological evidence:
* The average GHG trends from 7 previous interglaciations show CO2 and CH4 decreases, in contrast to the late Holocene increases;
* Interglacial stage 19, the closest Holocene analog, shows decreases in CH4 and CO2, and the CO2 decrease closely matches the 2003 EAH prediction;
* CH4 emissions from Asian rice paddies account for 70% of the observed CH4 rise from 5000 to 1000 years ago
* historical data show that early per-capita land use was at least 4 times larger than assumed in several phase-3 land use simulations
* a recent land use simulation based on historical evidence accounts for more than half the CO2 anomaly originally proposed in the EAH;
* pollen evidence shows nearly complete deforestation in north-central Europe before the industrial era began;
* δD and δ18O trends show anomalous late Holocene warmth compared to cooling trends in prior interglaciations, in agreement with A-OGCM simulations of the warming effect of the anthropogenic CO2 and CH4 trends.

Anyone seeking more detail on this issue should contact for pdf copies of the recent 2016 Ruddiman et al. paper in Reviews of Geophysics and an invited paper just submitted to Oxford University Press that summarizes the history of this debate, with full references to the papers shown in the table.

41 Responses to “The Early Anthropocene Hypothesis: An Update”

  1. 1
    Bill Ruddiman says:

    My mistake. John Mashey should have been a co-author on this post. He has been suggesting for years that I put together some kind of timeline graph like this. So this graph exists thanks to him.

  2. 2
    Charles Raguse says:

    The graph is not readable.

  3. 3

    Thanks, very interesting.

    One nit–the table is pretty much unreadably fuzzy. (Needs a higher-res original, I think.) But we get the gist.

    [Response: Fixed – Mike]

  4. 4
    Birte Boekel says:

    Thank you, that is very interesting – also a helpful overview for me as a science teacher. Naturally, my students ask me about the anthropogenic hypothesis over and over again, so I’m really grateful up-to-date, reliable information.

  5. 5
    John Mashey says:

    re: #1 Bill gives me too much credit – all I did was hassle him to do this, and help format the chart.

    I’ve been following this for a decade as an example of a *real* argument among credible *real* scientists that illustrates the progress of *real* science in action, as hypotheses are offered, challenged, evolve, bring in other researchers, respond to new data, inspire people to relook at old data, etc. One of these days, I’ll write that up as an observer, as I think it makes a really nice history-of-science example.

    I do think the chart makes a clear point: the evidence has piled up, most uncertainties have been closed enough.

  6. 6
    James McDonald says:

    I hav a naive question about pre-industrial land-use.
    I know that slash-and-burn farming has been common around the world for a long time.
    I also know that native tribes in Northern California would regularly burn the areas around oak trees, both to minimize the introduction of other species and to kill off the insect-infested acorns which tended to fall first. (The goal being a reliable harvest of acorns, a food staple.)
    So the question is simply whether we have any kind of certainty about the variety, range and regularity of anthropogenic wild-fires. Could ancient peoples have done more burning than we are aware of, for reasons we might not have guessed, or would such activity be clearly marked in the soil records?

  7. 7
    Hank Roberts says:

    Dr. Ruddiman, a tangential question — have you thought about whether, in previous ice/warm cycles, some species (likely planktonic) suddenly expanded its range or niche and contributed to some change in rate or pattern of climate change beyond what the astronomical cycle would be expected to cause?

    Seems to me this may be a repeating pattern, where something finds an opportunity as the ice rolls away and temporarily dominates the planet.

    I recall one of those cycles (I’ve forgotten which) saw the expansion of some varieties of plankton from their original successful habitat in shallow coastal waters — some of them found a way to make a good living in the upper photic zone of the deep ocean, a niche that hadn’t been filled previously.

    I don’t know of anyhting more complicated — did the grasses perhaps develop their ability to spread widely during such an opportunity?

    Ecopoiesis, I know. But it does make me wonder.

  8. 8
    Mal Adapted says:

    Hank Roberts:

    did the grasses perhaps develop their ability to spread widely during such an opportunity?

    Hank, are you perhaps thinking of Ellen Morris Bishop’s discussion of the evolution of C4 grasses and the spread of grasslands in the late Miocene, in In Search of Ancient Oregon? She cites Retallack 2001 (yay, no paywall), who argues for positive feedback between enhanced carbon sequestration by grass-dominated ecosystems and global cooling due to relatively low atmospheric CO2.

  9. 9
    Oale says:

    “#6 James McDonald Could ancient peoples have done more burning than we are aware of, for reasons we might not have guessed, or would such activity be clearly marked in the soil records?”

    Historical records (1500-1850) say for f.e. Finland that slash and burn was pretty regular near the settlements located in the eastern Finland. The settlements were regularly on south-facing slopes and the burns would have happened. The wash out from such fields would have ended up in the bogs or lakes under them. I’m not aware of specific studies of this relationship, but the regular cycle of slash and burn on this sort of hilly country would have been c. 40 years as the soil gets pretty poor in about 5 years of s&b agriculture, and the young trees would then felled for firewood and their stumps and branches burnt in s&b. The natural fires would likely have also been more common in these sorts of locations, the remaining old-growth forests in Finland are mostly on a moist ground, which is where the record of natural forest fires is obtained….

    So there might be some under-estimation of the extent of s&b, but I can’t come up with any other motives for burning forest, firewood would have been such an asset at least in the colder climates I see no reason why there would be any extra burning of the forests.

  10. 10
    Hank Roberts says:

    In left/red column, some of the papers are shown in italic font.
    What does that indicate?

  11. 11

    #7–For those of us who didn’t know:

    “Ecopoiesis: The artificial creation of a sustainable ecosystem on a lifeless planet.
    See also

  12. 12
    Terry Miesle says:

    RE 6:
    Native Americans burned prairies and woodlands pretty consistently. Periods vary, but tree ring data shows anywhere from 3-year cycles to longer cycles. Likely this started after observations from natural fires, and the plant life returning was probably more beneficial to the Native American population than light woods. Certainly the grasses and forbs were good forage for prey species.

    Returning America’s Forests to Their ‘Natural’ Roots
    Keith Kloor*
    Science 28 Jan 2000:
    Vol. 287, Issue 5453, pp. 573-575
    DOI: 10.1126/science.287.5453.573

    Fire is a very important factor in biodiversity in prairie environments.

    Ecological Determinants of Species Loss in Remnant Prairies
    Mark K. Leach, Thomas J. Givnish
    + Author Affiliations
    Science 13 Sep 1996:
    Vol. 273, Issue 5281, pp. 1555-1558
    DOI: 10.1126/science.273.5281.1555

    I’d have do to some more digging for additional articles. It’s pretty obvious to the researchers that fires were used for a few purposes, most importantly to maintain the mixture of woodlands and prairies.

  13. 13
    Richard Hawes says:

    Reply to #5, Dr John Mashey:
    “One of these days, I’ll write that up as an observer, as I think it makes a really nice history-of-science example”.
    Don’t leave it until it’s too late!
    Naomi Oreskes interviewed many, but not all, of the key players involved but not all for her book “Plate Tectonics – an insider’s history”. Tuzo Wilson and several other key players had already died.
    Start interviewing NOW!

  14. 14
    Hank Roberts says:

    > ecopoiesis
    More generally — it’s Lovelock’s word originally — it refers to the way [some form of] life changes something making the world more comfortable [for itself].

    Of course more comfortable for one particular kind of life may change the world in ways that hurt most of the rest of life, see e.g. the oxygen catastrophe or today’s CO2 change.

  15. 15
    John Mashey says:

    10: Hank
    Italic : authors include Bill, others don’t.

    One of the indicators for a hypothesis gaining real traction is that other author groups publish papers supporting it from different perspectives and with different kinds of evidence.

  16. 16
    Digby Scorgie says:

    If humans have been affecting the global climate since the inception of the Holocene, would this imply that the Holocene and the Anthropocene are one and the same?

  17. 17

    #16, Digby–Maybe. It’s already a suggestion before the relevant committee of the International Stratigraphers:

    There’s a relevant chapter in “The Sixth Extinction”, wherein we “meet” the chair of the “Anthropocene” committee. A vote on formalizing the term is expected this year.

  18. 18
    Bill Ruddiman says:

    My responses to previous comments up to #16:

    #10, #15: John is right — italics indicate papers on which I am author or co-author, to show how much my own publications contribute to the increase in ‘supportive’ papers during the 2009-2016 interval. There are 30 supportive papers listed for those 7+ years. I am author or co-author on 13 (4 as first author), and not an author on the other 17.

    #6, 9, 12: Humans have set fires since long before the Holocene, some of which ‘escaped’ and burned large areas. Because it is difficult to distinguish fires caused by lightning strikes from those started by humans, there is uncertainty about how large the earliest human contributions were. When populations were low, and people were hunter-gatherers and constantly on the move, forests likely had time to grow back before being cleared and burned again, leaving little if any net impact on atmospheric CO2. Later, as populations grew due to agriculture, that was no longer the case.

    #7, 8: You seem to be talking about evolutionary changes, which were minimal over the last 7,000 years. As for grasses, the major change in recent millennia that I can think of was the response in the relative abundance of C3 and C4 grasses because of weakening of the north tropical monsoons.

  19. 19
    Digby Scorgie says:

    #17 Kevin
    Thanks. That’s really interesting, and it’ll be interesting to hear what the final decision is.

  20. 20
    Phil mattheis says:

    Re: #7, #8, #18 and grassland evolution:
    Ooh, ooh, I know (waving hand in back of room)!
    There’s lots of literature on this topic, most of which I haven’t looked at for several lifetimes (and, admittedly, nothing recent).

    (Flagrant self-promotion warning) Mattheis etal, Annals of Botany 1970, reported looked at impact of lemming grazing on the various plants in arctic coastal plain communities (Barrow, AK). Grasses survive repeated cutting much better than other similar plant types (reeds, sedges) and with less cost to the target plant than shrubs and other plant types that invest in woody stems, etc. Most or all grasses have evolved strategies to survive and thrive under conditions of cutting, grazing, burning, etc, often as classic “domesticated” species in human agriculture history.

    “C3” and “C4” types refer to specific photosynthesis mechanisms, that mostly impact seasonality. C3 types are cool season, C4 warm season preference. For grasses, the C3 vs C4 distinction may be a relevant point in a warming world, as local microclimates see changed proportions of competing grass types due to temperature tolerances, in addition to the costs of having your expendable parts eaten…
    For what’s worth, repeated burning to maintain a grassland for agricultural (grazing) use, may not leave much visible ash for later discovery, as rains quickly turn that detritus into nutrients for recycling.

  21. 21
    James McDonald says:

    As a slightly off-beat update to my question above about early use of fire, it appears that some raptors may also deliberately set fires to flush out prey…

  22. 22
    DP says:

    Have heard that this interglacial is thought to have a longer duration than others, and may last thousands more years without human intervention. According to Ruddyman it should be virtualy over now.What was the answer to this.

  23. 23
    Simon C says:

    As far as I remember it was Marie-France Loutre and Andre Berger (2003) who suggested that the current interglacial was likely to be particularly extended even in the absence of anthropogenic forcing. Some other authors (noteably Lisiecki and Raymo 2005) have disagreed. Interglacials over the last 800,000 years have tended to be around 10,000 years in duration, with some variation. We are 10kyr into this interglacial, so on that basis it would not be surprising if it were drawing to a close; in the absence of anthropogenic forcings, previous interglacial durations seem to have depended more or less entirely on the orbital configuration. “Our” interglacial, of course, is liable to be a bit different … and as Prof Ruddiman has pointed out, may already have been different for some time …

  24. 24
    Alistair Connor says:

    Loved the book, Bill. Are the film rights available?

    I’m only half joking. A decent science documentary series on the subject of the early Anthropocene would blow quite a few minds. Personally, when I came across your work a few years ago, it was one of those moments when a bunch of half-known and half-understood fragments came together and made sense. I am glad that the carbon budgets are starting to balance, and hypothesis becomes plausible theory, well on the way to becoming established science.

    I’m from New Zealand, which is of course the last major land mass colonized by humans. It occurs to me that what is now understood about the deforestation that began with human colonization, in the last 1000 years, is an excellent proof-of-concept for the impact of stone-age humans on a virgin landscape. Is anyone working on this with respect to the anthropocene issue?

    A good TV vulgarisation of the subject would be an excellent boost to global consciouscness of the fragility of the planet, and the urgent need to start undoing some of the damage before it’s too late.

  25. 25
    Alf says:

    So global warming, to an extent, is something we need ? Or freeze ?

  26. 26
    John Mashey says:

    22, 23 Take a look at Ruddiman, Kutzbach, Vavrus(2011). Stage 19 looks a better analog than Stage 11, but is still not perfect.

    I think at this point it is very hard to escape thinking that indeed humans affected CO2 and CH4 starting thousands of years ago.

    There may be some room for argument about the reglaciation, in the sense that it’s a different kind of argument, and there is still uncertainty.

    Although we know the solar insolation curves, to be absolutely sure that there would have been a reglaciation we need to know:

    1) To great accuracy the level of CO2/CH4 that *would* have existed absent humans, just before the Industrial Revolution.

    2) To great accuracy, how much CO2/CH4 humans added.

    I think the estimates of these are much better than they were, but I don’t think the uncertainties are reduced enough to be sure yes or no. I.e., it is still plausible that humans stopped a reglaciation before the I.R., but not proven to the extent the other hypotheses are. My personal guess is “likely.

    3) Of course, absent massive continuing volcanoes or nuclear winter, the I.R., has guaranteed no real reglaciation for tens of thousands of years, at least, as per David Archer’s “The Long Thaw”, and maybe hundreds of thousands.

    4) See 4 My of climate change. Only in the last few million years has CO2 been drawn down enough to enable the wild gyrations of temperature driven by Milankovitch plus GHG amplification. … But that’s over, for a long time.

  27. 27
    Bill Ruddiman says:

    Responses to comments 16-26:

    #16, 17, 19: I am not in favor of a formal geologic definition of the ‘anthropocene’, which I see as being transgressive both in time and space, building from a slow start many millennia ago.

    #20, 21: Thanks for your input on things I mostly did not know.

    #22: We are now at or near a minimum in high-latitude northern summer insolation, yet a new glaciation has not begun. This has been called a ‘skipped beat’. My argument for the last decade has been that this insolation minimum would by now have led to the start of a new glaciation, but greenhouse-gas emissions from early agriculture prevented that from happening.

    #23: Thanks, nothing to add to your accurate summary.

    #24: Thanks for your enthusiasm. So far, a few nibbles but no bites as regards to a film. To me, the story of the development and spread of agriculture across the arable parts of the planet should be of very wide interest.
    New Zealand is indeed a prime recent example of anthropogenic deforestation. As you likely know, most of New Zealand’s natural forests were burned within 100 years of the arrival of Polynesians in 1280, even though population densities were very low.

    #25: Early greenhouse-gas releases from agriculture kept climate from cooling to the point of starting a new glaciation, but the current warming trend is so fast and so large that it is an altogether different phenomenon (at a minimum, worrisome; at a maximum, alarming).

    #26: Thanks John. Nothing to add to your comments.

  28. 28
    Richard Caldwell says:

    Alf: So global warming, to an extent, is something we need ? Or freeze ?

    Richard: Not really. The magnitudes of the forcing are completely different. It’s kind of like saying you need a supercar because you need to get to the closet from your bed in order to get dressed. James Hansen says, “The size of continental-scale ice sheets is mind-boggling. Although thinner toward the edges, ice over New York towered several times higher than the Empire State building–thick enough to crush everything in today’s New York City to smithereens. But not to worry–even though we sometimes hear geoscientists talk as if ice ages will occur again, it won’t happen–unless humans go extinct. Forces instigating ice ages, as we shall see, are so small and slow that a single chlorofluorocarbon factory would be more than sufficient to overcome any natural tendency toward an ice age. Ice sheets will not descend over North America and Europe as long as we are around to stop them.”

  29. 29
    Digby Scorgie says:


    It wouldn’t bother me if the term “Anthropocene” were ditched. However, it does have the advantage of pinning the blame squarely on humans for wrecking the Earth’s climate!

  30. 30
    E.D. Hilsinger says:

    The significance of anthropogenic burning as “purposeful” indigenous behavior in North America and elsewhere has been a topic of archaeology/anthropology/history/forestry for some time. Search S.J. Pyne.
    H.J. Lutz in Alaska, 1950s


    Intentional and unintentional (or “unintentional” where concepts of liability/responsibility have sway) use of fire has been a hallmark of preagricultural, agricultural, and pastoral cultures for tens of millennia. These behaviors were likely based on a variety of decision making inputs including empirical, inferential and explicitly cultural or traditional reasonings. A really fascinating topic with implications for several disciplines more amenable to statistical inquiry than direct historic or ethnographic means.

  31. 31
    Dp says:

    About early man altering the climate what about excavation of peat bogs? They are a major carbon sink but people have been excavating them since early times. What effect would this have?

  32. 32
    Bill Ruddiman says:

    Reply to #31: From what I know, peat burning has been quantitatively significant in northern Europe for at least the last 2000 years, and maybe earlier. Also, coal burning in China has been important for at least that along, especially during an almost-industrial interval in the 1200’s in southern China, until Ghengis Kahn and his cousins ripped up the infrastructure. Using very simple assumptions (number of people multiplied by number of “bricks’ of coal or peat burned per family per year), I have estimated that each source might have added ~1 ppm to total pre-industrial atmospheric CO2 concentrations, but there is certainly room for better quantification.

    Intriguingly, there is a report by John Dodson of coal being used for fuel in the Yellow River Valley 4500 years ago because the forests there had already been cut down (according to pollen records). Presumably the coal was dug out from surface outcrops. The Yellow river valley was already heavily populated 4500 years ago, so If that practice continued for the next 4500 years the amount of CO2 emitted would have been larger than my first crude estimates.

  33. 33
    Hank Roberts says:

    Dr. Ruddiman, have these studies been relevant for your estimates?

    Megafauna and ecosystem function from the Pleistocene to the Anthropocene
    “… drawing on special features of PNAS and Ecography that have been published as a result of an international workshop on this topic held in Oxford in 2014. Insights emerging from this work have consequences for our understanding of changes in biosphere function since the Late Pleistocene and of the functioning of contemporary ecosystems, as well as offering a rationale and framework for scientifically informed restoration of megafaunal function where possible and appropriate.”

    I found that while looking for the reference on whale poop and trophic cascades, which turns out to be just one of many similar examples.

    It seems that backing off our stripmining of the oceans may be a good idea, and that recovery of the larger ocean predators could be rather fast and make quite a difference in drawing down CO2

  34. 34
  35. 35
    Richard Caldwell says:

    Hank: It seems that backing off our stripmining of the oceans may be a good idea

    Richard: Dunno. That might be all that’s left to do once the reefs are finished toasting. You know what hit men say about double-tapping.

  36. 36
    Oliver Morton says:

    Hank @14 — ecopoiesis not a Lovelock term — coined by the late Canadian biologist Robert Haynes

  37. 37
    Mal Adapted says:

    Hank Roberts:

    Megafauna and ecosystem function from the Pleistocene to the Anthropocene

    That’s a very interesting article, Hank. It bears directly on a couple of discussions I’m in elsewhere in the blogosphere. I may have to go back to those and eat some crow. But thanks for the link anyway 8^}!

  38. 38
    JoeDOE says:

    do you see the pause?

  39. 39
    Ned Ford says:

    This is a wonderful dialogue for someone who isn’t often able to access serious scientific debate. I just wanted to suggest that people working on terminology for the anthropologic impact on climate might want to consider something that is inescapably clear to me and other people who spend a lot of time on solutions, as opposed to the science.

    Human activity certainly caused the boom in GHG’s and non-CO2 forcings. It seems reasonable that the length of time since the last warming peak 14K YBP might be extended or aggravated or whatever by human activity.

    But if (when) we solve this particular problem it will be the first time in human history that we have had to act as a species, as opposed to everything else we have done, including emissions and land use change – where we have acted as individuals and groups. I discount the CFC conventions, since there were such a small number of factories and nations involved in manufacture, although there is no real reason this phase change needs to be defined as a point in time.

    To me this is the next step in human evolution.

    And, since this seems to be a group which might benefit from the word, we are on the precipice of a dramatic move in the right direction. U.S. clean energy achieved a rate of new installation (efficiency plus wind plus solar) in 2015 which will end coal use in 2040, if unchanged. But this is on top of four years of stunning growth, particularly for solar, but wind too if you adjust for the effect of the premature end to the Federal wind tax credit in 2012. If this growth rate persists for another four years or a few months more or less than that we will achieve a rate of new clean energy which will end all fossil generation in 15 years.

    And by preserving that rate for another decade or so we will provide clean and affordable fuel for all EV’s, replacement power for retiring old nuclear plants, electricity to displace a significant amount of non-electric natural gas and space heating, and hydrolysis to produce ammonia instead of natural gas, to list the top items. This doesn’t by itself end petroleum or natural gas use, but it provides early fast reductions that make 450 ppm feasible, possibly even a little better if we don’t drop a few rows.

    It’s too bad we don’t have constructive conversations like this in the energy arena. Thanks!

  40. 40

    “…do you see the pause?”

    Yeah, it’s rapidly receding in my rear view mirror. Can’t spare it too much time, though; I’m looking for an important exit ramp.

  41. 41
    Digby Scorgie says:

    #39 Ned Ford

    I’m sure all regular visitors to this site appreciate your comments and would encourage you to visit regularly. A complementary website is I visit both websites almost every day, but it does chew up about an hour of my time. Still, one gains a vastly better idea of where the climate — and humanity — is headed. And in this regard, please pardon a cynical old man for thinking that what global action is currently occurring or in the pipeline is not going to be enough to counter the changes we’ve unleashed. It’s a race. Time will tell, and I don’t think we’ll have long to wait to see who wins the race — perhaps two decades?

Switch to our mobile site