Those who criticise anthropogenic global warming (AGW) science sometimes say we need to take the long view and if we do human activity is relegated to noise in the natural system. James Hansen and his colleagues reckon they have come up with a story about the relationship between atmospheric CO2, temperature and sea level change that fits the whole Phanerozoic Eon, which dates from about 545 million years ago.
The image above doesn’t give the temperature values as we normally understand them, but the variations are quite large. It’s hard to find a direct translation into the equivalent of the current global temperature average, but as far as I can make out the temperature at the Paleocene-Eocene juncture about 55mya was about 12C above now. That’s before it jagged upwards by about 6C in the so-called PETM event.
Currently we are at a cool point. We’ve been there before, but not for about 450 Million years.
One valuable insight to set the scene, I think, is that throughout the Phanerozoic the diversity of life on the planet has continued to increase, with setbacks from time to time.
The biosphere has a lot to lose if we stuff it up.
Hansen and co concentrate mainly on the Cenozoic Era where the important event was the transition to a glaciated world about 35mya.
2007 was the year that the Arctic sea ice coverage took a dramatic dive. It was year the IPPC seemed to make a special effort to take the issue of sea-level rise out of play, effectively downgrading its importance to irrelevance. It was also the year that James Hansen said that he’d bet $1,000 to a donut (we say “dollars to donuts” or “London to a brick”) that under BAU (business as usual) conditions we’d get several metres of sea level rise. In his paper with 8 colleagues Climate change and trace gases in May 2007 (pdf), he reminded us that three million years ago, when the temperature was 2-3C warmer than now, the sea level was 25 metres higher, plus or minus 10, with atmospheric GHG levels similar to what we have now.
Of course with 46 colleagues he had already said the same in an earlier paper, his Dangerous human-made interference with climate: a GISS modelE study (pdf) along with the notion that during the Eemian interglacial about 125,000 years ago the sea levels had been about 4 metres higher with temperatures 1-2C higher than present and CO2 levels of less than 300ppm. In the ‘trace gases’ paper they had traced the main forces at work in the paleoclimate history of the last 65 million years. The reduction in temperature from 55mya while the heat of the sun reaching us had increased by about 0.5C is quite stunning. It is illustrated in the following graphs:
A major part of the story up to the development of the Northern Hemisphere ice sheets lay in the imbalances created by the out-gassing of CO2 from the sea floor created by continental drift as against the CO2 taken out of the atmosphere by the weathering of rocks exposed by mountain growth, for example in the Himalayas and the Andes, together with vegetation changes. While minute in themselves such imbalances could produce dramatic long-term effects. A forcing one 20,000th as strong as is happening now can change CO2 composition by 100ppm in a million years.
Hansen and colleagues also looks at the effects of increased insolation through periodic changes in the earth’s orbit and feedbacks from the resulting ‘albedo’ (solar radiation reflectivity) changes. Tundra reflects more heat radiation than forest. Open sea absorbs markedly more than ice. Wet ice absorbs more than dry ice.
Orbital changes were nothing more than noise when temperatures and CO2 levels were high.
When the Northern Hemisphere ice sheets finally appeared, CO2 levels were correspondingly low. A change of say 100ppm has far more impact on temperature when the CO2 level is 280ppm than when it is 1100ppm because of the somewhat logarithmic relationship between CO2 levels and temperature taking into account short-term feedbacks. When the Northern ice sheets came into play the orbital changes became the main story, producing a ‘whip-saw effect on temperature.
What Hansen and his colleagues found has important implications for what is happening now with global warming and climate change.
First, when ice-sheets are in play the ‘albedo flip’ and other log-term feedbacks mean that on a time scale of centuries rather than decades, ‘climate sensitivity’, the temperature change resulting from a doubling of CO2, has to be recalculated. They estimate that it doubles, so the 550ppm seen as the stabilisation point by Stern and implicit in EU and Australian targets could see a warming of 6C, which would almost certainly trigger ‘tipping points’ such as methane release, soils giving up stored carbon etc.
Secondly, there is no time gap between short-term feedbacks, taking up to several decades to impact on the climate, and long-term feedbacks that come into play centuries hence. Long-term feedbacks are already a factor.
Hansen’s work goes a long way to explain why the climatic effects of global warming seem constantly to be racing ahead of expectations.
Third, ice sheets which have taken about 6,000 years to disintegrate fully in the past do not do so in terms of an unalterable internal dynamic at a steady pace. Disintegration occurs discontinuously and depends on how hard they are being forced. Current human forcing is many multiples stronger than here-to-fore through natural processes.
Disintegration and consequent sea level rise can occur at frighteningly rapid rates, as in the ‘Meltwater pulse 1A’ event of 14,000 years ago.
At that time the sea rose on average one metre every 20 years for 400 years.
It happened perhaps a quarter way in the ice sheet melt. We don’t know whether such an accelerated event will happen this time or when it would start if it did. We don’t know how long the remaining ice sheets would take to melt, but it seems likely that it would be considerably less than 6000 years.
But Hansen and mates found that 5C of warming would produce an ice-sheet free world, which involves sea level rise of about 75 metres. Even at a steady melt over 6,000 years we would have sea level rise of 1.25 metres per century.
Yet this is the scenario that present public policy is committing us to, if we decide to adopt it on a global basis. Stern had said that a 450ppm was too hard and went for 550. He is now saying that we must go for 450ppm and Garnaut is following him on this. But Hansen and co are saying that as far as they can make out 425ppm, plus or minus 75, will produce an ice-free world.
This is the science that so far Stern and Garnaut seem to be ignoring.
In November 2007 David Spratt of CarbonEquity and Philip Sutton of Greenleap Strategic Institute produced an impressively referenced report taking this new science on board entitled The Big Melt. They drew the conclusion that we needed to reduce CO2 levels and shoot for 320ppm.
While the nations of the world were having a bun fight in Bali last December James Hansen was giving a briefing at the meeting of the American Geophysical Union saying that we need to get CO2e ppm down to the 300-350 range in the first instance and see where we go from there. It was reported here for example.
Also in February 2007 Spratt and Sutton brought out their Climate Code Red report, incorporating The Big Melt, extending it and going into the kind of political and economic effort required to address the problem. Again impressively referenced.
Hansen and 8 colleagues have since produced a further technical paper, Target Atmospheric CO2: Where Should Humanity Aim? (pdf) addressing directly in some detail the issue of reducing CO2 to at least 350ppm. He hasn’t yet done a popular version of this paper, but much of the rationale is contained in Scientific reticence and sea level rise (pdf), Climate Catastrophe: 2007 (pdf), and his Iowa Testimony, bearing in mind that his position had been moving from a 450ppm target to the more stringent 350ppm during this period.
Hansen does believe that we can have some overshooting in relation to emissions targets and still attain satisfactory stabilisation. Bearing in mind that the longer it goes on the bigger the risk of irreversible change and losing any control entirely, plus the fact that we are already overshooting.
Lest there be any relaxation it is important to note that in the first of the Hansen technical papers cited above a major conclusion was that the guard rail to avoid dangerous anthropogenic interference (DAI) was 1.7C above pre-industrial levels or 1C above 2000, not 2C as previously thought. That’s 0.9C above now. It is generally considered that there is a 0.6C delayed effect at any given time, allowing for changes in emissions to work their way through to warming via short-term feedbacks. With a trend increase of 0.2C per decade that gives us less than 2 decades leeway.
In the second technical paper, which by including long-tern feedbacks doubled climate sensitivity to 6C, the momentum commitment inherent in the system now was estimated as 2C. In other words we are already in trouble, but don’t quite know it, at least on the official level.
One final point. Remember Rahmstorf concluded that we were likely to get one metre of sea level change this century just by extrapolating on what’s already happened. Hansen found that ice sheet degradation ‘normally’ takes about 6000 years, that’s 60 centuries. If you consider the 75 metres of sea level rise in prospect through complete deglaciation, that averages at just 1.25 metres per century.
It’s just that we are pushing the system many times harder than ‘normal’ and no-one knows when the fun is going to begin.
Update: The next in the series on sea level is Sea level rise: some real world implications.
I dont think we can put the blame on Stern or Garnaut for ignoring this science. Plenty more senior people to point the finger at. Strategically, I think we’d better just cross the line in the next minute.
But jesus, we are so fucked.
I think Garnaut in particular is conscious that because the issue is one of the global commons nothing any one country does to go beyond the international norm will have any appreciable effect in a physical sense.
Stern and Garnaut have the problem that if they go beyond what the politicians can wrap their heads around they’ll simply be ignored. The Europeans, who in a sense have been the standard setters, know that if they go too far they’ve got no hope of keeping the likes of the US, Russia, Japan on board, not that they are in a fully committed sense now. Then we have the major developing countries who say historically they didn’t cause the problem and have the right to pollute their way to prosperity too.
So it’s going to take something dramatic to change attitudes (closer to home than the north pole) and by that time we’ll be in deep shit.
My immediate response was,reading about the glacial melt matter,why does it have to be assumed it is carbon dioxide the gas or whatever,when it could be sand!?And what is known about the weather realities down wind of glaciers in terms of geology,wherever it can be visually seen that there is a downwind!?Wouldnt Australians for example reading from the map,find difficulties in understanding processes emanating other than from India Pakistan and China, and more to do with other locations!?.Red Herrings,these thoughts may well be,but has the person Hansen and the Contributer of Hansen’s stuff here been downwind of all glaciers!? Surely some observations have to be made first-hand !? I will have to read all the contributed stuff here later.I suspect I wont be able to make up my mind,as to its worthiness as an example,to be more concerned in a deeper sense.
I cut my teeth on palaeoclimate and would like to add a few “riders” to Hansen et al’s conclusions.
There are good reasons to suppose if we move back to Miocene temperatures (~6-8 million years ago and older) we would get major embayments all over the world (f’r instance the Murray-Darling Embayment went as far as Swan Hill in northern Victoria and the Gulf of Carpentaria was much larger). The Pliocene, following the Miocene was a period of sea level decline as the world cooled. There are a couple of reasons as to why that may also have been influenced by factors independently of the biogeochemical cycle.
The rising Tibetan plateau giving us ice near the equator and the closing of the Panama Isthmus changing ocean currents are two reasons. This would suggest that the climate sensitivity of the Earth is not fixed but will change as the Earth changes. The other is that the ice-albedo feedback is larger when there is more ice than less. Therefore the sensitivity according to this definition would have been larger at the peak of the last ice-age about 20,000 years ago than it is today, when there is less ice.
For those reasons, I am wary of using uniformitarian methods (a geological principle that suggests processes in the past are the same as they are today) to set atmospheric CO2 limits. Though I do think that 350 ppm CO2 would probably buffer Milankovitch-driven (orbitally-driven) glacial-interglacial cycles, which are unusual in the geological sense. The diagram in Figure 1 above is interesting, but I’m not going to use the past to set limits for the future, though it provides useful insights.
For me, the critical policy point is not 350 or 450 ppm (i.e. a stabilisation concentration). Rather it is getting to the point of overshoot in emissions, then concentrations, then temperature as quickly as possible , i.e. the peak of the increase, followed by decrease. It is still possible to gradually increase concentrations to 550 ppm and to stabilise there, though I don’t know why we would want to. This level would be high likely to give us new beaches well inland in additions to a host of other risks, exceeding most people’s criteria for “dangerous climate change”. It is not any long possible to get to 450 ppm by this pathway - a gradually increase in concentrations - it is too late for that. We must decrease emissions and concentrations, i.e. overshoot - to get to 450 from the high emissions pathway we are on.
Even if we can do that over the next couple of decades, we will know by then how far further we may need to go and how far we are permitted to go by physical limits in the Earth’s systems (short of geo-engineering). The key point now is not the target but the pathway and the dynamics over the next decade or two.
The assumption behind the Spratt and Sutton paper, and I suspect the Hansen et al. material too, is to wake people up and to get them motivated. I don’t necessarily agree that this is the right psychology and think that trying to understand the economics of overshoot, the trade-offs between the efficiency of carbon sinks, new technology and where we thought we were going in terms of society and the economy are far more relevent. From here, the target of 350 or 450 is less important than getting off the path we are on. Showing how this can be done successfully to me is more important than establishing emergency conditions. Even if we do the latter, we’ll end up with “energy watch”, “climate watch” and “two-bob watch” schemes to to add to the “fuel watch” scheme.
Targets are good, but we need bows, arrows and a better aim than we have now.
A relative and friend of mine have joined the ranks of climate skeptics. Can anyone recommend a place to start reading up? I’ve never bothered with the issue much, thinking it self-evident that we were making a mess of things and that it will come back to haunt us. Secondly, what’s the go with Ian Plimer? Heard him on ABC radio last year, and it sounded like he was a skeptic because he didn’t understand statistical confidence!
Chookie, http://www.realclimate.org/ is the place to start, though I think it’s a bit technical in places.
Chookie,
even better, go to the New Scientist special report on climate change.
I saw Ian Plimer in his first outing on climate change some years ago at an ABC “Science at the Pub”. At that stage, he honestly professed his ignorance, but suggested the whole theory didn’t sound right, and needed some stiff scepticism in any case. No problem with that, but he is no longer ignorant. He is staunchly contrarian and has to overlook a great deal of good science to remain so. Does not understand complex systems, feedbacks or how models work, despite using models of economic geology his whole professional life. In his book A Brief History of Planet Earth had to talk up the risk from meteors in order to talk down the risk from climate change.
Suggest that after his big fight with creationists, he is a bit Jeff Fenech/Azumah Nelson - can’t give up punching.
Roger, thanks for your input. I’m tired tonight, but I went looking for a post in RealClimate and found some other stuff instead, so got lost there. I’ll respond in two parts, general then a few specifics.
On reflection I probably should have chopped off the first two figures and started with the paragraph “2007 was the year…” Hansen did say somewhere that what he was saying applied to the whole Phanerozoic. I looked up Wikipedia and grabbed the pic from there. I was reacting to certain denialists who say that 100 or 150 years is far too short to mean anything (and then start a cooling trend from 1998). So I thought 545 million years should be long enough for them.
At RealClimate I found this one by raypierre. Ostensibly it’s about the Cretaceous, but it’s as much about the Eocene. It gives some cause for pause.
I’d always thought it best to exercise extreme caution going back past the mid-Pliocene, when the Americas joined at Panama. But Hansen reckoned that the continents were roughly in the same place as now 55mya, so I thought who am I to disagree?
Since I’ve been reading Hansen, which was from 2003, he’s had two big concerns. One is that the disintegration of ice sheets can be discontinuous and faster than has been commonly thought with the kind of forcing that’s happening now.
The second is to establish as closely as possible where the boundary for DAI (dangerous anthropomorphic interference) is.
I think he got most of what he needed from the last 3.5 million years, when the Northern Hemisphere ice sheets became firmly established as the dominant pattern with fairly regular whipsawing in and out of interglacials.
So accepting your cautionary note, and leaving the 300-350ppm target zone aside for a moment, I’m feeling that his conclusions as I identified them in the post stand.
I’m thinking that what kind of initial target you adopt is a matter of strategy. I recall Gavin Schmidt’s (who works for Hansen) response to 350ppm was to say that we should try to arrest the upward trend inside 450ppm CO2e, then see how we are placed and what we know and go from there. It should be noted that on Gavin’s count we are currently at about 375ppm. He counts the CO2 equivalence of aerosols also.
My interpretation of Spratt and Sutton is different from yours. I don’t think it’s a psychological ploy, rather a sincerely held conviction that we know enough to justify emergency mode. They suggest that we need to approach it like we approached WW2.
The point was also made on the RealClimate thread that action can also be justified in terms of the risks associated with uncertainty. It does seem apparent that in terms of climate sensitivity the risk is on the upside. So 5C becomes more likely than 1C and the risks associated with 5C are such as to justify emergency action.
Spratt and Sutton say that the norm in getting into an aeroplane is 1 in a million. Any greater risk and you’d worry about flying. Yet we are meant to be relaxed about say a one in 20 risk of completely wrecking the planet.
The specific points will have to wait until tomorrow night.
Brian, good post - it summarises the main points of these papers very well.
Note that the 12C temperature range is for the deep ocean (not quite the same as surface T), and the Vostok temperature reconstruction in your 3rd figure (c) is for Antarctica, so you need to halve this to get an approximation of global T. Most estimates put the global surface T during Last Glacial Maximum at 5-6C below current, and the Eocene greenhouse world T at 4-6C above the present day. That’s about as precise as we can get.
The fast+slow climate sensitivity makes a switch at ~35 million years ago, from 3C for doubled CO2 to 6C, due to ice-sheet albedo (no ice existed prior to 35 mya so fast=slow). So yes, there is accounting for changes in sensitivity during planetary evolution. A drop from ~1200 to 600 ppm CO2 from the Eocene to early-Miocene would have caused a 3C drop in global temperature. By contrast, a later drop from 425 to 212 would result in a 6C temperature drop due to the fast+slow feedback effect.
Re Plimer,
It’s not just creationists he’s been fighting all his life. He used to tell his students that conservation was rubbish because “you can’t conserve a dynamic system”. In other words the ecosystem is always changing, therefore no change can be undesirable.
However, he doesn’t apply the same logic to economics. Apparently if millions of people die because of sea level rise that’s ok because the sea levels sometimes rise naturally. However, for Australians to bear the slightest economic pain from cutting back on fossil fuels would be a disaster. He also always argues that any emission reductions will happen in the most damaging way possible. He said that to meet our Kyoto targets we would need to stop all motor vehicle travel 4 days a week and since this was unacceptable we should just keep polluting - the idea that the same reductions could be a achieved through energy efficiency and driving smaller cars was something he flatly refused to accept.
and feral,
he is not even consistent within that framework, worrying about “all natural” meteors hitting the earth. Presumably because it might damage the economy, but I thought economic damage through human agency was the only risk worth worrying about?
Bit OT but Steve McIntyre over at Climate Audit has been taking a look at some of Hansens data manipulations and is not impressed. Steve isn’t a skeptic either.
Thanks, Barry.
I can’t actually recall where I got that information about 12C from, but it wasn’t from reading of the graph posted. I did know that it was calibrated in deep ocean T. It bugs me that I can’t find a graph of the Cenozoic era temperatures where the values are translated into global surface temperature. The nearest I’ve found is Fig 2 in an article by Zachos et al.
Today I went back and had another look at Hansen’s Iowa testimony (large pdf) which, Chookie, is an excellent primer on the whole topic of AGW and climate change. A couple of points.
On page 17 he says that he included Fig 14 (page 41) because it showed estimated CO2 in relation to glaciation for the whole Phanerozoic eon. The graph he showed was quite different from my Fig 1. I thought about using it, but it was not self-explanatory enough for an article like this.
His main focus in explaining the Cenezoic (my Figure 3) is on the period from 15 million years BP (before present) when Antarctica iced up completely again after having been partly iced since 27my BP (page 15ff). He emphasises the role of CO2 throughout, but in the Pleistocene (1.8my BP) he nominates orbital change, a very weak forcing, as the chief intigator of climate change and GHGs and ice sheet area as the chief mechanisms as a feedback.
Roger, Hansen’s big point, I think, is that the glaciologists were getting nowhere with ice sheet models, because the melting they were studying as it happened was minute and no guide to the complete disintegration of an ice sheet. He thought it better to look to the paleoclimate record. His Iowa stuff is very accessible and reads well, IMHO.
In the end he’s looking for insights. Even there he can’t study the complete melting of Greenland and Antarctica, because it didn’t happen in the last 35my.
So we can’t assume that Meltwater Pulse 1A, for example, will happen just that way again, because we are dealing with different ice sheets, smaller volume of ice and in higher latitudes.
Peter, I don’t know whether I’ve got the energy for Steve McIntyre because the effort:reward ratio is usually pretty low. He was the one who announced with shock horror that NASA GISS had changed the temperature record so that 1934 and 1998 for the US (not the world) went from being statistically the same to being statistically the same. It took him to spot a sneaky change to cover up their presumed incompetence.
The significance of it all was so close to zilch it didn’t matter.
Well I don’t really think you can blame him for people making a mountain out of a mole hill. His hockey stick expose was pretty good and his efforts to expose the shoddy record keeping and data is commendable. I would have thought that the most important thing in this whole episode was good data but it seems there is little interest in improvements in this area. It’s also pretty obvious that the historical surface temperature record is badly compromised and was never designed to measure climate change (where .3 deg is significant). So this data has to be massaged in various opaque ways with some climate researchers pretty well refusing to reveal their methods. Bit of an eye opener if you ask me.
Peter, if you ask me, even the hockey stick thing was only a fraction as important as he thought it was.
I’m not an expert on how temperatures are recorded, but when I’ve put my head into that area I was convinced that scientists are sincerely doing the best they could. You can’t go back and re-measure the temperatures in the 1930s.
Also there are several groups doing the sums.