Fighting for useful climate models

  • This is part two of a two-part post. Read part one here.
Princeton University's Suki Manabe published his latest paper in March this year, 58 years after his first one. Credit: Princeton University

Princeton University’s Suki Manabe published his latest paper in March this year, 58 years after his first one. Credit: Princeton University

When Princeton University’s Syukuro Manabe first studied global warming with general circulation models (GCMs), few other researchers approved. It was the 1970s, computing power was scarce, and the GCMs had grown out of mathematical weather forecasting to become the most complex models available. “Most people thought that it was premature to use a GCM,” ‘Suki’ Manabe told interviewer Paul Edwards in 1998. But over following decades Suki would exploit GCMs widely to examine climate changes ancient and modern, helping make them the vital research tool they are today.

In the 1970s, the world’s weather and climate scientists were building international research links, meeting up to share the latest knowledge and plan their next experiments. Suki’s computer modelling work at Princeton’s Geophysical Fluid Dynamics Laboratory (GFDL) had made his mark on this community, including two notably big steps. He had used dramatically simplified GCMs to simulate the greenhouse effect for the first time, and developed the first such models linking the atmosphere and ocean. And when pioneering climate research organiser Bert Bolin invited Suki to a meeting in Stockholm, Sweden, in 1974, he had already brought these successes together.

Suki and his GFDL teammate Richard Weatherald had worked out how to push their global warming study onto whole world-scale ocean-coupled GCMs. They could now consider geographical differences and indirect effects, for example those due to changes of the distribution of snow and sea ice. Though the oceans in the world they simulated resembled a swamp, shallow and unmoving, they got a reasonably realistic picture of the difference between land and sea temperatures. Their model predicted the Earth’s surface would warm 2.9°C if the amount of CO2 in the air doubled, a figure known as climate sensitivity. That’s right in the middle of today’s very latest 1.5-4.5°C range estimate.

Comparison between the measured sea surface temperature in degrees C calculated by the GFDL ocean-coupled GCM, from a 1975 GARP report chapter Suki wrote - see below for reference.

Comparison between the measured sea surface temperature in degrees C calculated by the GFDL ocean-coupled GCM, from a 1975 GARP report chapter Suki wrote – see below for reference.

At the time no-one else had the computer facilities to run this GCM, and so they focussed on simpler models, and fine details within them. Scientists model climate by splitting Earth’s surface into 3D, grids reaching up into the air. They can then calculate what happens inside each cube and how it affects the surrounding cubes. But some processes are too complex or happen on scales that are too small to simulate completely, and must be replaced by ‘parameterisations’ based on measured data. To get his GCMs to work Suki had made some very simple parameterisations, and that was another worry for other scientists.

Asking the right questions

Exploring Wally Broecker's theory that a giant meltwater lake, known as Lake Agassiz, flooding into the Atlantic could have caused an abrupt cooling spurred Suki Manabe and Ronald Stouffer to push their ocean-coupled GCMs further. Image credit: Julian Murton/University of Sussex

Exploring Wally Broecker’s theory that a giant meltwater lake, known as Lake Agassiz, flooding into the Atlantic could have caused an abrupt cooling spurred Suki Manabe and Ronald Stouffer to push their ocean-coupled GCMs further. Image credit: Julian Murton/University of Sussex

“They thought, ‘You’ve got to improve the parameterisations first’,” he recalled. “Improving parameterisations is a very important issue, and I think that we have to do everything we can to do that. But on the other hand that model was a little bit like a stereo set – when you have a $100 speaker, there’s no point using a $10,000 amplifier. Don’t spend disproportionate time on the complicated parameterisation of some process that you don’t fully understand.”

Instead, Suki stayed with his simple parameterisations for decades, applying them to many types of climate simulations. He, Richard and other scientists would look in more detail at what changing CO2 levels did in their models, and the role of natural variability and clouds. Into the 1980s they also increasingly simulated ice ages, and the role of ocean flows in climate, driven in part by Wally Broecker at Columbia University in New York. Just as Wally credited Suki’s global warming models for convincing him that human CO2 emissions were worrying, Suki now responded to a stark possibility Wally raised.

Willi Dansgaard at the University of Copenhagen, Denmark, and other scientists had found evidence of abrupt cooling in the Northern half of the planet 12,800 years ago. In this ‘Younger Dryas’ period gradual warming at the end of an ice age reversed dramatically, and then was followed by similarly rapid warming 1,300 years later. In the 1980s, Wally suggested a massive meltwater lake had suddenly emptied into the North Atlantic, shut down an ocean flow carrying warmth northwards, and flipped climate into a different state. If that were true, perhaps similar threats might come with the warming that human CO2 would bring.

Using their ocean-coupled GCM, Suki and fellow GFDL scientist Ronald Stouffer showed in 1988 that two different climate states – with and without the warming flow – could indeed be stable. By 1995, they had also shown that Wally’s melting lake idea could make the flow weaken abruptly, intensify, weaken again and then slowly recover. That wasn’t quite enough to reach a stable state without the warming flow, but was enough to cause sharp temperature rises and falls.

Emerging from the swamp

When including ocean movement, Suki and his GFDL colleagues had to make controversial 'flux adjustments' where the ocean and atmosphere met in their GCMs to stop them becoming unrealistic. This diagram, showing the areas where the different model parts meet, is from the 1975 GARP report chapter Suki wrote - see below for reference.

When including ocean movement, Suki and his GFDL colleagues had to make controversial ‘flux adjustments’ where the ocean and atmosphere met in their GCMs to stop them becoming unrealistic. This diagram, showing the areas where the different model parts meet, is from the 1975 GARP report chapter Suki wrote – see below for reference.

To include realistic ocean flow in these studies, Suki and Ronald had to tackle the problem of ‘flux imbalance’. Moving beyond swamp oceans produced tiny differences in movement, energy and fresh water moving between the atmosphere and ocean that could push ocean-coupled GCM simulations into unrealistic states. So they introduced ‘flux adjustments’ to stop the simulated climate drifting away from a realistic state. But these changes lacked a basis in physical theory, another factor that drew criticism. Yet because Suki and Ronald made the same adjustments both when simulating the changes they’re interested in and the baseline simulations they compare against, they felt they were justified.

“It’s a fair game in that sense,” Suki said. “And suddenly I am able to do all kinds of interesting calculations. Everybody else is complaining because when they are working so hard to improve the model, here is a guy who came up with this gimmick, and without pain gets all kinds of interesting results, which may or may not be true in their opinion. They are angry, because instead of incorporating their parameterisation, I am still using my old 30-year-old parameterisations in the atmosphere, and then using this ad hoc method I started having a great time.”

Soon other scientists started recognising the power of the approach, and using Suki’s ocean-coupled GCMs themselves. And when the UN Intergovernmental Panel on Climate Change (IPCC) began preparations for its first assessment report, that would be published in 1990, Suki couldn’t avoid it. “I avoid these committees like pests,” he admitted. “They told me I better come.” As a result the report relied heavily on a paper GFDL had published the previous year, modelling the world’s atmosphere, land and oceans with the best combination of duration and detail then available.

Though Suki has since been less involved in global climate assessments, and flux adjustment is less widely used today, he has remained active in improving climate models. In March this year he published a paper in Proceedings of the National Academy of Sciences of the USA aimed at reducing uncertainty over climate sensitivity. “The uncertainty is attributable in no small part to our inability to model reliably the feedback processes of cloud,” he told me. “Our study proposes a promising approach for evaluating modelled feedback, using the top-of-the-atmosphere fluxes of outgoing radiation observed from satellites.”

Beyond modelling, Suki today sees assessing the impact of climate change as one of the most challenging tasks ahead. “Under the current situation, it is extremely difficult, if not impossible, to mitigate emission of CO2 enough to prevent 2°C warming,” he warned. “Increasing emphasis should be placed, in my opinion, upon the adaptation to global change such as increased frequencies of drought in semiarid regions, those of floods in moist regions, and sea level rise. Also, we could increase our effort to exploit the benefit of global warming such as increased plant growth from increasing CO2, and increased availability of the Arctic Ocean due to the rapid reduction of sea ice.”

Suki focussed on using his models to answer questions rather than improving them unnecessarily, a fact the similarity between this map from 1999 and the ones from 1975 above underlines. Here, Suki and Ronald are exploring Wally Broecker's idea that a meltwater lake flooding into the Atlantic could have shut down a warming ocean flow. The map shows how much warmer a control model with normal ocean circulation is than a model where a large discharge of freshwater runs into the Atlantic after both were run for 6,000 years. The temperature difference of the shaded areas is labelled in degrees centigrade. Figure taken from their 1999 Tellus A paper, used under Creative Commons license, see reference below.

Suki focussed on using his models to answer questions rather than improving them unnecessarily, a fact the similarity between this map from 1999 and the ones from 1975 above underlines. Here, Suki and Ronald are exploring Wally Broecker’s idea that a meltwater lake flooding into the Atlantic could have shut down a warming ocean flow. The map shows how much warmer a control model with normal ocean circulation is than a model where a large discharge of freshwater runs into the Atlantic after both were run for 6,000 years. The temperature difference of the shaded areas is labelled in degrees centigrade. Figure taken from their 1999 Tellus A paper, used under Creative Commons license, see reference below.

Further reading:

Apart from the quotes in the last two paragraphs, which are from questions I asked Suki directly, all the quotes in this blog entry are from an interview with Suki Manabe by Paul Edwards on March 14-15 1998, published by the Archives for the History of Quantum Physics Collection, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA. They are copyrighted by the AIP, and used here with its permission.

This year I’ve already written about the following pivotal climate scientists who came before Suki Manabe, or were around at the same time: Svante Arrhenius, Milutin Milanković, Guy Callendar part I, Guy Callendar part II, Hans Suess, Willi Dansgaard, Dave Keeling part I, Dave Keeling part II, Wally Broecker part I, Wally Broecker part II, Bert Bolin part I, Bert Bolin part II, Suki Manabe part I

Syukuro Manabe and Richard T. Wetherald (1975). The effects of doubling CO2 concentration on the climate of a general circulation model Journal of the Atmospheric Sciences : 10.1175/1520-0469(1975)0322.0.CO;2
Syukuro Manabe (1975). The use of comprehensive general circulation modelling for studies of the climate and climate variation The Physical Basis of Climate and Climate Modelling, Report of the International Study Conference, GARP Publications Series No. 16, World Meteorological Organization
Wallace S. Broecker, Dorothy M. Peteet & David Rind (1985). Does the ocean–atmosphere system have more than one stable mode of operation? Nature DOI: 10.1038/315021a0
Syukuro Manabe and Ronald J Stouffer (1988). Two stable equilibria of a coupled ocean-atmosphere model Journal of Climate DOI: 10.1175/1520-0442(1988)001
Syukuro Manabe and Ronald J Stouffer (1995). Simulation of abrupt climate change induced by freshwater input to the North Atlantic Nature DOI: 10.1038/378165a0
Syukuro Manabe and Ronald J Stouffer (1999). Are two modes of thermohaline circulation stable? Tellus A DOI: 10.3402/tellusa.v51i3.13461
Syukuro Manabe and Ronald J Stouffer (2000). Study of abrupt climate change by a coupled ocean-atmosphere model Quaternary Science Reviews DOI: 10.1016/S0277-3791(99)00066-9
Yoko Tsushima and Syukuro Manabe (2013). Assessment of radiative feedback in climate models using satellite observations of annual flux variation. Proceedings of the National Academy of Sciences of the United States of America, 110 (19), 7568-73 PMID: 23613585

Spencer Weart’s book, ‘The Discovery of Global Warming’ has been the starting point for this series of blog posts on scientists who played leading roles in climate science.

About these ads

9 Responses to “Fighting for useful climate models”

  1. rogerthesurf Says:

    I remember when this was the topic that scientists and governments were pushing. Anyone who remembers those days is not likely to be convinced by AGW theories.

    Cheers
    Roger
    http://www.rogerfromnewzealand.wordpress.com

    • andyextance Says:

      Loving the Leonard Nimoy voiceover! But it’s a bit unforgiving of you to still hold a grudge over an idea that’s 35 years old – I was 2 when this programme was made! How much have you looked into why people thought this was going to happen? It’s really interesting stuff and part of the point of this series of historical/biographical entries is to explore the evolution of climate theories like that. There are three key elements as I see it.

      One is the slight global cooling that went on from the 1950s to 1975, thanks largely to sulfate aerosol pollution, produced by burning coal, that reflects the Sun’s energy back out into space. One idea to fight global warming through geoengineering now is to deliberately put sulfates into the air, though it’s seen as risky.

      Another element is that scientists at this point had only just got direct data on the sequence of historic ice ages. That helped confirm the Milankovic theory of how they’re largely driven by changes in Earth’s orbit. That theory suggests that today we should be very slowly slipping towards an ice age, when temperatures in 2013 so far are in fact 0.6°C higher than the 20th century average. According to NOAA October was the 344th consecutive month with a global temperature above the 20th century average (the last time it was below that average was almost as long ago as when this video was made!). But in 1972 this really worried some scientists, who wrote to Nixon in warning. However, around the same time scientists like Wally Broecker suggested the warming effect of CO2 might be about to burst through – as indeed it did.

      The last element, which I think is especially relevant to this blog post, is that some of the evidence came from models that were set up wrong. Having looked into this “Coming ice age” movie, I note Stephen Schneider was involved. He’s going to be my next profile, and his book “Science as a Contact Sport” is a great read – some brilliant stories, including his role in this cooling debate. As I’ll discuss in a couple of weeks, he was active in raising the idea that sulfate pollution could cause so much cooling we were going to create an ice age. These Suki Manabe entries show the computing capabilities at the time were limited and scientists were working hard to put physical climate processes into models. Stephen Schneider and his teammates hadn’t included the effect of CO2 in the stratosphere, an important oversight. When they did they got climate sensitivity figures similar to Suki’s,

      All of these things happened in the 1970s, and in the time since the evidence for human-driven warming has grown ever stronger. Science is supposed to be able to come up with ideas, test them, and reject them when they’re wrong. By contrast, my experience is that we as people don’t do this quite so easily – as you say, you’re unlikely to be convinced. But to me, the reasons behind why we thought an ice age might be coming fit well with the modern warming consensus.

      • rogerthesurf Says:

        Andy,

        Thanks for your reply.

        First of all by showing you that video, I am not saying that the world is about to plunge into an ice age. What I am saying is that we have heard climate change hysteria before. You were too young to remember this. No one took much notice of it anyway but those of us who remember it know that the shepherd boy has cried “wolf” before.

        Your reply is complicated and convoluted, and as usual without consistent references.

        It is fairly obvious that the theories are being twisted around to fit the facts. Does it ever occur to you that AGW may simply be wrong?

        I’m surprised by what you quote from NOAA. Do you know that if you take NOAA’s data for the last 15 years or so, and do a least squares regression using Excel you will find that any warming is less that the error coefficient.
        I won’t give a reference for this fact, perhaps you should do it for yourself.

        Below is a link which covers most things very competently. I hope you can find the effort to read it with an open mind.
        http://wattsupwiththat.com/2013/12/16/open-letter-to-the-executive-producers-of-years-of-living-dangerously/

        Cheers

        Roger http://www.rogerfromnewzealand.wordpress.com

        PS: AGW has had a lot of effort put into it (biased effort I would say) but maybe you should look into the real reasons why this huge and expensive effort has been made.

    • andyextance Says:

      Hi Roger, I did the regression that you suggested. Indeed the warming from 1998-2013 was lower than the standard error, but for 1997-2013 and 1999-2013 warming was higher than the error. I think your point is misleading on that basis, and also in other ways. You can see in detail what I did and a more thorough discussion of its implications in this blog entry: http://wp.me/pLahN-1ne

  2. When the climate change fight got ugly | Simple Climate Says:

    […] Broecker part I, Wally Broecker part II, Bert Bolin part I, Bert Bolin part II, Suki Manabe part I, Suki Manabe part II, Steve Schneider part […]

  3. The warrior who gave his life to climate change | Simple Climate Says:

    […] Broecker part I, Wally Broecker part II, Bert Bolin part I, Bert Bolin part II, Suki Manabe part I, Suki Manabe part II, Steve Schneider part I, Steve Schneider part […]

  4. Climate change science anyone can play with | Simple Climate Says:

    […] opportunity to rise to a challenge set for me by a climate skeptic, rogerthesurf, commenting on this blog entry. “Do you know that if you take NOAA’s data for the last 15 years or so, and do a ‘least […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 160 other followers

%d bloggers like this: