Wednesday, July 6, 2011

Reconciling anthropogenic climate change with observed temperature 1998–2008

My colleague Robert Kaufmann (one of my PhD advisors and coauthor on several papers) has published a paper in the Proceedings of the National Academy of Sciences about the slowdown in global temperature increase between 1998 and 2008. The Guardian has pretty balanced coverage but all the bloggers I read seemed to be upset about it both people who support mainstream science and the climate change sceptics/deniers. Maybe that's a sign of doing something right? :) I talked for half an hour on Friday evening with a reporter from Science News about the paper though I don't get quoted in the report.

Joe Romm criticizes the authors for saying there was a period of flat temperatures when other data sources now show an increase for the period in question. There is more than one temperature series available and it is a fair point that they could have considered more than one data source. But the main point of the paper is that given the evolution of climate forcings over the period and inherent uncertainty the temperature series they look at is within the range of what you would expect to see. Actually, it tracks the model prediction quite well though temperatures were still colder than the expected value but within the confidence interval. So the sceptics claims that a halt in warming of this duration proves that anthropogenic climate change is rubbish is wrong.

Judith Curry who appears to be a climate skeptic also doesn't like it on the grounds that sulfur emissions didn't increase again from the their post-1980s decline until 2004 and that she thinks the authors aren't qualified. Based on the Smith et al. data, sulfur emissions hit a minimum in 2002 and increased by 9% by 2005. Actually, Robert has a pretty good background on climate change and James Stock is a leading econometrician.

The strangest claims are on WattsUpWithThis. They argue that either the paper is wrong (e.g. "it is out of date" because it only uses data up till 1998 to estimate the model!) or it proves them right that global warming is crap. In the comments people seem to be really confused about who even wrote this paper as well as the usual comments about government funding. The authors do thank the NSF but that is for a fellowship that presumably funded grad student Michael Mann. Kaufmann and Stock did the work as part of their jobs at private universities.

Then there is this Fox News headline (via Climate Progress).


  1. Looks to me from the SI that they relied on sulfur data from Smith et al. 2004 not 2011 that you have linked here (and which they cite). Not clear that it would make a difference in their analysis, but there are big differences in estimates for 1990s between the two datasets. Also unclear why they did not use the more recent data.

  2. Kaufmann et al, David and Roger jnr - talk about clutching at straws in the winds from China that do not reach most parts of the globe, and especially Australia, where there is indeed only a completely flat temperature trend since 1948 (and likewise zero trend in annual changes in temperature) (source data NOAA-ESRL). All BoM data showing otherwise are confections.

    But why rely on non-global data for sulphates (most of are contrived derivatives not actuals as in Kaufmann et al.), and leave out ALL mention of atmospheric water vapour, which provides a highly stat sig (99%)explanation for Australian temperatures (and elsewhere).

    BTW, I have just been invited (unsolicited) to submit on this to an international meteorological journal

  3. Roger - I am guessing the idea is that they took their existing model and then projected it out of sample and see how it matches the observations.

  4. David, Thanks, but they probably should have used the updated data on sulfur ... ignoring it is odd, especially since the 1990s saw such a change between datasets. Perhaps I'll email the authors to ask why they did it this way. Thx.

  5. The abstract of Kaufmann's paper says in part, "it has been unclear why global surface temperatures did not rise between 1998 and 2008." However, if he had used any other starting year between 1990 and 2000, he would have had to write: it has been unclear why global surface temperatures have risen between year X and year Y (1999 and 2009, for example).
    Why did he use a far-above-average year (1998) as the starting point? Seems like obvious cherry-picking to me, based on the data in this chart:

  6. They picked 1998-2008 because that has been the period that has been the big talking point for climate skeptics. The paper is meant to be a rebuttal to climate skeptics.

  7. Agreeing with someone is a strange way to rebut them. Denialists are now using Kaufmann's paper as expert "proof" that the world stopped warming in 1998.
    It's a simple case of faulty methodology; the so-called "pause" in warming is an artifact of picking an above-average value in 1998 and drawing a straight line to a below-average value 10 years later. By contrast, the trend line (the 5-year running mean in the chart I linked to previously) shows an obvious temperature increase from 1998-2008; there is no pause in warming for that period.
    Kaufmann then compounds the cherry-picking error by offering an explanation for something that doesn't exist (i.e. the "pause" in warming).
    If Kaufmann's aim was to provide further fuel for climate change deniers while calling into question the credibility of climate scientists, I have to say: Mission Accomplished.

  8. I agree with Anonymous, Kaufmann et all is a remarkably inadequate paper.

    Its Abstract states "Declining solar
    insolation as part of a normal eleven-year cycle, and a cyclical change from an El Nino to a La Nina dominate our measure of
    anthropogenic effects because rapid growth in short-lived sulfur emissions partially offsets rising greenhouse gas concentrations."

    Then it ends by stating:

    "The finding that the recent hiatus in warming is driven largely by natural factors does not contradict the hypothesis: 'most of the
    observed increase in global average temperature since the mid 20th century is very likely due to the observed increase in anthropogenic
    greenhouse gas concentrations'".

    So what would contradict the hypothesis? Apparently it is incontrovertible!

    But there are other oddities:

    we are asked to believe that "rapid (sic) growth in short-lived sulfur emissions partially offsets rising greenhouse gas concentrations". That is rather surprising when the "rapid growth" in sulfur emissions brings their total to only about 65 million tonnes in 2005, to be compared with emissions of CO2 in that year of 32.5 BILLION tonnes (gross, including LUCF), associated with co-emissions of water vapour of c.15 BILLION tonnes seemingly unknown to all climate scientists, with their evident disregard of the formula for combustion of a typical hydrocarbon:

    C3H8 + 5O2 → Energy + 3CO2 + 4H2O

    In this context it seems unlikely that just 65 million tonnes of sulfur could offset the radiative forcing from over 32.5 BILLION tonnes of CO2 and around 15 BILLION tonnes of H2O.

    But I suppose anything is possible in the pages of PNAS.