Welcome to the live Q&A with Richard L. Smith, L. Mark Berliner, and Peter Guttorp, the authors of the March Amstat News article, “Statisticians Comment on Status of Climate Change Science”. Starting at noon EDT, they will answer your questions. ASA Director of Science Policy Steve Pierson will be moderating.
Click on the black bar to enter your question. (Note: For carriage returns, hit shift-enter.)
The authors will give preference to questions over comments.
Thanks everyone! Now closing!
Full References for article (http://magazine.amstat.org/2010/03/climatemar10/)
1. Duffy, P.B., Santer, B.D. and Wigley, T.M.L. (2009), Solar variability does not explain late 20th-century warming. Physics Today, January 2009, 48-49.
2. Friis-Christensen, E. and Lassen, K. (1991), Length of the solar cycle: An indicator of solar activity closely associated with climate. Science 254, 698-700.
3. Hegerl, G.C. et al (2007), Understanding and Attributing Climate Change. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (S. Solomon et al, eds), Cambridge University Press, pp. 663-745.
4. Mann, M.E., Bradley, R.S. and Hughes, M.K. (1998), Global-scale temperature patterns and climate forcing over the past six centuries. Nature 392, 779-787.
5. Mann, M.E., Bradley, R.S. and Hughes, M.K. (1999), Northern Hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations. Geophys. Res. Lett. 26, 759-762.
6. McIntyre, S. and McKitrick, R. (2005), Hockey sticks, principal components, and spurious significance. Geophys. Res. Lett. 32, L03710, doi:10.1029/2004GL021750.
7. National Research Council (2006). Surface Temperature Reconstructions for the Past 2,000 Years. The National Academies Press, Washington, D.C. 2006.
8. Scafetta, N. and West, B.J. (2008), Is climate sensitive to solar variability? Physics Today, March 2008, 50-51.
9. Scafetta, N. and West, B.J. (2009), Interpretations of climate-change data. Physics Today, November 2009, 8-10.
10. Solomon, S. et al. (2010), Contributions of Stratospheric Water Vapor to Decadal Changes in the Rate of Global Warming. Science Express, Published online 28 January 2010 [DOI: 10.1126/science.1182488]
11. Wegman, E.J., Scott, D.W. and Said, Y.H. (2006), Ad hoc committee report on the "Hockey Stick" global climate reconstructions. Committee on Energy and Commerce and Subcommittee on Oversight and Investment, U.S. House of Representatives, Washington, D.C.
@Frederick K. Childers [Question 1 reply:] The issue of regional climate effects has been studied by the US Climate change Science Program. The basis for the statement was one of their synthesis reports
@Frederick K. Childers : [More on Question 1:]The argument is simple and reasonable over a wide range of quantifications: warming>more precip as rain in fall and spring>faster, sooner snow melt>decreased snow pack>decreased soil moisture>drought>fire
@Frederick K. Childers Comment on Susan Solomon's paper - yes, thank you for the correction, the paper did refer to a decrease in water vapor after 2000, not an increase.
@Guy [Question 2 reply:] The other studies that we were referring to are studies that have re-analyzed the data essentially taking account of Wegman's criticisms of the principal components analysis - so the datasets are not independent, but the analyses are new. In particular, Bo Li, Doug Nychka and Caspar Ammann have written two papers using Bayesian hierarchical models - one in Tellus (2007), the other accepted for JASA but currently available (I believe) as a preprint. Other papers in process include one by Tingley and Huybers, and one to appear by Mark Berliner and a student.Richard Smith
@petercraigmile [Question 3 reply:] The obvious direct issue is to compare the distribution of gridded weather data to the distribution of climate model output (climate, in this context, being the long-term distribution of weather).
@petercraigmile [Question 3 reply continued:] The issues are change of support, in addition to weather not being directly comparable to climate model output.
@petercraigmile [Question 5 reply:] This issue is discussed in the forthcoming Li, Nychka, Ammann paper. The basic issue is that it is necessary to smooth over a longer time period to reconstruct temperature from the borehole, especially going further back in time, Therefore, I suspect the smooth shape of the reconstruction is a consequence of the type of data rather thsan a real climatological effect. Bo Li has a list of preprints on her webpage
www.stat.purdue.edu though when I tried just now I did not succeed in downloading that paper.
@PaulM [Question 4 reply:] Not really. From the description in the paper it seems to be related to the percent variance explained, but the formula given does not make sense to me. I think my pdf file is damaged somehow.
@J. Arthur [Question 6 reply:] The IPCC report (www.ipcc.ch, 2007, working group 1) has a fairly strong statement to this effect in its summary for policymakers (they say evidence for warming is "unequivocal", whereas the evidence that it is human caused is merely "very likely"). Besides, several of us statisticians have analyzed temperature data from numerous different points of view, and I don't think any of us doubt that there is an overall warming effect.
@J. Arthur [Question 6 thread:] Here is one
Trend assessment in a long memory dependence model using the discrete wavelet transform
Environmetrics
Volume 15, Issue 4, Date: June 2004, Pages: 313-335
Peter F. Craigmile, Peter Guttorp, Donald B. Percival
@J. Arthur [Question 6 thread:] One of my own is: R.L. Smith (1993), Long-range dependence and global warming. In Statistics for the Environment, edited by V. Barnett and F. Turkman, John Wiley, Chichester, 141-161. There were a couple of other papers around 1991-1992 by Peter Bloomfield (and one by Bloomfield and Doug Nychka) in the journal Climate Dynamics
Thank you all for watching and commenting. Thanks also to the authors for their time. Feel free to follow up through Steve Pierson: pierson@amstat.org. We will wrap up shortly.
[Question 7 reply:] I have spent the last decade or two promoting interest in statistical climatology, and trying to get statisticians involved in IPCC. The history is interesting, but I don't have time to do it here. However, there is now for the first time a statistician as vice chair of IPCC WG 1 (Francis Zwiers from Canada), More are definitely needed.
@Guy There is a by now very large field known as "detection and attribution" analysis. See chapter 9 of working group 1 of the 2007 IPCC report (available online through www.ipcc.ch). The essential idea is to decompose the climate signal into (typically) four components, due to greenhouse gases, sulfate aerosols, solar radiation and volcanic effects. There is quite a sophisticated statistical methodology, mostly developed by climate scientists (see e.g. Allen and Stott, Climatic Change 2003) that performs a multiple regression in a high-dimensional space of observations. If the regression coefficient for greenhouse gases is statistically significant, then the effect is said to be "detected". The practical interpretation of this is that the observed warming cannot be explained by natural forcings alone. When the IPCC says "very likely", they mean the probability of the effect is more than 90%. They usually interpret this to be the same thing as rejecting a null hypothesis at the 0.1 level of significance. As a statistician, I am not so comfortable with this definition of "very likely", but it is one very widely used in IPCC and other summary reports.
Steve Pierson
lucymerf
Copyright 2012 Scribble Technologies Inc. All Rights Reserved.