Sunday 23 November 2008

Faces of Climate Science - the Why?

With the onset of winter I have WTMTOMH (way too much time on my hands) and rather than burn it off playing Civ 4, Rise of Nations, or Sim City Societies, I thought I'd finally tackle a project I'd had in mind for some time: creating a website listing the names of all the experts called upon as authors in the 2007 Fourth Assessment Report of the IPCC.

If you've already read my blog post below, or don't really care how or why I did this, here's a shortcut to my intro page for my List of Climate Scientists, with stats plus links to author's homepage. The intro page sets up what I'm doing, then links to some variant forms of my listing: one with just IPCC AR4 authors, another with a longer list of climatologists that I've collated. I've also created variants with only the photos. The intro page links to these, but with a caution that those pages link to several hundred photos, so they eat a lot of bandwidth and can take a long time to load (esp. the first visit when the photos are not in your browser cache).

Okay, why?

Well, last year I spent a fair bit of time working on Wikipedia, on a really wide scattering of topics that grabbed my interest, but always coming back to
List of scientists opposing the mainstream scientific assessment of global warming - a motley crew of varied qualifications or relevance to the scientific debate. As of today there are 42 names on the page.

This page has a checkered history of edit wars, interminable debates about what screen to use for inclusion and who meets the bar, etc. There are at least a half-dozen active editors who appear to be strong climate change 'skeptics' eager to include as many names as possible. They seem to engage in so-called 'quote mining' to find statements in support of their 'side.'

If you read over that page, the thing that I find striking is how scattered are the views of this list of people. Some think solar variation explains most recent change so well we can dump all our research about CO2; others--almost half--say the task of projecting future changes is too daunting for us to make any judgments at all - they are in effect 'agnostics,' often implying it is simply beyond human knowing to foresee what effects our atmospheric CO2 pulse may have.

Anyway, the kind of argument I often see on this topic frequently involves some kind of claim that the number of names on this list should count as evidence that the "science is not settled" and that there is an ongoing debate over whether humans actions have or could impact on climate in any detectable way.

That claim seems patently false to me. The number of scientists who still maintain that adding CO2 to the atmosphere does little to nothing to global temperatures is almost vanishingly small, in the context of the number of scientists qualified to speak as experts on this topic.

To demonstrate that, we can point to the authors of the four IPCC Assessment Reports. This report did not waver on that question; it very clearly states that adding CO2 is leading and will lead to rising temperatures (and sea levels.) There are a lot of authors credited in this report--a whole lot. The list in Annex B to the 2007 Fourth Assessment Report names 619 authors in "Working Group I" -- the group responsible for summing up the scientific basis: the greenhouse effect, quantifying human-caused emissions of CO2, CH4, CF6, N2O, etc. and quantifying their radiative forcings; working out the other forcings such as sooty aerosols from coal and diesel, jet contrails, etc; and feedbacks, both positive and negative, that either amplify or dampen the initial forcings.

So I set about gathering all their names - getting the list proved quite simple as they are in the PDF file of Annex B, and indeed the climatologist who blogs under the nom-du-clavier "Eli Rabbett" has already extracted this list, numbered, and made a small start at annotating. His list is here.

I thought I would take things a few steps further. First, when I found that handy list I had already begun assembling a list by hand, picking the author names out of highly-cited papers on climate that I'd find via Google Scholar. This was labour-intensive but still rather interesting, and I've carried on tracing co-authors and departmental co-workers to find yet more names in this field.

Google Scholar can select all works by a specific author. There is an 'advanced' search' page with a box for this, but I found a quick way to enter this search condition is to type author:fm-surname (by observing what Google filled in as the URL once I'd run an 'advanced search').

Google returns the papers by that author sorted by how many times that work has been cited by others, with the most cited at the top. I started collecting just the number of cites for the top articles by each author as a simple indicator of that person's impact within the discipline.

I ended up gathering the number of citations for each author's top four listed papers. Why four? Because I could see that many on the first results page without scrolling, and it felt like a useful number.

I've settled on sorting the authors on the number of cites for their #4 most cited work (sub-sorting on #3 in case of ties). This is a totally arbitrary metric, but the results seem reasonably representative of the authors' standing. For all the ones I haven't gathered the stats on, I've just fallen back on alphabetical order.

Here's what I've got so far (the page is still a work-in-progress for me; I've only got up to letter K looking up home pages, and I've only done the citation stats for about a quarter of these names.)

Here is the intro page explaining the project, with links to the various formats.
The list in table form has name, year of Ph.D., country of birth (or residence), cite stats from Google Scholar, research area, and institutional affiliation. Each name links to the author's homepage.

The 'How'

I'd start with a name generally in the form "F.M. Surname", with just first and possibly middle initials. For some of the top names, I already knew what the initials stand for; otherwise, I'd have to follow the links to the journal paper in search of more clues. Many journals actually never give the full first or middle name of the authors, but they almost always add footnotes showing their institutional affiliations. A few journals now show the full name - much easier for me - but if not, I would next google for the surname plus the institution. Often this would lead straight to the author's academic homepage. If so, great; if not, keep hunting.

Some people - shocking! - don't actually have a homepage. Proving a negative is tough, so given the size of my self-defined challenge, I set a time limit on how long I'll search for a link for any given person. As a rule, anyone teaching at a university has at least some sort of locator page, with their address, phone number, and email. Most fill these in with statements of their research interest, academic history, and current group members. By contrast, some of the largest government-run research bodies such as the Hadley Centre and the U.S. nation research labs, and especially military labs, don't make this a practice. At best they may have a table of member names with contact info (phone and email); they simply don't create member home pages. The same holds for scientists at commercial service providers in the private sector, who from time to time get listed as co-authors on papers they helped on with such services as instrumentation, satellite communications, data assimilation and management, software delevopment, etc.

Fortunately, most experts at the civilian labs typically are cross-appointed at a nearby university also (Hadley->Reading/U.of East Anglia; NCAR->U.Colorado@Boulder, etc.) That usually gets me out of those blind alley.

The step of filling out the author's given names from just initials can take couple of minutes per person. Once armed with the full name and academic affiliation, it's usually a short search to get their home page. Finding the university or government lab, of course, gave me a new source of (full) names of all their research colleagues to add in. So the list just keeps growing - about to pass 1300, as of today.

Almost everyone in academia has a P.R. photo head shot online - if not on their homepage, then elsewhere on their institution's site or in an online brochure for a conference where they've spoken. So I started saving the URL of each author's photo as well.

This led me to create a second version of my project: the "Faces of Climate Science" all on a single page. Viewers with a slow link will probably want to skip this one. I've written a script to convert my tabular list into HTML, and I can tweak any design details such as the assigned image height to which all pictures get scaled:

Just photos with links to author's homepage.

I've been compiling the list for a couple of weeks now, filling up much of my spare time. I've found just under 1300 names, so almost 680 of my own that were not on the IPCC AR4 wg1 authors list.

I welcome any suggestions on how to improve these pages - alternate formats, ways to make them more readable/accessible/useful, whatever.

Jim Prall
Toronto, Canada


Unknown said...

I am in awe of the yeoman's work you have produced here. Please keep it up, it may prove pretty useful for the quote mine wars.

Unknown said...

Jim, I saw your post on RC and I must say I'm very, very impressed!

EliRabett said...

I've added an update to my list with pointers to yours. I am floored.

rpauli said...

Jim, Just coming from your RealClimate post... where you offered the number 4% of scientists in doubt.

Yours is the ONLY source that seems to study the issue - you're the defacto expert. Good going.

And yes the language is the key - whether to say:
man-made problem
human caused
human contributed
human amplified

Somehow I recall a lecture by Naomi Oreskes saying that most Americans believe global warming is real, and yet also think scientists are divided on the subject.

Wonder what other controversial science beliefs might we compare? Like How many believe in theory of plate tectonics, evolution, or even heliocentric orbits. (I was shocked to hear that many deny plate tectonics...) Could 4% be very low compared to other issues?

Thanks for all that you do. Let me know what you learn.

Richard Pauli