Sunday, October 23, 2011

The BEST results: the scientific method works.

Recently released results by an independent research team ("BEST") confirm something that had been obvious for a long time: the Earth is warming.

The recent results by the Berkeley team (BEST) confirm that the Earth is warming. That's no surprise; we had known that for decades. So, what's so special in these results?

One point is, of course, the evident disarray of the skeptical tribe, as they had clearly put great hopes in this study. But that is a short term phenomenon as they are rapidly closing ranks and restarting the doubt-creating machine. Rather, what is interesting in the BEST study, I think, is a further demonstration of how well the scientific method works (*).

Think about that: the BEST study had started with great fanfare as a new study performed by people who defined themselves as "skeptics". It was supposed to be the final world on whether the earth is warming or not and clear hints were sent by the performers that they had strong suspicions that decades of work by climate scientists had been badly affected by an overlooked phenomenon known as "Urban Heat Island" (UHI). Considering some previous statements by the BEST team leader, Richard Muller, and some of the financing sources of the study, it was not an auspicious start.

But, instead, the team was staffed by professionals and worked professionally; applying the scientific method. At least three different teams had worked before BEST in examining the data provided by the temperature measuring stations. They all had used the scientific method; just as the BEST team did. In the end, all four teams arrived to the same results. The "UHI" bias does not exist (or, better said, it is correctly accounted for in the treatment of the data). See? The method works.

But then, why so much discussion? What made the BEST team think that previous studies were wrong? And what made critics of the BEST effort think that it would be biased in favour of anti-science theses? Well, it is a fact that scientists are human beings and they have their personal biases.

There are two kinds of typical scientific biases: one is when an aged scientist mistrusts everything new; it is the "not measured here" syndrome. Within some limits, that is a syndrome shown by Richard Muller in several of his public statements - but, in the end, it didn't affect the work of the team.

The other kind of bias occurs when scientists turn out to be easily gullible on matters they are not experts on. This is shown, among many examples, by the recent case of the "E-Cat," the device that was claimed to be able to produce energy by nuclear fusion reactions. This kind of bias is specular to the one described before; here, a scientist may take position on the basis of incomplete data, but "measured here." Eventually, however, also this bias can be corrected by the scientific method.

So, we have a good method that we can use to understand what's happening around us and what problems we will be facing in the future. We can use the scientific method to take action in order to avoid the negative effects of climate change and resource depletion. The problem? We are not using it.

(*) But what is exactly the scientific method? It is not so easy to say as it could seem, since different fields of science require different approaches and the complete description of the method needs a rather long article in Wikipedia - to say nothing of the many books and studies that have been dedicated to the subject. But I think there is a single fundamental point in the method: experimental results always take precedence over theory. In other words, reality always trumps hopes. It is this approach that defends us from the ideological bias that is part of our way of thinking.


  1. Thanks, Prof. Bardi. Timely post for me ;)

  2. Actually, Ugo I think they have made an error. I have taken their method (which is qualitative rather than quantitative) and applied it to the full set of station data for the USHCN data set, breaking the analysis down by states. However I have used actual numbers for the population densities around the stations - and as a result I think that they got it slightly wrong.

    What they show is not always what folk think - consider the refuse thrown at me for pointing out what they now note "While the trend distributions are broad, with one-third of the stations in the US and worldwide having a negative trend, both distributions show significant warming." - note the bit about the US!

    The problem is that the relationship to population is logarithmic, so that a village growing from 1 person to 100 has more change than a town growing from 100,000 to 150,000. The effect is also thus evident across the full spectrum of sites that are used, rather than just the 1% that have a population of more than 50,000. (I just spent 3 days checking all this and would be glad to share the numbers).



Ugo Bardi is a member of the Club of Rome, faculty member of the University of Florence, and the author of "Extracted" (Chelsea Green 2014), "The Seneca Effect" (Springer 2017), and Before the Collapse (Springer 2019)