Science Journalism: A Case Study

Jay Manifold points to an important article on why science journalism is so bad. Go read it. Then come back here for a not untypical case study.

Done?

O.K.

This started out last Shabbos evening when I was stopped in shul by someone who wanted my response to an item he'd heard from Rush Limbaugh. Apparently some "Swiss and German" scientists had discovered that the Sun was currently "hotter than it had been for a thousand years", with the implication that this was what was causing global warming. Now I'd not heard anything about this remarkable study, and my informant was unable to to give me any more details. Some things rang false about it though. First, the sun's temperature isn't the issue, what is important is the total energy output of the Sun, its luminosity, or brightness. Second, how could they possibly know what the temperature, or luminosity of the Sun was, a thousand years ago? Third, even if they did, how big an effect did they actually measure?

A quick Google search led to this Telegraph article from July 2004 with the following lede:

Global warming has finally been explained: the Earth is getting hotter because the Sun is burning more brightly than at any time during the past 1,000 years, according to new research.

So, we have a year old study, and the issue is "brightness" as expected.

A study by Swiss and German scientists suggests that increasing radiation from the sun is responsible for recent global climate changes.

Here are the Swiss and Germans.

Dr Sami Solanki, the director of the renowned Max Planck Institute for Solar System Research in Gottingen, Germany, who led the research, said: "The Sun has been at its strongest over the past 60 years and may now be affecting global temperatures.

60 years, not 1000.

"The Sun is in a changed state. It is brighter than it was a few hundred years ago and this brightening started relatively recently - in the last 100 to 150 years."

Or maybe a few hundred.

Dr Solanki said that the brighter Sun and higher levels of "greenhouse gases", such as carbon dioxide, both contributed to the change in the Earth's temperature but it was impossible to say which had the greater impact.

So maybe it is something down here after all.

There then follows some boilerplate on Kyoto et al. followed by some actual information on what Dr. Solanki actually studied.

To determine the Sun's role in global warming, Dr Solanki's research team measured magnetic zones on the Sun's surface known as sunspots, which are believed to intensify the Sun's energy output.

The team studied sunspot data going back several hundred years. They found that a dearth of sunspots signalled a cold period - which could last up to 50 years - but that over the past century their numbers had increased as the Earth's climate grew steadily warmer. The scientists also compared data from ice samples collected during an expedition to Greenland in 1991. The most recent samples contained the lowest recorded levels of beryllium 10 for more than 1,000 years. Beryllium 10 is a particle created by cosmic rays that decreases in the Earth's atmosphere as the magnetic energy from the Sun increases. Scientists can currently trace beryllium 10 levels back 1,150 years.

And here we have the thousand year number.

Note the following about the described study:
a) The luminosity of the Sun is measured with two proxies: sunspots and beryllium 10.
b) The "dearth of sunspots" and "cold period" refer to the Maunder minimum, which corresponded with the period known as the "Little Ice Age". This occurance is the main data point supporting a connection between sunspots, solar luminosity, and terrestrial climate.
c) While the sunspot proxy isn't too bad an indicator of solar luminosity, the beryllium 10 measurement is actually a proxy for the sunspot number, so you have two sets of uncertainties to get through before you can draw a causal connection.

The article then goes on to discuss just how much this effect would affect global temperature, but never states how big an effect there was.

So, armed with the author's name, I tracked down the original research on the Astrophysics Data System and found two papers, one in Solar Physics and one in Nature. (Alternate link) It is the latter paper that appears to be the source of the Telegraph article, although it wasn't published until 28 October 2004, having been submitted 20 Feb 2004 and accepted 1 Sept. 2004. So, presumably, it was really a press release of some sort. (Odd though. Nature really frowns on any advance publicity for papers submitted there. Indeed, there is a press release, but it is dated 28 October 2004, as it ought to be.)

I won't go into all the details of the study, except to note the last sentence of the Nature abstract

Although the rarity of the current episode of high average sunspot numbers may indicate that the Sun has contributed to the unusual climate change during the twentieth century, we point out that solar variability is unlikely to have been the dominant cause of the strong warming during the past three decades.

Indeed, in a previous paper they write

In particular, the Sun cannot have contributed more than 30% to the steep temperature increase that has taken place since [1970].

I doubt the poor journalist ever saw this though, and depended on some suitably filtered version of the study. So, a paper that finds small, and cyclic, variations in the solar luminosity, and specifically states that it is not the dominant cause of global warming is transmuted into a claim that the Sun is causing global warming.