The Guardian and the Economist are among the media reporting on a press release from the Berkeley Earth Surface Temperature Group (BEST) lead by Professor Richard Muller. They report results of BEST’s analysis of temperature records which show the world is indeed warming. Thus, it is claimed, weakening the case for those who deny or are skeptical of “Climate Change”.
Here is the graph of temperature change from berkeleyearth.org/analysis.php
There are many strange things about these media reports. Here are some of them.
Very few, if any people, actually deny that the climate changes, they question the claim made by UN IPCC that recent climate change has been overwhelmingly caused by human activity. The BEST press release specifically states
What Berkeley Earth has not done is make an independent assessment of how much of the observed warming is due to human actions, Richard Muller acknowledged.
So it is immediately clear that contrary to what is claimed BEST have done nothing at all to weaken the case of skeptics AND have gone out of their way to explicitly state this. You might have thought anyone writing a review of the press release would have read this, as it’s in the press release.
The manner of the press release is unusual as it reports that 4 scientific papers have been submitted for peer review, then goes on to state that these papers will form part of the literature for the next IPCC report on climate change.
As the papers have only been submitted for review, this means they haven’t been reviewed yet, let alone accepted for publication. In fact the papers may never be published at all. So it is bizarre to claim the they will be included in the next IPCC report.
It is also unusual to announce results from papers that have not yet been reviewed or published as the reviewers may find errors or request corrections. This is the purpose of review.
It is not simple to measure the temperature of the earth, let alone how it has changed over hundreds of years. There are thousands of weather stations around the world. But the number has changed over time, sometimes stations are moved. Sometimes development takes place near a weather station so it changes from being in a rural setting to an urban one. Urban settings tend to be warmer than rural, so if the environment around a weather station changes from rural to urban it will distort the temperature record. Some countries have many weather stations (e.g. US) others have few (e.g. Africa).
So there is not a continuous complete record of temperatures but rather a hodge podge of incomplete records, with some parts of the world having a good record and others poor.
It is important to keep in mind that BEST have only analysed temperature records for land weather stations. What happens over the oceans, 2/3 of the worlds surface, has not been analysed.
The tools that are used to piece together the temperature record come from statistics and whilst they may be known to people who have studied statistics as an undergraduate or graduate, they are not known by all scientists. In fact it is bizarre the way politicians and media refer to scientists as though they are all equally skilled, equally knowledgeable and generally interchangable, like Cybermen.
Richard Muller is a physicist as are most of the people on the BEST team according to their about us web page. Though they do have 2 statisticians and a climatologist.
The statistical methods used by BEST have been severely criticised by Doug Keenan writing at BishopHill blog. Doug’s comments are worth reading, even if you skip the details of the maths, to see how he emails Richard Muller and points out the mistakes Doug feels Richard has made and gives a list of undergraduate textbooks to help stop Richard making the same mistakes in the future.
Matt Briggs (aka William M Briggs) who is a statistician, has also posted a critique of BEST.
Broadly speaking both Doug and Matt agree that temperatures may have risen in the 200 years or so, but are both highly critical of the statistical methods BEST have used. In particular they feel that the uncertainties in the BEST results are way to low (BEST quote figures to a 1/1000 of a degree C). If the uncertainties are greater, say +/- 0.5, it becomes much less clear if temperatures have risen let alone what the trend is.
What this means is the grey areas on the first graph become a lot larger so it becomes a lot less clear what the trend of the graph is.
Everything that I have written ignores the key part of the UN IPCC climate change claims, the predictions for the next 100 years. I have posted here how predictions using fitted functions can be very wrong. The error in the predictions will increase along with the uncertainties in the data used to make the predictions.
But there’s something even more important. The choice of function that you choose to fit. Looking back over temperature records for last few hundred years it can be seen that temperature wiggles up and down usually around +/- 0.5 degree. Why anyone would then assume that in the next 100 years the temperature will go up in a straight line has always confused me. Surely it will carry on wiggling about like it has always done? And if someone attempts to fit a straight line to some data which is not a straight line but oscillating (like a sine wave for example) their future predictions will be way off.
Michael Palmer, writing on Watts Up With That Blog has taken a different approach to the problem of incomplete temperature records. He noticed that there are 600 weather stations in US that have continuous records from 1900 to 2000 so he looked at those. The full details are in his post but here is his summary graph where he has shown rural and urban stations separately. (Note by continuous Michael means at least 90% complete, the completeness is indicated by the two dash lines at the top of the graph and the scale on the right).
And surprise surprise there is no warming over the 100 years, there is a slight fall.
Now if we take a load of incomplete data, munge it with some statistical trickery to try and make it complete and get a different result from a set of complete continuous observations which have not been statistically munged it poses the very real question.
Has the statistical munging produced a false effect?
Michael also points out Central England Temperature and Hohenpeissenberg, Bavaria are two examples of long running temperature records which do not show any particular long term warming.
Steve McIntyre at climateaudit.org has pointed out the BEST results show that temperatures from 1810 to 1820 were about 2 degrees cooler than today. But temperatures warmed about 1.5 degrees in the 10 years from 1820. Presumably this was not due to human influence.
There has been much discussion of attempts to hide previous warm periods such as the Medieval Warm Period (MWP) from 950 to 1250 which allowed the Vikings to colonise Greenland. Some have claimed that MWP never existed and the talk of living in Greenland was the medieval equivalent of time share salesmen! So it will be inconvenient for BEST to produce a rise of 1.5 degrees in 10 years.
There have been other warm periods, for example around 3000BC, when Skara Brae in the Orkneys was inhibited. It is known that wheat and barley were grown then, yet it is too cold for these crops today.
I remember watching a BBC program called “Making Scotland’s Landscape” presented by Iain Stewart, professor of Geoscience communication at Plymouth University, and learning about Skara Brae. Climate Change does strange things to people as having explained that in 3000BC the Orkneys must have been much warmer than today, maybe as much as 2 degrees warmer, Iain Stewart then went on to talk about how serious Climate Change is!
All this talk of past and future temperatures may seem boring and not at all relevant to everyday life. Nothing could be farther from truth. The UK government is using the IPCC predictions to decide how it manages cost lines (knock down sea walls as they will be too costly to maintain) and artificially raising the price of electricity so that windmills and other forms of green energy can be subsidised. The cost of this runs to billions and it is tax payers and energy consumers who will pay.
All of which means it is really important to make sure the “facts” and “predictions” are actually correct.
