"BEST" temperature record study surprises skeptics

Last month, a team of scientists from Berkeley called the Berkeley Earth Surface Temperature (BEST) group released results from research they did on the Earth surface temperature record. Though there have been numerous studies and time series created on surface temperature, they wanted to take an independent look at the data and create a new temperature record. What they found was surprising to some in the "skeptic" community, though not surprising to most climate scientists.

Dr. Richard Muller is the founder and scientific director of the BEST group, which is made up of physicists, statisticians, and climatologists. Though Dr. Muller has been described as a climate change "skeptic" and "denialist," he has an impressive and extensive curriculum vitae in physics, including being a consultant for the U.S. Department of Defense, and a MacArther Foundation Fellow, and the recipient of the National Science Foundation Alan T. Altman Award. His skepticism is evidenced most frequently in the press by his funding from the Koch brothers, who have made billions of dollars in the oil industry. The BEST project also accepted funding from Koch, among many other organizations, though the funders had no influence over methodology or results, which is almost always the case in peer reviewed science. The BEST group also includes Dr. Judith Curry, the chair of the School of Earth and Atmospheric Sciences at Georgia Tech, who has recently been vocal about the need for a more transparent scientific process, and more eyes on the data, especially when it comes to research on man-made global warming and the temperature record.

The BEST team was open with their hypothesis: they expected to find that, when using temperature stations that other organizations failed to include, the warming trend wouldn't be present, or at least not as dramatic. Their objectives are listed on their website (which also includes access to data and submitted papers), which include:

-- Merging land surface data into a raw dataset that's in a common format and easy to use
-- Developing new and potentially better ways of processing, average, and merging the data
-- Creating a new global temperature record
-- To provide not only the raw data and the resulting record, but also the code and tools used to get there, making the process as transparent as possible



Figure 1. Locations of the the 39,028 temperature stations in the Berkeley Earth data set (blue). Stations classified as rural are plotted on top in black.

The BEST project collaborators combined data from 15 sources that, wherever possible, did not include the tried and true data that the "big three" (NASA, NOAA, or HadCRU) used in their analyses, mainly the GHCN Monthly dataset, which is widely used because of its requirements that the each station in the data set have plenty of observations, no gaps, and no erroneous data. However, the BEST project was born to create a new global surface temperature record, and to "see what you get" if you use observations that other institutions have weeded out. BEST looked at data from 39,028 different temperature measurement stations from around the globe (Figure 1), and developed an averaging process to merge the stations into one record, which you see below in comparison to previous records that have been constructed.



Figure 2. Temperature time series from the big three: NASA Goddard Institute for Space Science (NASA GISS, blue), NOAA (green), and the Hadley Centre and Climate Research Unit of East Anglia (HadCRU, red) along with the results from the BEST project (black).

The result was a new land surface temperature series to be added to the well-cited records of NOAA, NASA, and HadCRU, in addition to some truly independent, amateur compilations. The new temperature record agrees with the records from "the big three," and agrees with them on a warming of 1°C since 1950. BEST also addressed concerns raised by the skeptic community about station bias and urban heat island effect. They conclude that the urban heat island effect does not contribute significantly to the land temperature rise, given that urban area is only 1% of the land area in the record. Also, they looked at the stations that Anthony Watts has reported as "poor" quality, and have found that they also showed the same warming as the stations that were reported as "OK." This helps to show that temperature stations were not "cherry picked" in previous studies for warming trends, but for honest station quality.

The addition of another (eventually) peer-reviewed temperature series is good, and more eyes looking at the data is good, but the result is not surprising. However, it might have changed the minds of some skeptics who have been wanting to see an analysis from scientists that they find trustworthy. I think Dr. Muller sums their results up nicely in his Wall Street Journal opinion article:

When we began our study, we felt that skeptics had raised legitimate issues, and we didn't know what we'd find. Our results turned out to be close to those published by prior groups. We think that means that those groups had truly been very careful in their work, despite their inability to convince some skeptics of that.

The BEST project has four papers out for review in various journals. Having released the results to the public eye before undergoing the scrutiny of peer review, they've also made some updates to the analysis since these papers were submitted, thanks to a peer review process of its own: the internet.

Links and references:

Berkeley Earth Surface Temperature

BEST FAQ

BEST Press Release

Angela