Posted by Eli Rabett has been investigating Ian Plimer’s claim that climate scientists were cooking the books on the CO2 record. Plimer wrote:
The raw data from Mauna Loa is ‘edited’ by an operator who deletes what is considered poor data. Some 82% of the raw data is “edited” leaving just 18% of the raw data measurements for statistical analysis [2902,2903]. With such savage editing of raw data, whatever trend one wants can be shown. [p 416 of Heaven and Earth]
The raw data is an average of 4 samples from hour to hour. In 2004 there were a possible 8784 measurements. Due to instrumental error 1102 samples had no data, 1085 were not used due to up slope winds, 655 had large variability within 1 hour but were used in the official figures and 866 had large hour by hour variability and were not used.[2102] [p 418]
This drew a correction from NOAA’s Pieter Tans:
To illustrate how misleading
Plimer is I made a plot of 3 years of all hourly data, with 2004 in the
middle because Plimer discussed 2004. …
In the plot, “selected” data means that we have used it in constructing
the published monthly mean because those hours satisfy the conditions
for “background” measurements. The red stripes are extremely close to
the published monthly means. …
Also plotted in purple-blue are all non-background data. If one
constructs monthly means from ALL data, incl. non-background, one
obtains the purple-blue stripes. The differences are only slight, with
the seasonal cycle becoming a bit larger due to upslope winds, esp.
during the summer.
Tans concludes that Plimer is a con man, but the story doesn’t end there. Plimer’s reference 2102 is ftp://ftp.cmdl.noaa.gov/ccg/co2/in-situ/. I analyzed the 2004 Mauna Loa data from there and found there were some minor errors in Plimer’s numbers: In fact, due to instrumental error 1103 samples had no data, 1097 were not used due to up slope winds, 655 had large variability within 1 hour and were not used and 881 had large hour by hour variability and were not used.
No big deal, complaining about minor mistakes like this really would be nitpicking. But look at this page, where Ferdinand Engelbeen makes exactly the same mistakes as Plimer and has the numbers in the same order:
For 2004, 8784 hourly average data should have been sampled, but:
1102 have no data, due to instrumental errors (including several weeks in June).
1085 were flagged, due to upslope diurnal winds (which have lower values), not used in daily, monthly and yearly averages.
655 had large variability within one hour, were flagged, but still are used in the official averages.
866 had large hour-by-hour variability > 0.25 ppmv, were flagged and not used.
Now if they both got the numbers right they could have just independently analyzed the same data, but when two people make exactly the same four errors it is proof of copying. The Internet Archive shows that Engelbeen’s page has had those numbers since at least 2007, well before Heaven and Earth was published, so it is Plimer who has copied from Engelbeen. And since Plimer does not cite Engelbeen as his source, that’s plagiarism.
But why plagiarize Engelbeen? All Plimer had to do avoid academic misconduct was cite Engelbeen instead of the NOAA FTP site. Well, directly before the numbers that Plimer copied are two graphs, one showing the raw data and one showing the selected data. And directly after the numbers Engelbeen writes:
As one can see in the trends, despite the exclusion of (in the above second graph) all outliers, the difference in trend with or without flagged data is minimal, only the number of outliers around the seasonal trend is reduced and the overall increase in 2004 in both cases is about 1.5 ppmv.
And to end any doubt on this point, Engelbeen continues:
Does discarding of “contaminated” data influence the trend over a year or several years? I have asked that question to Pieter Tans, responsible for dataprocessing of the Mauna Loa data. His answer:
The data selection method has been described in Thoning et al., J. Geophys. Research, (1989) vol. 94, 8549-8565. Different data selection methods are compared in that paper, including no selection. The methods give annual means differing by a few tenths of 1 ppm. I assume that you have read the README file [4] when downloading the data. The hourly means are NOT pre-processed, but they are flagged when the st.dev. of the minute averages is large.
That’s the same Pieter Tans quoted earlier.
So it seems that the reason why Plimer didn’t cite Engelbeen was that Engelbeen conclusively refuted Plimer’s claims about data selection at Mauna Loa being used to manufacture a trend. (Engelbeen, by the way, is a global warming skeptic and wrote his page in an attempt to end the Keeling-curve denial of folks like Plimer.)
October 27, 2009
October 27, 2009
October 27, 2009
October 27, 2009
October 27, 2009
October 27, 2009
October 28, 2009
October 28, 2009
October 28, 2009
October 28, 2009
October 28, 2009
October 28, 2009
October 28, 2009
October 28, 2009
October 28, 2009
