What Evidence for “Unprecedented Warming”?
Posted by Jeff Condon on January 20, 2011
This is a new paper which takes a look at the statistical uncertainty of the long term warming trends by Pat Frank. He looks at the uncertainty of the data including that created by non-stationary errors and comes to the reasonable conclusion that global trends for the length of the temp record are statistically indistinguishable from zero. This has particular implications for model verification and especially to the real need for vetting errors in station measurements. Pat asked me to post on it here, and he has written a blog style explanation of his results below.
I want to copy the abstract and a bit of the conclusion here before his post just to help frame the discussion.
Sensor measurement uncertainty has never been fully considered in prior appraisals of global average surface air temperature. The estimated average ±0.2 C station error has been incorrectly assessed as random, and the systematic error from uncontrolled variables has been invariably neglected. The systematic errors in measurements from three ideally sited and maintained temperature sensors are calculated herein. Combined with the ±0.2 C average station error, a representative lower-limit uncertainty of ±0.46 C was found for any global annual surface air temperature anomaly. This ±0.46 C reveals that the global surface air temperature anomaly trend from 1880 through 2000 is statistically indistinguishable from 0 C, and represents a lower limit of calibration uncertainty for climate models and for any prospective physically justifiable proxy reconstruction of paleo-temperature. The rate and magnitude of 20th century warming are thus unknowable, and suggestions of an unprecedented trend in 20th century global air temperature are unsustainable.
Pat Frank –
First, my thanks to Jeff for his interest and for posting this essay.
Steve McIntyre’s ClimateAudit (CA) blog has been a terrific stimulus to zillions of its readers, including me. Back in May 2008, Steve posted on Nature’s “discovery” of the bucket-adjustment discontinuity in the Sea Surface Temperature record, three years after it had been discussed in detail at CA.
Brohan, 2006  (B06) came up in the discussion of Steve’s post. Brohan, 2006 was the most recent compilation of the global average surface air temperature, HadCRUT3. HadCRUT3 was a production of the UK Met Office, which had taken responsibility from the Climate Research Unit at University of East Anglia. Warwick Hughes, a real hero in the fight to bring transparency to the global temperature record, discussed this transition, here and here.
The discussion at CA led me to read Brohan, 2006, where I noticed that they had described measurement noise as strictly random and didn’t mention systematic error at all. That seemed doubly peculiar, and that led to the analysis I’m presenting here.
Reading the air temperature literature, it became clear that this double peculiarity typified the approach to error right back through the 1980’s and before.
What I found was that Folland et al, had made a guesstimate back in 2001  that the average measurement error was (+/-)0.2 C. This (+/-)0.2 C was applied by B06 and treated as random and uncorrelated among surface stations. So, following the statistics of random errors, B06 decremented the (+/-)0.2 C as 1/(sqrtN), where N = the number of temperature measurements, and as N got large the error rapidly went to zero. And that was the whole B06 ball of wax for measurement error.
To make the long story short, assessment of error methodology showed that guessing an average error is an explicit admission that you have no real physical knowledge of it. Random error is “stationary,” meaning it is defined as having a constant average magnitude and a mean (average) of zero. When one has to make a guesstimate, one doesn’t really know the magnitude, and doesn’t really know whether the error is stationary.
In short, if one doesn’t know the error is random, then applying the statistics of random error is a mistake.
Guesstimated errors don’t go as 1/(sqrtN). They go as 1/(sqrt[N/(N-1)]). That means at large N, the error rapidly goes to 1´(the original guesstimate). So, that (+/-)0.2 C error from Folland, et al., 2001, has to enter unchanged into the surface air temperature record. Since the guesstimated (+/-)0.2 C error enters both into every measurement, and propagates into the 1961-1990 normal used to calculate the anomaly trend, the total measurement error in every point in the global average temperature anomaly trend is sqrt[(0.2)^2 + (0.2)^2]=0.28 C. And so that correction alone more than doubles the nominal uncertainty in the global air temperature anomaly record.
But it’s worse than that. Kenneth Hubbard and Xiaomao Lin at the University of Nebraska, Lincoln, showed that there is a large amount of systematic error in surface station temperature measurements [3-5]. This systematic error mostly comes from solar loading on the radiation shield and from wind speed effects. These effects cause very significant deviations in measured temperatures.
Under very ideal conditions of siting and maintenance, Hubbard and Lin found that a standard Minimum-Maximum Temperature System (MMTS) sensor produced an average daytime bias of 0.43 C away from the correct temperature, with a standard deviation (the uncertainty width) of (+/-)0.25 C.
Cotton Regional Shelters (CRS) – the usual shelter for the older mercury thermometers – produced twice as much systematic error in high precision resistance thermometers, with an uncertainty width of (+/-)0.53 C. The older mercury thermometers inside CRS shields are likely to be even less accurate and less precise. But mercury thermometers inside CRS shields provide the bulk of the 20th century temperature record.
So, at the end, I estimated a lower limit of uncertainty in the 20th century global surface air temperature anomaly record by combining the guesstimated measurement error and the ideal MMTS systematic error. To do this, one has to also propagate these uncertainties into the temperature normal used to calculate the anomalies. This all produced a lower limit of uncertainty of 1-sigma = (+/-)0.46 C.
The Figure below shows what happens when this 1-sigma lower limit of uncertainty is plotted onto the GISS global temperature anomaly trend.
Legend: (•), the global surface air temperature anomaly series through 18 February 2010, (http://data.giss.nasa.gov/gistemp/graphs/). The grey error bars show the annual anomaly lower-limit uncertainty of (+/-)0.46 C.
The message is clear: including the lower limit of instrumental uncertainty, the trend from 1900 through 2000 is indistinguishable from 0 C, at the 1-sigma level.
“Unprecedented” 20th century temperatures? Hardly. It appears no one really knows the rate and magnitude of warming. Once again, climate alarm appears to be rooted in neglect of uncertainty. As with GCM projections. As with proxy paleotemperature reconstructions (which in any case aren’t even science).
This analysis is now out in Energy and Environment , and anyone who’d like a reprint can contact me at pfrank830 AT earthlink DOT net.
1. Brohan, P., Kennedy, J.J., Harris, I., Tett, S.F.B. and Jones, P.D., Uncertainty estimates in regional and global observed temperature changes: A new data set from 1850, J. Geophys. Res., 2006, 111 D12106 1-21; doi:10.1029/2005JD006548; see http://www.cru.uea.ac.uk/cru/info/warming/.
2. Folland, C.K., Rayner, N.A., Brown, S.J., Smith, T.M., Shen, S.S.P., Parker, D.E., Macadam, I., Jones, P.D., Jones, R.N., Nicholls, N. and Sexton, D.M.H., Global Temperature Change and its Uncertainties Since 1861, Geophys. Res. Lett., 2001, 28 (13), 2621-2624.
3. Hubbard, K.G. and Lin, X., Realtime data filtering models for air temperature measurements, Geophys. Res. Lett., 2002, 29 (10), 1425 1-4; doi: 10.1029/2001GL013191.
4. Hubbard, K.G., Lin, X. and Walter-Shea, E.A., The Effectiveness of the ASOS, MMTS, Gill, and CRS Air Temperature Radiation Shields, J. Atmos. Oceanic Technol., 2001, 18 (6), 851-864.
5. Lin, X. and Hubbard, K.G., Sensor and Electronic Biases/Errors in Air Temperature Measurements in Common Weather Station Networks, J. Atmos. Ocean. Technol., 2004, 21 1025-1032.
6. Frank, P., Uncertainty in the Global Average Surface Air Temperature Index: A Representative Lower Limit, Energy & Environment, 2010, 21(8), 969-989.