According to the British Meteorological Office and the World Meteorological Association, the past decade was ‘by far’ the warmest since records began -a claim which totally ignores overwhelming empirical evidence to the contrary.
For example, in a supplement to the August 2008 Bulletin of the American Meteorological Society, Dr Jeff Knight et al claim that that the mean global surface temperature has changed by the statistically insignificant amount 0.07 +/- 0.07 deg C from 1999 to 2008.
The dicrepancy between this finding and the Met Office claim likely arises from the averaging processes used. Three temperature sets are typically used in estimating temperature, one in the UK Met Office, and two in the US, maintained, respectively by the National Oceanic and Atmospheric Administration and by the US space agency NASA. These three agencies (who are all involved with the Climategate scandal) take raw data (which they refuse to make public) and ‘homogenise’ it to deal with the anomalies caused by local variations and discontinuities. The existing satellite data sets are calibrated against the earth data sets.
Temperature “homogenization” of this type can be physically meaningless, as explained here and more rigorously here. When non linear transformation are applied, the same data set can consistent with both warming and cooling (See pennance.us/?p=4).
As a simple example: Imagine a planet, call it X, that has only two weather stations. Suppose that for many centuries the two stations both record a constant temperature of 16°X. Suppose that in a certain year one station records an average temperature of 0°X and the other 36°X yielding a mean of 18°X. Suppose the temperature scale °X is related to another thermometric scale °Y by an order preserving transformation. Suppose for simplicity that under this change of scale: 0°X converts to 0°Y; 16°X converts to 4°Y; 36°X converts to 6°Y. When °X are used, the data from the two stations shows a rise of 2 °X in the mean. The same data when converted to °Y shows a fall of in the mean of 1°Y.
Thus whether we deduce warming or cooling depends very much on the scale used. Nonlinear averaging behaves exactly like a nonlinear change of scale in this respect.
References
- A Fable of Global Warming by Philip Pennance
- Global Warming I —On the non existence of a global mean surface temperature by Philip Pennance
- Appendix to Global Warming I — The ordinality of temperature by Philip Pennance
- Global Warming II—On the non existence of a global mean surface temperature by Philip Pennance