The Temperature of 2010
Now that the relevant data for the temperature of the Earth’s surface for the past year are available, it is instructive to examine the claims made by some that it was the hottest year ever, and claims made by others that it was in second place.One way of looking at the year’s data we have used before is to see how the individual months fare. This is only one way of interpreting the data, there are others, but this way will show if any record temperature is a result of a general yearlong rise or just a few exceptional months.These five temperature databases I examine give the monthly temperature to thousandths of a degree which is superfluous. When rounded up to a more physically sensible 0.1 deg almost all of the differences between the years of the past decade go away, but that is another story, and not the subject of this post. Quoting temperatures to a thousandth of a degree is not physically meaningful. I round the figures to a hundredth of a degree which I think is a good compromise, but still superflous in reality. This occasionally causes some ambiguity when for example a particular month in one year may have a temperature anomaly of 0.400 and the same month in another year 0.402. This is deliberate and is not sloppyness. If the analysis carried out below is pedantically performed strictly to an accuracy of 0.001 degree it will cause some of the monthly sequences to be altered. To a scientifically untrained eye this can look like a succession of errors. However, what this is doing is naively overestimating the importance of noise in the dataset and is not a scientifically sensible basis for criticism, and given the obvious fact that we cannot measure the global temperature to a thousandth of a degree is a meaningless exercise, and unphysical in its conclusions. Also it should be noted that these datasets are sometimes modified slightly after the initial analysis is performed.However, it is interesting to note that if the analysis is done with no rounding at all, even though some of the years are rearranged, it actually supports the conclusions reached with the analysis of the figures rounded to 0.01 degree. My intention is to look for meaningful patterns in a sensible way and not to scrabble around in the noise of a distribution that may not be normal in the statistical sense. I think my approach has the right balance of caution with regard to the precision of the dataset. I have also limited my comparisons between the datasets. If one of them has a record it is noted even if it is not reproduced in any of the others. Another crucial factor to bear in mind is to be clear when a particular month is the warmest on record seen from 2011, or was the warmest on record at the time.2010 was an El Nino year and that had a profound effect on global temperatures, as we shall see. Before I examine the monthly temperature for the year I thought it would be instructive to see what an El Nino year looks like. In this case the strongest El Nino on record.1998Data from Hadcrut3 produced by the Climatic Research Unit.January; (at the time the warmest on record) cooler than 07, 04, 03, 02. February; warmest on record. March; (at the time the warmest on record) cooler than 10, 02.April; warmest on record. May; warmest on record. June; warmest on record. July; warmest on record.August; warmest on record. September; (at the time the warmest on record) cooler than 09, 07, 06, 05, 04, 03, 02, 01. October; (at the time cooler than 97) cooler than 09, 08, 06, 05, 04, 03, 02, 01, 97. November; (at the time cooler than 97) cooler than 10, 09, 08, 06, 05, 04, 03, 02, 01, 97. December; (at the time cooler than 97) cooler than 06, 03, 97.The data shows just how exceptional was 1998 (and 1997). Nine months broke a record, and the other three were close. The warm Spring is typical of El Nino.The subsequent warm decade has altered things somewhat. We see that even for the hottest year on record, by most global temperature datasets, half of the months of the year were unexceptional in the context of the recent (and warmest) decade. 1998 now has the top 4 of the warmest months on record, and another entry in the top ten. Curiously, in the top ten warmest months, only two are after 2002 (Jan 07 in 6th place and March 10 in 10th.)For comparison consider a non-El Nino year.2003January; cooler that 07, 02. February; cooler than 10, 07, 06, 04, 02, 99, 98. March; cooler than 10, 08, 07, 05, 04, 02, 01, 98. April; cooler than 10, 07, 05, 04, 02, 01, 00, 98. May; cooler than 10, 05, 02, 98. June; cooler than 10, 09, 05, 02, 98.July; cooler than 10, 09, 06, 05, 02, 98.August; 09,98.September; warmest on record.October; warmest on record.November; 10, 09, 08, 05, 04, 01, 97.December; cooler than 97.The differences are clear.Moving onto 2010 there are five temperature databases to examine.Dataset: Crut3vJanuary; cooler than 09, 08, 06, 00, 99.February; cooler than 09, 08, 06, 03, 01, 00.March; warmest on record.April; cooler than 07, 05.May; cooler than 03, 98.June; warmest on record.July; cooler than 06, 05, 98.August; cooler than 09, 01, 98.September; cooler than 09, 07, 05.October; cooler than 09, 08, 07, 06, 05, 04, 03, 98.November; cooler than 05, 04.December; cooler than 09, 08, 07, 06, 05, 04, 03, 02, 01, 99, 98, 97, 96, 94, 93, 90, 89, 88, 87, 81, 79, 39,1852.Note: Two record months, though not statistically significant. El Nino warmth in Spring. La Nina cooling later in the year.2010 overall; cooler than 2005, 1998.Dataset: Hadcrut3 from CRUJanuary; cooler than 07, 04, 03, 02.February; cooler than 07, 02, 99, 98.March; cooler than 02.April; cooler than 98.May; cooler than 98.June; cooler than 98.July; cooler than 05, 98.August; cooler than 06, 05, 03, 01, 98.September; cooler than 09, 07, 06, 05, 04, 03, 98,97.October; cooler than 09, 08, 06, 05, 04, 03, 98, 97.November; cooler than 09, 06, 05, 04, 01, 97.December; cooler than 09, 08, 07, 06, 05, 04, 03, 02, 01, 99, 98, 97, 92, 87, 79.Note: No single month broke a record. Warm Spring evidence of El Nino.2010 overall; cooler than 2005, 1998, equivalent to 2003.“Met Office” Hadcrut3 (the Met Office also has a database it calls Hadcrut3 which it calculates a different way from CRU)Jan; cooler that 07, 04, 03, 02, 98.Feb; cooler than 07, 04, 02, 99, 98.March; cooler than 02.April; cooler than 98.May; cooler than 98.June; cooler than 98.July; cooler than 98.August; cooler than 09, 06, 05, 03, 01, 98.September; cooler than 09, 07, 06, 05, 04, 03, 02, 01, 98.October; cooler than 09, 08, 06, 05, 04, 03, 98.November; cooler than 05, 04, 01.December; cooler than 09, 08, 06, 05, 04, 03, 02, 01, 99, 98, 97.Note 5 warm months in the Northern Hemisphere Spring – sign of a strong El Nino. No single month broke a record.2010 overall; cooler than 1998.Dataset: NOAAJanuary; cooler than 07, 03, 02.February; cooler than 04, 02, 99, 98.March; warmest on record.April; warmest on record.May; warmest on record.June; cooler than 05.July; cooler than 05, 98.August; cooler than 09, 06, 05, 03,02.September; cooler than 09, 07, 06, 05, 04, 03, 02, 98.October; cooler than 09, 08, 06. 05, 04, 03, 02.November; cooler than 04.December; cooler than 09, 08, 07, 06, 05, 04, 03, 02, 01, 99, 98, 97, 94, 90, 87, 82, 79.Note: Three record months, though not statistically significant. Evidence of El Nino warmth in Spring.The NOAA Press Release said 2010 tied with 2005. However, the quoted errors are +/- 0.07 which means that 2010 is statistically equivalent to 09,08,07,06,05,04,03,02,01,98.Note that in the NOAA dataset 2010 is 2nd warmest land temperature and 3rd warmest ocean temperature.Dataset: NasaGissJanuary; cooler than 07, 05, 02.February; cooler than 98.March; cooler than 02.April; warmest on record.May; cooler 98.June; cooler than 09, 05, 98.July; cooler than 09, 08, 07, 05, 02, 01, 98.August; cooler than 09, 01, 98.September; cooler than 09, 06, 05, 03, 98.October; cooler than 05, 03.November; warmest on record.December; cooler than 09, 08, 06, 05, 04, 03, 01, 99, 97.Note: Evidence of warm El Nino Spring. Record warmest months not statistically significant.According to NasaGiss 2010 set a record with only two record months.A NASA Press Release said 2010 tied with 2005 which at a temperature anomaly of 0.63 was 0.01 above 2005. The Press Release went on to say that 1998 was in third place with 09,07,06,03 and 02. However, the reality is that 1998 and the other years are statistically equivalent being spread over a range of 0.03 deg C within the errors of 2010 and 2005. NasaGiss is thus statistically equivalent to no change since 1998 (note that in 2005 NasaGiss announced that 2005 was a dead heat with 1998. In 2007 they put 1998 behind 2005.)Many press reports said that 2010 was a near-record breaking year despite the cooling influence of a La Nina later in the year. What was omitted however was mention of the fact that the reason why the year was marginally warmer than previous years was because of the warming El Nino.Clearly we live in the warmest decade of the instrumental record. But contrary to press reports the evidence is that 2010 was a year no different from all of the years 2001-2009 with the exception of a moderate to strong El Nino that elevated temperatures in the Northern Hemisphere’s Spring, and a cooling La Nina later in the year. The standstill seen in global temperatures since 2001 continues.Finally, the temperature anomalies for the past 13 years.
| Year | Met Office Hadcrut | CRU Hadcrut | CRUTem | Giss | NOAA |
| 2010 | 0.50 | 0.47 | 0.71 | 0.63 | 0.62 |
| 2009 | 0.44 | 0.44 | 0.64 | 0.58 | 0.56 |
| 2008 | 0.31 | 0.33 | 0.53 | 0.44 | 0.48 |
| 2007 | 0.40 | 0.40 | 0.68 | 0.58 | 0.55 |
| 2006 | 0.43 | 0.43 | 0.67 | 0.55 | 0.56 |
| 2005 | 0.47 | 0.47 | 0.75 | 0.62 | 0.62 |
| 2004 | 0.43 | 0.44 | 0.61 | 0.48 | 0.54 |
| 2003 | 0.47 | 0.47 | 0.65 | 0.55 | 0.58 |
| 2002 | 0.46 | 0.46 | 0.66 | 0.56 | 0.58 |
| 2001 | 0.40 | 0.41 | 0.55 | 0.47 | 0.52 |
| 2000 | 0.24 | 0.28 | 0.36 | 0.33 | 0.39 |
| 1999 | 0.26 | 0.30 | 0.49 | 0.32 | 0.42 |
| 1998 | 0.52 | 0.53 | 0.82 | 0.56 | 0.60 |
Note; Met Office Hadcrut, CRU Hadcrut and CRUTem are all with respect to 1961–90. Giss is 1951-80. NOAA 1901-2000.
Feedback: david.whitehouse@netzerowatch.com