2010-11-28 21:15:47Met Office to revise global warming trend upwards
John Cook



The increase in average global surface temperatures has been around 0.16C per decade between 1970 and 2000 but has ranged between 0.05C and 0.13C in the last ten years.  The trend figures for the period between 2000-2009 are based on data from three different source; one from NASA (0.13C), one from the US National Oceanic and Atmospheric Administration (0.07C) and one from HadCRUT3 (0.05C) - the dataset managed by the Met Office and by the University of East Anglia's Climate Research Unit.

However, the increasing number of measurements of sea surface temperature (SST) from ocean buoys and the decreasing proportion of measurements from ships in the years since 1980 may have had the effect of depressing the recorded increase in temperatures. As a result, Met Office scientists have reviewed the whole sea surface temperature data set between 1850 and 2006 to take account of this bias. A paper has been submitted to the Journal of Geophysical Research which looks in more detail at all the biases in sea-surface temperature measurements from ships and buoys between 1850 and 2006 and reassesses the sea surface temperature records and their uncertainties in the light of this.  And since sea surface temperatures feed into the bigger picture this could lead to a reassessment of the global average surface temperature.

2010-11-29 08:47:39
Glenn Tamblyn


Here is a link to an earlier paper by Thompson et al


Looking at extracting various perturbing factors from the temp record to see the underlying signal - Volcanoes, ENSO etc. One result is that there seems to be a distorting of the SST record during WWII due to a significant switch between the proportion of US vs UK ships doing the measurements. This is most noticeable in Aug 1945 when the ratio changed quite suddenly. This may be one contributing factor to the difference between the warm period up to the 40's and the coolwer period afterwards.

2010-11-29 11:20:17worth a post
Dana Nuccitelli
It will be well worth doing a post on this when the paper comes out.  You know the deniers are going to spin this as "climate scientists fudging the data to show more warming."  It would be good to explain exactly why the revisions were made in a blog post.
2010-11-29 12:54:00Earlier Thompson paper
John Cook


The Thompson paper looks at data in the 1940s while this latest result shows a cooling influence just over the last decade. So two quite separate results.

I'm really keen to see the 1940s data fixed but it appears to be taking ages for the corrections to make their way into the published data. It really frustrates me seeing that spurious warming in the 1940s (I know, I need to get a life).

Similarly, will be interesting to see how this affects things - will it make 1998 no longer the hottest year on record even in the HadCRUT record? Will it affect the other records, GISS and NOAA? I'm not sure if they get their ocean data from the Hadley Centre but I imagine there is some overlap.

Of course deniers will spin this. On the other hand, it's like a reverse UHI effect - they like to focus on the influences that inflate the warming trend like microsite influences - not liking the fact that there are influences that impose spurious cooling trends.

2010-11-29 16:40:45
James Wight


John, the reason why the data haven’t been updated might be because the scientists at CRU have been a little bit busy being investigated for fraud. It frustrates me too though.