The quote I had problems with was:

The NCDC ranked September as fifth warmest since 1880 for land and ocean surface temperatures combined.

This is immediately contradicted in the paragraph below (which is what I quoted above).

I’ve probably confused you as well, by now.

I was a bit surprised at this:

The global surface temperature for the combined January-September year-to-date period was the fourth warmest January-September on record

I do follow the statistics posted by NASA GISS, and (for combined land & sea) they have September as 4th warmest, not 5th.

I’d also calculated (from the NASA GISS figures) that we probably had the second warmest Jan-Sep period in the record, and (since 1998 actually had quite a cold October and November) we were quite likely to see 2007 as the second warmest year to date.

This doesn’t tie in with the NCDC, though. I could accept that my maths is probably clumsy compared to theirs, but they put 1998 as warmer than 2005 (in contrast to NASA GISS).

I know that NASA GISS, WMO, and the UK’s Hadley Centre all monitor global temperatures with slightly different results – I’d kind of assumed that the NCDC and NASA used the same data, but this now appears not to be the case.

Can anyone point me to anything that explains why the NCDC’s figures are different from NASA’s?

Not that this is important – the trends are pretty much the same no matter which source you use, and we’re clearly warming. I’m just curious, and would like to know more.