Really they should make this harder. In a story purporting to "prove" that absent knowing what they were looking at or taking into account what can be considered climatic change in a scientific sense, statisticians can say we are warming even though we have been cooling (yeah, doesn't make sense to me either), we are given whoppers such as this one:
2008 was still the ninth hottest in 130 years of NOAA records.
Of the 10 hottest years recorded by NOAA, eight have occurred since 2000, and after this year it will be nine because this year is on track to be the sixth-warmest on record.
This is an outright falsehood. There is no way to compare the numbers NOAA uses now, which incorporate land and sea temps completely unavailable for most of the 130 period, with all of the historical record. (It simply is not an apples to apples comparison, and idiots who don't understand what that means shouldn't be writing about science in the press.)
This is the data the AP is using:
(I know this is the data because this NOAA states "For the year to date, the global combined land and ocean surface temperature of 14.7°C (58.5°F) was the sixth-warmest January-through-September period on record.") If you really think we have data from all of the places marked on the above graphic for 130 years, well, then you are smoking something.
Now, if one wanted to just to look at a more homogeneous data set, like the US temps one could make a comparison for 115 years. Let us see what that looks like:
Turns out this ranks as only the 29th warmest Jan-Sep in the last 115 years.
This was the 37th warmest.
NOAA doesn't have the reports for 2001 and 2000 online at the moment, of the years we have 2009 (trend only) to 2002 only two of them (2006-07) have been in the Top 10 warmest years in the last 115, not 8 of them.
Additionally, I've notice the time period 1961-1990 being used to provide a "base line mean" to measure anomalies (as it is in the blended land/sea graphic I showed earlier.)
So, how representative is that 1961-1990 period? Well, not very it seems.
So, 21 years of that 30 year period (or 70%) were below the long term average. It was a pretty cold period. Therefore it isn't surprising that we find most other temps hotter then that, and that is a good part of the reason the blended land/sea graphs are misleading (to put it mildly.)
Compare the US Temps on this graph for 2008 (using the 61-90 baseline):
...with the US Temps using the 115 yer averages:
So, according to the long (115 year averages) 61 regions in the US had "much cooler" or "record cool". Only one region (Long Island) was "much warmer". But according to the other graphic, 25 regions were warmer then "average" and only 15 were cooler.
As I said, it's misleading.
Roger Pielke Sr. looks at this as well:
This article, however, (which is not a true independent assessment if the study was completed by NOAA scientists) is not based on the much more robust metric assessment of global warming as diagnosed by upper ocean heat content. Nor does it consider the warm bias issues with respect to surface land temperatures that we have raised in our peer reviewed papers....to state that the “[t]he Earth is still warming” is in error. The warming has, at least temporarily halted.