NOAA breathlessly reports, “July 2015 was warmest month ever recorded for the globe.” CNN, USA Toady—oops, that’s USA Today—, BBC, and lots of other mainstream media lapdogs all obediently reported it.
But laymen need to pause for a deep breath before panicking.
First, “ever recorded” means back to 1880 (conveniently leaving out the Holocene Climate Optimum, Minoan Warm Period, Roman Warm Period, and Medieval Warm Period, all of which exceeded modern temperatures), and “recorded” means by land- and sea-based thermometers, whose siting (subject to contamination by urban heat islands and other phenomena) and sampling (very few to represent the whole earth, and not randomly located around the globe) make them notoriously less reliable than satellite measurements, and the amount by which July purportedly topped the next-hottest month on record was within the margin of error.
Second, the much more reliable satellite data firmly contradict NOAA’s claim. John Christy and (Cornwall Alliance Senior Fellow) Roy Spencer of the Earth System Science Center at the University of Alabama in Huntsville, in partnership with NASA, NOAA, and the U.S. Department of Energy, collect those data, currently in two database versions, 5.6 (which shows slightly more warming over the period 1979 to present) and 6.0 (which, ipso facto, shows slightly less).
Version 6.0’s record from December 1978 through July 2015 graphs this way (click the graph for higher resolution):
Note that the last bar (you really do need to click to see higher resolution), which is for July 2015, is well below the 0.2 degree line, but as you can see the bars for many months reach well above it. So, UAH v. 6.0 doesn’t find July the hottest month on record. As Spencer reports here: “The Version 6.0 global average lower tropospheric temperature (LT) anomaly for July, 2015 is +0.18 deg. C, down considerably from the June, 2015 value of +0.33 deg. C.”
So not only is July not the hottest month “on record,” it’s not even the hotter of the last two months!
Here are the raw numbers from v. 5.6 for July for every year, listed in descending order of temperature anomaly from the 1980-2010 norm:
1998 | 0.44 |
2009 | 0.37 |
2011 | 0.37 |
2010 | 0.36 |
2014 | 0.3 |
2005 | 0.28 |
2015 | 0.24 |
2002 | 0.22 |
2007 | 0.22 |
2013 | 0.17 |
2006 | 0.15 |
2012 | 0.15 |
1988 | 0.12 |
1991 | 0.12 |
2003 | 0.11 |
2001 | 0.06 |
1983 | 0.05 |
1987 | 0.04 |
1995 | 0.03 |
1997 | 0.01 |
2008 | 0.01 |
1980 | 0 |
1981 | -0.03 |
1990 | -0.03 |
1994 | -0.03 |
1999 | -0.07 |
1996 | -0.09 |
2000 | -0.09 |
1993 | -0.11 |
1979 | -0.12 |
1989 | -0.14 |
2004 | -0.19 |
1986 | -0.26 |
1982 | -0.33 |
1984 | -0.34 |
1992 | -0.4 |
1985 | -0.45 |
As you can see, July 2015 comes in seventh, behind 1998, 2009, 2011, 2010, 2014, and 2005, and it is above 2002 and 2007 by only 0.02 degrees, which is considerably smaller than the margin of error, so it’s essentially in a statistical tie with them.
So, breathe easier—and remember, take every claim of greater heat from NOAA and NASA (which are, after all, federal agencies, bound to serve whoever’s in the White House) with more than a grain of salt.
Ed Millerski says
Still, it is somewhat curious that 11 of the 12 hottest Julys in the UAH record have occurred in this century.
E. Calvin Beisner, Ph.D. says
Not really curious at all. By analogy: When you climb up to a plateau and then walk along it, every step is higher than the last till you reach the top, and then every stop on the top is higher than all the steps before reaching the top. Similarly, since about 1850 we’ve been rising, in fits and starts, with minor ups and downs along the way, out of the Little Ice Age. At the moment we’re on a plateau stretching back, according to the satellite data, about 18 years and 7 months (http://wattsupwiththat.com/2015/08/06/the-pause-draws-blood-a-new-record-pause-length-no-warming-for-18-years-7-months/). Nobody KNOWS whether we’re going to start climbing again, or begin descending (although some scientists think rising CO2 will take us upward again, while others think a cooling Sun will take us back down). Only time will tell. Meanwhile, the facts that (a) on average the models simulate twice as much warming as observed, (b) over 95% of the models simulate more warming than observed (implying that the errors are not random but driven by some kind of bias), and (c) none of the models simulated the complete absence of statistically significant warming over the last 18 years and 7 months) together entail that the models are wrong, that CO2 concentration doesn’t control temperature nearly so much as the modelers thought, and that the models therefore provide no rational basis for any predictions about future global temperature and therefore also no rational basis for any policy.