As we’ve seen in this report from John Goetz, GISS: worlds airports continue to run warmer than ROW a significant portion of the GHCN (Global Historical Climate Network) surface temperature record is measured in airports, not rural open fields. Airports, airport expansion, and air travel frequency tend to be linked with the population, growth, and wealth trends of a city. It stands to reason that since the majority of thermometers in the GHCN record are at airports, they’d have a broad application of UHI. Joe tries out a simple method of approximating what the signal might look like with a UHI removal. – Anthony
By Joseph D’Aleo, CCM, AMS Fellow
Hadley Center Annual Mean Temperature since 1895 shows a warming of about 1C since 1895. – Click for larger image
The long term global temperature trends have been shown by numerous peer review papers to be exaggerated by 30%, 50% and in some cases much more by issues such as urbanization, land use changes, bad siting, bad instrumentation, and ocean measurement techniques that changed over time. NOAA made matters worse by removing the satellite ocean temperature measurement which provide more complete coverage and was not subject to the local issues except near the coastlines and islands. The result has been the absurd and bogus claims by NOAA and the alarmists that we are in the warmest decade in 100 or even a 1000 years or more and our oceans are warmest ever. See this earlier story that summarizes the issues.
No one disputes the cyclical warming from 1979 to 1998 that is shown in all the data sets including the satellite, only the cause. These 60-70 year cycles tie in lock step with the ocean temperature cycles and solar Total Solar Irradiance. The annual mean USHCN temperatures are shown below along with the annual TSI and PDO+AMO.
Click for larger image
One needs simply to look at the record highs for the United States and globe to see that the warmest years are not all in the last two decades (although some were to be expected given it is one of two peaks in the cycles). The first image below shows the decadal state record all-time highs. The 1930s still clearly dominates (24 state all time records) with only one state (South Dakota) in the 2000s tying a 1930s all-time heat record.
Click for a larger image
The following image (enlarged here) shows the record monthly highs by individual year. Note the 1930s and 1950s dominate and this decade showing the least record highs than any decade since the 1800s.
Here is the NCDC compilation of the continental all-time records (enlarged here), note for all the populated continents, the records were in the 1800s and early 1900s.
TRYING TO GET AT A BETTER LONG TERM TREND
NCDC removed the UHI effect for the US in 2007 in version 2 of the USHCN. GISS maintains their version of a UHI adjustment of this NCDC USHCN data. By differencing the two, I found the following (enlarged here):
NOAA USHCNV2 -vs- GISS – click for larger image
It shows an artificial warming of about 0.45 C or 0.75F for the NOAA data for removal of the urbanization adjustment. Phil Jones of the Hadley Center, co-authored a paper that showed the UHI contamination of China was 1 degree Celsius (1.8F) for the century, so this contamination appears not to be unreasonable, in fact it may be conservative.
I then took that UHI adjustment for the United States and applied to the global data. The Hadley center data is dominated by land areas with their ocean temperatures mainly coming from ships and in the northern hemisphere. Here’s what Hadley says about marine data “For marine regions sea surface temperature (SST) measurements taken on board merchant and some naval vessels are used. As the majority come from the voluntary observing fleet, coverage is reduced away from the main shipping lanes and is minimal over the Southern Oceans.”
I subtracted the UHI annual contamination from the annual Hadley CRUT3v global temperatures. I got the following (enlarged here):
This gives a much more believable view of global temperatures, consistent with the natural forcings and more in line with records shown. The greatest warming was in the early 20th Century. The warming since 1930s and 1940s was negligible (0.2C). It suggests much to do about nothing in DC and Copenhagen. See PDF here.
UPDATE: This post has been changed to include a raw Hadley CRUT3v global plot, a NOAA-GISS difference plot and a corrected adjusted Hadley plot now all in Celsius. This is a work in progress and an attempt to see what Hadley plot might look like with an adjustment for UHI that numerous peer review papers suggest is necessary. Your suggestions are welcome (jsdaleo at yahoo.com).