The Global Temperature Records

Every January, the global mean temperatures for the previous year are announced.  Four institutions worldwide maintain running temperature records, and sometimes they draw slightly different conclusions.  That was true in 2017, as one institution called it the second warmest year on record while the other three called it the third.  (Last year, I had predicted it would be the third warmest year.)  So who are these four institutions, and why don’t their records produce uniform agreement?

Two of the four institutions that maintain temperature records, NASA’s Goddard Institute for Space Studies (GISS) in New York and the National Oceanographic and Atmospheric Administration (NOAA, headquartered in Silver Spring, Maryland) are U. S. government agencies.  The other two agencies are the British Meteorological Office and the Japan Meteorological Agency.  These records, based primarily on thermometer readings from many weather stations and ships worldwide, enable analysis of local temperature changes over time as well as changes in the global mean temperature.  When talking about global warming in general terms, though, the annual mean variations in global mean temperature are usually presented.  NASA GISS reported a global mean temperature anomaly of 0.90ºC (1.62ºF) relative to the mean 1951-1980 temperature, their second largest anomaly behind 2016.  NOAA had a result of 0.82ºC, the Met Office 0.70ºC, and Japan 0.73ºC, when normalized to their 1951-1980 means.  These three agencies had 2017 third in temperature, between 2016 and 2015.  The reason these are different has to do with the methodologies that the four organizations use.  A detailed explanation of the different methodologies can be found here. I will summarize the NASA GISS methodology by saying that it is more aggressive in counting ocean temperatures in the northern part of the world, where direct measurements are scarce.  The rationale behind this approach is that this part of the world has been the most sensitive to the influence of global warming so far.  This also explains why the GISS record tends to have higher numbers for the global mean temperatures than the other records do.

You might also be wondering why temperatures are expressed in terms of an anomaly, rather than a specific value for the temperature.  The reason for that is that it is easier to measure the change in temperature at a specific location, after accounting for factors like the urban heat island effect or proximity to ship engines that might skew the specific value of the temperature, than it is to define the specific value of temperature at a site unambiguously over time.  Furthermore, in regions where coverage is sparse, the average change in measured temperature is more likely to accurately represent that of the full region than the average measured temperature itself.

Figure 1

Figure 1 shows the plot, from 1890 to the present, of global mean temperature anomalies as determined by all four agencies normalized to their mean values between 1951 and 1980.  (GISS is black, NOAA is red, the Met Office is green, and Japan is blue.)  While they don’t agree on the exact numbers, they certainly do reflect the variations in temperature very similarly, in terms of both year-to-year shifts and the underlying long-term trend.

Regardless of which data set you look at, a very important conclusion can be drawn.  As 2017 was not significantly influenced by an El Niño event like 2015 and 2016 were, it is the warmest year on record without such an influence.  As I’ve discussed in a previous blog entry, the El Niño/La Niña cycle is the primary driver of natural variability in the temperature record.  So keeping natural variability in mind, the sharp short-term surge in global mean temperatures that became noticeable in 2014 is continuing.

(For updates on new posts, please click the "Follow" button.)