Fake News and 2017 Near-Record Temperatures

Guest essay by Dale Leuck

“Fake news” is the process of misleading the public through an inaccurate or incomplete depiction of reality, either deliberately or unintentionally. Often the fake news involves the use of statistics, particularly graphic illustrations of complex processes, as discussed in a February 2017 Forbes article. Such has long been the case with the issue of “global warming”, as, for example, in a January 2018 New York Times article, in which the following graph was presented, showing a 1.2 degree Celsius increase in annual global surface temperature data since a “base” period of 1880-1899. The chart is deceptive both for what it contains, what it hides, and what it excludes. Thus, the New York Times article is fake news.

As NASA, GISS notes at the top of it’s webpage, the data are from “…current data files from NOAA GHCN v3 (meteorological stations), ERSST v5 (ocean areas), and SCAR (Antarctic stations),” where NOAA is the National Oceanic and Atmospheric Administration.

First, for what the above chart contains. Kip Hansen, at “Watts Up With That” notes the Time’s doctoring of the NASA GISS data. The data, updated through December on January 17, are available in both text and csv formats, at the NASA, GISS website under the heading “Tables of Global and Hemispheric Monthly Means and Zonal Annual Means.” Clearly stated under that heading is that the data are “…deviations from the corresponding 1951-80 means. Nowhere on either the NASA, GISS or the NOAA websites are such data presented as deviations from 1880-1899. Kip Hansen explains why, and why the New York Times doctored the data:

“NASA never ever made a graph of global temperature anomaly with a base period of 1880-1899 — two decades, during which Global Average Temperature and its [nearly imaginary] anomaly are based on rough guess-work at best.    The NASA standard is 1951-1980, the usual 30-year climatic period.   The clever little eager-beavers at the Times have moved the zero point down to the 1880-1899 level thereby increasing the “anomaly” to nearly 1.2 °C.

Apparently, NASA’s official figures of “1.62 degrees Fahrenheit (0.90 degrees Celsius) warmer than the 1951 to 1980 mean, according to scientists at NASA’s Goddard Institute for Space Studies (GISS) in New York” did not seem “hotter” enough, so they felt it necessary to sex-it-up.”

Ed Straker promptly (AT, January 19) identified “adjustments” in raw NOAA data that hide a reality of little or no temperature increases, since 1890, and, as well as periods of temperature declines, including from the late 1990s, until about very recently. The USHCN data, of which the below chart is comprised, is from the U.S. Historical Climatology Network, and quantify “…temperature changes in the contiguous United States,” which is to say that it excludes data from Alaska, Canada, and the entire rest of the world, including the oceans.

Moreover, the percentage of missing station data that is “fabricated” has been rising in recent years, reaching 42 percent in 2016, in an article on Tony Heller’s Real Climate Science website, with discussion starting at the sixth chart down. The remainder of the article is an esoteric discussion of NOAA’s defense of using such data. The important point is that only a small percentage of the public is even aware that data are adjusted at all. And if adjustments of such magnitude occur for the continental United States, how reliable are presented data for the rest of the world at representing temperatures? This is only one of many questions a blue-ribbon appointed congressional committee should investigate.

Annual data itself hides extremely important information. Charting the NASA, GISS monthly data, for example, from 2013 through December 2017 reveals 2016 to have been a record hot year entirely because of a very hot period the first four months early that year, a phenomenon that began in late-2015. Notice in the chart below that temperature anomalies the first four months of 2017 were significantly below those of 2016; that the June anomaly was below those of the previous four years; and anomalies from August through November were below those of 2016. So, nine of the twelve months of 2017 were below those of 2016; one month of 2017 was below those of the previous four years; and one month (September) was below that of three of the previous years (2013, 2014, 2016). One could argue that the real “anomaly” in this data is the aberration of a few months of hot temperatures in late-2015 and early-2016, as 2017 anomalies appear on the same downward trend begun in late-2016.

Indeed, the graphical depiction of alternate facts, hidden by the main stream media such was the New York Times, supports such a more benign conclusion. The below chart depicts monthly anomalies, centered on the NASA, GISS 1951-1980 base period, but only from January 1998 through December 2017, a period some have argued, such as Sheldon Walker (January 17, 2018), has been a hiatus in global warming, and others that warming began again after 2012. While Walker uses an acceptable statistical technique to correct for what is known as “autocorrelation,” this is unnecessary to reach the conclusion of a hiatus if one only removes the “anomalous” data for the months of December 2015 through April 2016, above the red line. Intuitively, the anomalies for the succeeding months appear only marginally above the range of remaining values between January 1998 and November 2015 around the 1998 – 2017 average.

Aside from all of the above, the New York Times, other main stream media, and indeed many climate researchers themselves exclude exculpatory data in their quest to prove global warming. Surface temperatures have only been widely recorded since 1880, the year the NASS, GISS data set begins, and as noted earlier data for the earliest years are considered sufficiently unreliable that NASS, GISS uses an average calculated over 1951-1980 as its base for comparing anomalies. Also noted is the paucity of data from some stations in the continental United States, not to mention the rest of the world and a significant part of the globe that is not covered because there are no measurement stations. Moreover, the some-120 years dubiously reliable data is therefore not sufficient for drawing conclusions over geological time measured in terms of thousands of years.

A September 16, 2017 AT article discussed some of the above issues, starting with the need to show current temperatures setting records since at least early civilization. The 1990 United Nations Intergovernmental Panel on Climate Change (IPCC) published such a chart showing temperatures during the twentieth century below those of the medieval warm period. By the 2001 IPCC report, a “hockey stick” shape showed modern temperatures at a record, largely attributed to the work of Michael Mann and his use of proxy data, itself of dubious reliability, as well as data “adjustments of the sort summarized above, and recounted in the AT article.

With all of the complexity of the topic and potential costs of policies, there need to be more hearings, such as the one chaired by congressman Lamar Smith, of the Committee on Science, Space, and Technology. That hearing featured Judith Curry, John Cristy, Michael Mann, and Richard Pielke, all experts in the field. Appointment of a blue-ribbon commission would also seem advisable.

Dale Leuck retired from over 38 years of research and forecasting for the U.S. Dept of Agriculture in September 2017, having completed his Ph.D. in Agricultural Economics at the University of Tennessee in 1979.  He has written some for “American Thinker”, a farm magazine, and to various senators and congressmen.