Jerry Lundry: Annual Average Temperatures for the Forty-Eight Contiguous United States, and Atmospheric CO2

This is a guest post by Jerry Lundry

Two plots are presented for annual average temperature in the United States Historical Climate Network (USHCN). This data set is highly regarded by some in climate science and is sometimes used as a surrogate for world-wide temperatures. Among its attributes are its coverage of a large land mass (the forty-eight contiguous United States), dense coverage of that land mass (1218 stations), and records that are complete to 1912 and missing only about eighty stations back to 1895. Temperatures for all stations are also provided for 1908.

In 2012, the author downloaded and produced annual average temperatures for this data set. The first figure below provides average annual temperatures for 1908 and 1912-2011. The curve faired through the data is a standard Excel sixth-order polynomial. This curve shows minima in years 1914 and 1970, and maxima in years 1940 and 2004, give or take a year or two.

Figure 1

The second figure below has the same type of results for the period 1895-2011, but with temperatures for all years removed from the analyzed data subset for those stations missing one or more temperatures. This broader span of data shows minima in about 1908 and 1972, and maxima in 1942 and 2003.

Figure 2

The author prefers the second plot, as its longer period seems to better define the temperature trend in the early years, but would like to read opinions of others on his preference, from both those who agree and those who do not.

Regardless, both figures show a current downward trend in temperature that goes largely unreported, at least in the popular press. Both also reflect two climate cycles that, again, are rarely, if ever, mentioned in the popular press. The maxima and minima seem to indicate the so-called thirty-year cycle (the author thinks of thirty years of warming followed by thirty years of cooling as a sixty-year cycle, but this is a nit) that seems to be related to the Atlantic Pacific Decadal Oscillation.

The effect of the second climate cycle is also apparent. It is a general increasing trend between the earlier extremes and the later extremes in both figures. This seems to reflect the 1,400-year climate cycle (also referred to as the 1,500-year cycle by some).

The temperatures that determine this second cycle are derived from surrogates for temperature, as thermometers have been in use for less than three centuries. Various sources indicate minima in the periods 300-500 and 1600-1800, and maxima in the periods 200-400 BC and 1000-1200. Based on these results, one can project another maximum in the period 2300-2500.

Finally, the author wishes to present a third figure. This figure below shows atmospheric CO2 data for 1895-2011, obtained from the web site of the Carbon Dioxide Information Analysis Center (CDIAC). Atmospheric CO2 has risen monotonically for this period except for the years around 1940, where it was approximately constant.

Figure 3

Relative to what we are told in the popular press, these three figures, viewed together, indicate to the author an obvious inconsistency. For the last three decades, atmospheric CO2 has been increasing at an almost constant rate. For the first two of those decades, temperature was also increasing. Yet, for the past decade, temperature has been decreasing.

This latter observation appears to contradict what we are told about the effects of human CO2 emissions on temperature. We are told human CO2 has caused increases in temperature with no mention of any other mechanisms. Further contradiction is provided by the observation that during the three decades 1940-1970, temperature was also clearly decreasing while CO2 was either constant or increasing.

From the observations in the preceding two paragraphs, the author can only conclude that, whatever the effects of human CO2 on temperature, they are over-powered by the natural effects related to the thirty-year climate cycle.

The author would greatly appreciate reading comments made by others about the data presented in this post and about the author’s conclusion, both those who agree and those who disagree. For the latter, he would greatly appreciate reading also the reasons for disagreement.