D. There are Biases?
A significant bias may be associated with a arbitrarily selected starting Period or Zero line HadCRUT4 used the time period between 1961 and 1990 as the reference period or zero line. A selection of a shorter or longer period to establish the zero line will significantly change the amount of temperature anomalies (changes).
The Tokyo Climate Center plotted the time period of 1890 to 2018 and shows that a temperature anomaly of 0.67°C occurred in 2018 above the 20th century average. The 2018 anomaly was .31°C from the average of the 1981-2010 period. This illustrates that changing the zero line dramatically affects the anomaly. It also exemplifies the problems associated with selecting a particular starting date or zero line. For example, if one selected 1000 years ago as the reference point or zero line, the current anomaly would be about a minus 0.1°C. That is, it was warmer 1000 years ago than today. This is shown in Figure 8 from [Christiansen, B. & Ljungqvist, F. (2012) The extra-tropical Northern Hemisphere temperature in the last two millennia: reconstructions of low-frequency variability; Climate of the Past, Vol 8, pgs. 765-786.]
The temperature plots that go back 500 million years ago also illustrate the arbitrary nature of selecting 1850. [Veizer, J. et al. (1999) 87Sr/86Sr, d13C and d18O Evolution of Phanerozoic Seawater. Chem. Geol. 161, 59-88; and Veizer J, Godderis Y, and Francois L, (2000) Evidence for decoupling of atmospheric CO2 and global climate during the Phanerozoic eon, Nature, Vol 408 pgs. 698-701. http://dx.doi.org/10.1038/35047044]
In the above Glen Fergus figure, along the horizontal scale, the far right hand side at point 0 is the present day and the far left hand side is 500 million years ago. The horizontal scale changes with each section that is why there is a double line between the sections. The plot is not based a logarithmic scale, but a linear scale within each section. But, each section has its own scale. That is, every double vertical line represents change in the scale. For example, the far right hand side, the scale is from 0 to 20,000 years representing 20 thousand years. The next section illustrates a scale from 20,000 to 1 million years. The next section is from 1 million to 7.5 million, the next is from 7.5 million to 75 million and the final section is from 75 million to 540 million. It shows that for the last 11 thousand years the earth has been in an exceptionally stable period.
This illustrates that in the HadCUT4 figure the selected zero line between 1961-1990 is essentially touching the same point as 2019. The above Glen Fergus graph illustrates that the HadCUT4 anomaly of 0.8°C it is well within the normal and expected fluctuations. The Glen Fergus graph also confirms that selecting the period going back 11 thousand years would make more sense since this time period shows a stable pattern. If someone wanted to show deviations from the stable pattern, it would be far more meaningful to select the entire 11 thousand year time period rather than an arbitrary 30 year period.
Professor Philip Jones is the co-author and observer for the HadCRUT4 data and analysis. He is professor at the University of East Anglia as referenced in the above figure. See (https://people.uea.ac.uk/p_jones) In an interview with BBC News on February 13, 2010 he referenced that the following was significant:
This table shows that during the 21 years between 1860 and 1880 the temperature was rising at a rate of 0.163 °C per decade. During this time the CO2 concentration in the atmosphere was approximately 280 ppmv. [See https://commons.wikimedia.org/w/index.php?curid=49535289]. The rate of increase in temperature for the period of 1975 to 2009 was lower at 0.161 °C/decade. During this industrial period the CO2 concentration was 330 to 380 ppmv. During the period of 1860 to 1880 the amount of CO2 emissions from fossil fuels was near zero. [CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=438191.%5D The fossil fuel CO2 emissions between 175 and 2009 were about 3 to 6.5 Gigatons of Carbon/year. This illustrates a disconnect between both CO2 and Fossil Fuel Emissions and temperature rise.
This Illustrates that the same data can be viewed in different ways to promote or dispute various positions. These exceptionally narrow time periods are not representative of the Earth’s geologic history. They are unreliable in making meaningful projections.
Using Diurnal Temperatures as the Average Temperature. The HadCRUT4 method as well most others use diurnal temperatures as the average temperature. The diurnal temperature is the sum of the highest temperature of the day and the lowest temperature of the day divided by two. This is not the average temperature. The average temperature is the total of all of the temperature measurements during the cycle, often a 24 hour cycle, divided by the number of measurements. That is the definition of average.
If there is an unusual spike in the high or low temperatures caused by some external event that would show up significantly changing the diurnal temperature but not the average temperature. Quorvita has concerns that true average temperatures were not used even when they have that data. Quorvita has not seen actual data in order to determine if there is a problem. However, temperature data from the Gale Crater on Mars illustrates there could be a potential problem as shown in the table below.
The average of the all the monthly highs is shown at -5.7°C and the average of all the monthly lows is shown as -78.5°C. Adding those values together and dividing by two equals -42.1°C. Next, taking the record high (20°C) and the record low (-127°C) and adding them together dividing by two yields an average temperature of -73.5°C. That a deviation of about 100%. It is recognized that record numbers and diurnal numbers are not the same. But neither is the comparison of the average temperature with the average diurnal temperature.
The fact that anomalies are involved makes the issue more complex. In any event, this point is a concern and a study should be conducted to determine if there is a statistical difference and if so how much.
Averaging Air Temperatures with Water Temperatures is not appropriate. The air has a very low heat capacity (ability to hold thermal heat) as compared with water. The heat capacity of air is 3300 times lower than the heat capacity of water. Hence comparing air to water is not appropriate without factoring in the heat capacity difference. However, the surface air temperature is being used as a proxy for surface land temperature. This makes the comparison between land temperature anomaly and ocean temperature anomaly more appropriate. But, the average heat capacity for land is slightly lower than water. That is, land can hold about 25% less heat that water. Therefore, averaging surface land temperatures with ocean temperatures should be adjusted, i.e. weighted. The land temperature anomaly portion should be reduced by 25%.
As discussed, earlier, the use of surface air temperatures as a proxy for land temperatures is not appropriate without applying a discount factor. This discount factor must be based on a statistical comparison study between surface air temperatures and surface land temperatures.