Central England Temperature – Model

CSS-16 One of the many problems with the available global surface temperature record is the short time period. Ideology, accuracy, homogenization and areal coverage (especially in the 19^th century) would be other problems that we will not be discussing here. HadCRUT4 is the longest global record, starting around 1850. The NASA/GISS data set conveniently starts in 1880 (the last major cold period of the Little Ice Age (LIA)), making NASA/GISS the go to option for the Catastrophic Anthropogenic Global Warming (CAGW) alarmist crowd. But there is a longer term measured record that shows the cold of the Maunder, Dalton and Centennial solar minimums. None of which could be due to CO₂. The Central England Temperature (CET) begins in 1659, deep in the depths of the Maunder Grand Solar Minimum (GSM). The temperature fluctuates significantly throughout the data period. Those fluctuations are due to natural fprcings, not CO₂. Those natural forcings were active throughout this period. They were active pre-CET (review my Holocene Logic posts) and they will continue to active in the future despite the modelers insistence that CO₂ is the essentially the dominant driver. That would be the same modelers that have admitted that their models forecast way too hot. The relevant links are available in my OPS-55 – The State of Climate Science post.

#climatechange #delaythegreen #globalwarming #showusthedata

The model consists of three main drivers (Total Solar Irradiance (TSI (as a proxy), Ocean Cycles (the Atlantic Multi-decadal Oscillation) and atmospheric CO₂ concentration). The Pacific Decadal Oscillation (PDO) and the el Nino Southern Oscillation (ENSO) were also considered but their cycles could not be as easily modeled as the AMO. The AMO 20-year moving average Momentum (AMO_M20) fits an inclined sinusoidal curve very well, allowing an estimate back to the mid-1600s. The more recent PDO_M20 and ENSO_M20 can be added in and will be in a later post.

So, what were the results? The model correlated very well except for three temperature spikes at the beginning of each TSI_M20 increase out of the three solar minimums and the post 1990 period. The post-1990 period is the most complicated to explain, so I will leave that to the last. There are obviously additional natural forcings acting during the first three temperature spikes and CO₂ had virtually nothing to do with those temperature spikes. The obvious candidates would be the PDO and ENSO which are not currently included in this model. The volcanic eruptions (Tambora and Krakatoa) probably tempered the magnitude of the 1820 and 1900 temperature spikes. There may also be other short term natural forcings (global and/or local) that act during the early TSI_M20 increases that manifest as the sun, moon and planets interact through their various cycles.

The AMO, PDO and ENSO likely all play a role in the separation between the CET and model temperatures, post 1990 (as shown in the UAH and ocean cycle data). Another significant contributor to the separation is most likely “homogenization”. In the US, the difference between the measured and homogenized temperatures over the satellite period is roughly 1 °C. That is more than twice the difference between the CET and my model temperatures. So, how much of that homogenization is justified? The global difference may be lower, but the homogenization process is being applied globally and aggressively. For example, the measured temperatures at Calgary’s airport weather station have been declining since 1973 at a rate of 1.25 °C/century. After “homogenization”, Calgary’s “temperature” is magically increasing at a rate of 1.62 °C/century (a difference of +2.87 °C/century). That data (along with other global data) is presented in my CSS-13 – A Look at Homogenization post. And just for another perspective, the separation between the UAH (measured satellite data) and HadCRUT4 (homogenized surface data) has continually increased over the satellite period to almost 2 °C. You could almost argue that the difference between the CET and the model results was all “homogenization”, could you not? Are computer/spreadsheet/whatever models perfect? No, but using a simplistic, unscientific narrative that focusses on one small parameter (CO₂) of thousands is just ridiculous. There are obviously other natural factors acting on our climate system. Ignoring those other factors is much more dangerous than any of the beneficial warming that CO₂ (a FECKLESS Greenhouse Gas, CSS-7) might provide. Climate models are only as good as their programming. The new CMIP6 protocol has added additional solar forcings (cosmic ray flux and high energy particles) to the radiative forcing arsenal. The CMIP5 protocol was limited to the absolute value of the TSI alone (a minor solar forcing). During beta testing the Modern Temperature Record (MTR, 1850 to the present) could be modeled with no CO₂ influence. Is that the correct solution? It could be, but I suspect CO₂ plays a minor role and some CO₂ influence should be included. Remember, just because the model gives you an answer you like, does not mean the answer is right. Especially when the modelers have already conceded that their models run too hot (OPS-55 – The State of Climate Science). There is also that little problem that there is no empirical CO₂/Temperature data that shows CO₂ driving the climate on any statistically significant historical time scale. Empirical data is required for a theory, hypothesis or in this case the CAGW narrative to be considered scientific.