MaxCon 1.0 Model

[duwayne]

I've copied the post below from the Cycle24 board in case that board disappears some time in the future.  It contains my Global Temperature forecast made in 2007.  The post itself is from 2018 and is a summary of previous posts dating back to the original predictions from the model in 2007 which haven't changed except for being adjusted to allow comparison to the current UAH anomalies. The current reported anomalies use 1991-2020 as the base years.

______________________________________________________________________________________________

"My MaxCon 1.0 model for predicting global temperatures is described in detail in many posts on the “Global Warming Temperature Prediction” thread. That thread started in 2008 but the model actually dates back to 2007.

In 2007 some alarmist predictions were calling for global temperatures to increase by 10 degrees centigrade by the end of the century if fossil fuel burning wasn’t stopped right away. That got my attention so I decided to see if that was logical.

I started with a plot of the monthly Hadcrut3 anomalies shown below.


Since atmospheric CO2 changes slowly and steadily from year-to-year, it can’t be the reason for the short-term variations. So, I applied 7-year smoothing to get a better look at the trends.


What jumped out at me was that there were 2 very similar periods of approximately 30 years each with steadily rising temperatures preceded by periods of approximately 30 years where temperatures actually fell a little. I looked for a cause and found what I call the Ocean Current Cycle. Here is a plot of the Atlantic Meridional Oscillation which is a significant part of the Ocean Current Cycle versus global temperatures.

Not only is it a good match, it makes sense that this natural cycle would affect the warming from CO2 in a cyclical manner.

I used a periodogram to confirm the existence of a 60-year cycle in the global temperature data.

The pattern of global temperature change is logical and clear. When the Ocean Current Cycle is in the cooling phase, temperature trends are flat because the warming is offset. When the Ocean Currents are in their warming phase, global temperatures grow at a rate of about 0.5C per 30 years which includes the CO2 warming and the Ocean Current warming. (I'm giving CO2 all the credit for the general warming trend but that's not a certainty.)

Applying this to the future and assuming that atmospheric CO2 continues to grow at rates consistent with the last 30+ years my prediction was that the 1977-2007 uptrend would end in 2007 and it would be followed by 30 years of flat temperatures (2007-2037). Then an upleg over the next 30 years (2037-2067) would show a 0.5C temperature increase. Then temperatures would be flat for 2067-2097. Temperatures at the end of the century will be about 0.5 to 0.6C above the 2007 uptrend level.

In 2008 using the technique described above I predicted an average global temperature for 2007-2037 of 0.3C based on the UAH anomaly 2007 trend value calculated from the 1977-2007 uptrend. Since then the UAH anomalies have been revised using a more recent base period. Using the UAH6 anomalies and the identical technique my prediction is an average temperature for 2007-2037 of 0.2C.

With 2018 almost finished, it’s now a certainty that the average UAH6 anomaly for the first 12 years of the 2007-2018 flat period will be 0.2C."

[duwayne]

Below is a copy of a February 2021 post from the Cycle24 board MaxCon 1.0 thread.  It explains how an update by UAH of their base years required an update to my MaxCon 1.0 prediction to allow the numbers to be compared to the current UAH anomalies. 

_______________________________________________________________________________________________

"My Maxcon 1.0 Model is explained on this thread and I am using it to predict the UAH6 average anomaly for the 30-year 2007-2037 cool ocean cycle period. The prediction was temperatures would remain flat at 0.2C. 0.2C was the 2007 trended temperature for the 1977-2007 warming period - the 30-year warm ocean cycle.

Roy Spencer has announced a change in the base period for the UAH6 calculations from 1981-2010 to 1991-2020. This means all historical and future UAH6 anomalies will be (rounded to the first decimal point) 0.1C lower than they were and will be. To remain comparable, my 0.2C prediction is now 0.1C average for the 2007-2037 cool ocean cycle period.

On the updated basis, the actual average UAH6 anomaly for the 14-year 2007-2020 period is 0.1C, as predicted. The December 2020 anomaly was also 0.1C.

After 14 years, so far so good. The prediction was based on atmospheric CO2 growing at the rate it grew prior to 2007. The rate has been somewhat faster over the last 14 years despite the talk. We'll see what happens in the future, but a continuation of current growth rates could push the average temperature up by 0.1C."

[duwayne]

My prediction made in 2007 is that UAH global temperatures would remain flat for 2007-2037 at the 2007 ocean cycle driven global warming trend temperature.  Note that this trend temperature is not the 1-year average of the temperatures in 2007, but it is the 2007 temperature taken from a least squares trendline for the warm ocean cycle period from 1977-2007. Using the current UAH anomaly base that would give a 2007 trendline anomaly of 0.1C. 

We are now 14.5 years into the 2007-2037 forecast period. The June 2021 number is in. The average so far, nearly halfway through the period,  is 0.1C.

[code]

How is that model working out?

[duwayne]

Code, thanks for asking about the latest results of the MaxCon 1.0 model. It’s doing very well.

The original MaxCon 1.0 based prediction was made in 2007. It predicted  that the average of the UAH monthly anomalies over the period 2007-2037 would be 0.1C, the same as the trend number for 2007 which was at the end of the 1977-2007 Ocean Current warming cycle.  The prediction was that anomalies would remain “flat” due to an expected 30-year Ocean Current cooling cycle which would offset  any upward forcing due to rising CO2 levels.

Original Prediction                                   0.1C

Actual through June 2022 (15.5 years)      0.1C

The 2022 average for the first 6 months of 2022 is also 0.1C and June itself is 0.1C.

A description of the Model is provided above in this thread and on the old Cycle24 site.

[missouriboy]

Jul 2, 2022 13:22:06 GMT duwayne said:

Code, thanks for asking about the latest results of the MaxCon 1.0 model. It’s doing very well.

The original MaxCon 1.0 based prediction was made in 2007. It predicted  that the average of the UAH monthly anomalies over the period 2007-2037 would be 0.1C, the same as the trend number for 2007 which was at the end of the 1977-2007 Ocean Current warming cycle.  The prediction was that anomalies would remain “flat” due to an expected 30-year Ocean Current cooling cycle which would offset  any upward forcing due to rising CO2 levels.

Original Prediction                                   0.1C

Actual through June 2022 (15.5 years)      0.1C

The 2022 average for the first 6 months of 2022 is also 0.1C and June itself is 0.1C.

A description of the Model is provided above in this thread and on the old Cycle24 site.

Congrats.  Perhaps NOAA would like to review your results in comparison to those of their in-house wizards.  In addition to accuracy, I think that yours is also possibly less expensive. 

[duwayne]

The original MaxCon1.0 Global Warming Model which I developed in 2007, as noted at the time, was based on the presumption that atmospheric CO2 growth would continue similar to the growth in the years prior to 2007. I have noted for some time now that CO2 was growing considerably faster than the presumed rate and as a result the MaxCon1.0 global warming predictions would likely turn out to be on the low side. This post lays out the new MaxCon1.1 Model which is MaxCon1.0 adjusted for the faster CO2 growth.

MaxCon1.0 is described in detail earlier on this thread. The Model predicts that UAH global anomalies will be flat during the 30-year cool Ocean Cycles and grow by 0.6C over the 30-year warm Ocean Cycles for an average of 0.3C over each 30-year Cycle.

In the year 2000, atmospheric CO2 had grown at a rate which would result in a doubling in 200 years.   Looking back over the past 11 years the rate would result in a doubling in 117 years. I’ve decided to adjust the Model warming to a 150-year CO2 doubling rate which is less than the most recent rate and this results in an average warming of 200/150 times 0.3C or 0.4C per 30 years.  This is an increase of 0.1C per 30 years for MaxCon1.1 versus the MaxCon1.0 Model.

This works out to be warming of 0.1C of warming over the cool Ocean Cycle periods (rather than flat for MaxCon1.0) and 0.7C during the 30-year warm Ocean Cycle periods rather than 0.6C, all due to the higher growth in atmospheric CO2.

The resulting UAH anomaly predictions for the end of the 30-year cooling/warming periods then are;

                                                          MaxCon1.0 MaxCon1.1

2007-2037 (Cooling Ocean Cycle)            0.1C      0.2C
2037-2067 (Warming Ocean Cycle)          0.7C      0.9C
2067-2097 (Cooling Ocean Cycle)            0.7C      1.0C

The predicted UAH anomaly average over the current 2007-2037 period now is 0.15C (0.1C at the beginning, 0.2C at the end). Since I round everything off to the nearest tenth of a degree and not pretend that I can predict things to the nearest 100th of a degree, my official UAH anomaly prediction for the average for 2007-2037 using MaxCon1.1 is now 0.2C, up from 0.1C.

As a result of the adjusted warming rates, for the year 2100 UAH anomalies are predicted to be 0.9C higher than the current level.

If atmospheric CO2 continues to grow at the even faster recent rate, the anomaly predictions for 2100 would be a little higher. In any event, this anomaly growth would be about 1/3 of what was being predicted by the “Climate Scientists” back when I came out with my original predictions.

[Sigurdur]

Duwayne: Smart man.

Instead of sticking to your guns, you adjust for current conditions.

That is SUCH a novel idea!!

I certainly look forward to how close your model is in 10 years, God willing.

I have been searching for global cloud coverage variability. Missouri had some success a while back finding those numbers.

I personally can't discount Svenarks cosmic theory.

Another item of interest is Stratospheric H2O vapor. The amount does have climate implications.

So much I don't know, keen on finding reliable readings.

[neilhamp]

Hi Duwayne,
Following your model closely.
I think AMO is also closely linked to Arctic sea ice extent.
We have seen a recovery in sea ice over the past two years
Neil