By David Archibald

[This looks more like a technical analysis of a stock than a discussion of physical processes to me.  Any time series will have a lower bound.  Selecting a few of those data points that seem to fall on a straight line and asserting that they are bounded by an undefined feedback mechanism seems arbitrary at best.  But I’m posting for discussion anyway.  Have at it.~ctm]

This post was on the subject of how consistent the lower bound of the UAH satellite temperature series is. The lower bound is defined by a total of six monthly data points plotting on the green line in this graph of near 40 years of data:

Figure 1: UAH Temperature of the Global Lower Atmosphere

To investigate this seemingly impenetrable bound let’s first detrend the data:

Figure 2: Detrended UAH Temperature of the Global Lower Atmosphere

The slope of the lower bound, shown by the green line in Figure 1, is 1.1°C per century. UAH calculates the slope of this temperature series as 1.3°C per century. The difference is probably due to the two big El Ninos in the second half of the record.

The lower bound of the UAH temperature range is well defined with six tests of it over the near 40 year period.

The next step is to examine the distribution of the detrended monthly data:

Figure 3: Cumulative distribution of monthly data points

Figure 3 shows that the detrended temperature record spends most of its time in a band 0.4°C wide from -0.4°C to 0.0°C. The excursions from that band are bigger on the upside than the downside.

Figure 4: UAH Temperature Distribution by 0.2°C band

Figure 4 shows the number of data points per 0.2°C band in the current UAH record. There is no long tail in the lower end of the series – consistent with it bouncing off the lower bound as shown in Figures 1 and 2. The record appears to be lumpier than it should be, particularly at -0.10°C to -0.06°C. Perhaps the lumps will even out when we have accumulated another couple of decades of data.

The lower bound in the record to date is determined by six data points:

March 1984 -0.51°

September 1985 -0.51°

August 1992 -0.45°

March 1993 -0.43°

May 2008 -0.26°

February 2012 -0.22°

Figure 5 shows what happens when we plot up just those five points:

Figure 5: The six points that determine the lower bound of the UAH record

As figure 5 shows, the lower bounding temperatures of the UAH record are far from a random walk. Some feedback mechanism is tightly controlling temperature once it departs too far from the preferred band. We will know that global warming is over once that bounding line is broken.

David Archibald is the author of American Gripen: The Solution to the F-35 Nightmare