The temperature trend is not as simple as Garnaut makes out

Figure 2: Annual CO₂ measurements at the South Pole. The solid trend straight line is drawn from the first to last measurement of the time series. For each year the difference of the measurement from the trend line is shown in Figure 3. Source Scripps Institute.

Figure 3: Residual differences for each year of CO₂ measurement from the trend line. The three straight lines show periods of constant increase and their intersections in 1975 and 1998 mark years of significant change. The changes are coincident with the changes in temperature trends at the time of the Great Pacific Climate Shift of 1976-77 and the temperature break of the later 1990s.

The conclusion to be drawn from Figure 3 is that a break in the CO2 trend occurred at about 1975. This is coincident with the temperature series break seen in Figure 1 in 1975-76. It turns out that there is also a break in the humidity time series in 1975-76. All this occurred at the time of the GPCS.

So when you look again at Figure 3 there is a break in the late 1990s in the CO2 time series. This coincides with the apparent break in global temperatures seen in Figure 1. There are also breaks at this same point in time series for humidity and methane. The break in methane comes about as more water vapour (humidity) in the atmosphere reduces the methane concentration. (The atmospheric methane measurement series only started in the 1980s.) This is a four way coincidence not a random noise event in the global temperature series. This coincident break point has the signature of the breaks seen at the time of the GPCS. It marks another ocean overturning in the Pacific Decadel Oscillation and perhaps the start of a cool phase?.

The presence of the break in a series of atmospheric measurements as a result of ocean influences creates a fundamental problem for modelling future temperatures. The long ocean cycles, such as the Pacific Decadel Oscillation, are not well understood but are not driven by increasing CO2. Mechanisms connected to changes in the speed of the earth's rotation have been suggested but at present there is no way of forecasting these ocean changes. A further problem is that a change of the surface temperature of the oceans leads to a rebalancing of CO2 in the atmosphere thus further complicating the modelling.

This analysis illustrates the complicated interaction of the oceans and atmosphere. The failure of Garnaut to undertake a proper statistical analysis of the behaviour of the atmosphere suggests selective use of evidence to sustain his conclusions.

It also raises the question of the usefulness of computer models of the atmosphere for policy development if forecasts of future temperature trends and other important measures such as sea level changes are uncertain.