Home » Comment and analysis » James Hansen’s climate models versus observations 1958-­2015

James Hansen’s climate models versus observations 1958-­2015

James Hansen’s climate models versus observations 1958-­2015

Jan-­‐Erik Solheim, professor (emeritus), University of Tromsø, Norway

Projections based on climate models are used to convince politicians to reduce the emissions of CO2 to save the world from overheating. The latest warning is that a global temperature increase of 2.7K is the lower limit possible in 2100, if the emission control works as proposed for the Paris meeting.

I will describe how climate models developed by James Hansen et al. (1988) have failed to predict how the global temperature has changed until now.

If we cannot stop the building of more coal-fired power plants, those coal trains will be death trains – no less gruesome than if they were boxcars headed to crematoria, loaded with uncountable irreplaceable species.
James Hansen, 2007, in “Super Models, Old King Coal II, & Civil Disobedience”
Coalfired power stations are death factories. Close them. James Hansen, 2009. theguardian.com


Figure 1. Global temperature predictions based on 3 scenarios (A, B and C) as described in the text. The hatched area represents the maximum temperature in the present (Holocene – 6000 yr ago) and previous (Eemian-­120 000 yr ago) interglacials (1). The solid black curve represents observations.

James Hansen was Director of NASA’s Goddard Institute for Space Studies (GISS) in New York. He was the first scientist to warn the US Congress of the dangers of climate change. This was based on climate models published by Hansen and co-­‐ workers in 1988 (1). Simulations with GISS model II were presented, and 3 projections for future global temperature was shown (figure 1):

A: Yearly CO2 –emissions increase exponentially with 1.5 per cent per yr. B: CO2 emissions are stabilized to a yearly constant increase.

C: Strong regulations lead to no increase of CO2 emissions after 2000.

In scenarios B and C some volcanic eruptions are strategically added. They cool the Earth for a limited time. In scenario A there are no volcanoes.

Hansen et al. estimated their uncertainty to be 0.13K, and wrote that a change of three times this uncertainty, i.e. 0.4K warming is the “smoking gun” which proves that humans warm the Earth. In their graph (figure 1) we find that scenario A showed 1K warming in 2005, while B showed 1K warming in 2010. Scenario C showed a maximum warming of 0.7K in 2005 and then cooling.

fig2Figure 2: Scenarios A, B, and C from Hansen et al. (1998)(blue, green and red curves), compared with HadCRUT4 estimated global temperature (black). The arrow suggests a possible prognosis (P) based on more emissions than in scenario A observed the last years. All the curves show running averages over 5 yr with reference period 1961-­90.

Figure 2 shows Hansen et al. scenarios compared with HadCRUT4 estimated global temperature until August 2015. Since the CO2 emissions now are increasing 3 per cent per yr (2), a possible higher estimate for expected global temperature in 2015 than scenario A, is marked with an arrow and letter P. This is about 1.6K increase, while the estimated observed value is 0.6K. The prognosis is 170 per cent too high.

An increasing amount of CO2 seems to be absorbed in the sea or biosphere (2), so even if emissions increase with 3 per cent yearly, the atmospheric CO2 increase is constant with about 2 ppmv per year. This is close to scenario B, which estimate

a temperature increase of 1.1K in 2015. This is about 80 per cent higher than observed.

However, the global temperature follows closely scenario C.

We have the remarkable situation that actual CO2emissions predict a higher temperature than scenario A, while the atmospheric CO2 content indicates scenario B, and the observations follow scenario C.

In a recent paper Hansen et al. (3) suggest that the missing increase in atmospheric CO2 in the atmosphere can be explained by the sharp increase in coal use in China and India. emission of Nitrogen from coal burned power stations in China and India. Fossil fuel emissions increase carbon uptake by fertilizing the biosphere via provision of nutrients essential for tissue building, especially nitrogen. More nitrogen leads to an increased growth of plants, which absorb more CO2. If these power stations are cleaned, and the nitrogen-­‐ emissions reduced, CO2 will again increase exponentially and the Earth warms more. Hansen et al. calls this a Faustian bargain. There are many other explanation of the temperature hiatus – but to deny people of cheap electricity, which leads to development, better health and longer life for most people on the Earth, may be the real Faustian bargain.



Figure 3. Extrapolation of the Hansen et al. scenarios A (blue) and B (green) until the year 2100, compared with the COP 21 goal of 2.7 K increase (red).

Estimated temperature increases since 1750 is shown as the red scale on the right. HadCRUT4 observations (black) are shown with extrapolations based on a simple model with a trend and periodic variations (magenta).

Before the COP 21 Paris meeting, it has been announced that it is too late to keep the global warming below the 2 K limit, but if all the announced reductions in the CO2-­‐emissions are done as promised, it is estimated that the global warming is kept below 2.7 K  (since start of industrialization in 1750?). In figure 3 this is shown as a red, broken line, together with linear extrapolations of scenario A

and B from Hansen et al. (1). This is compared with a simple model proposed by Syun-­‐Ichi Akasofu (4), consisting of a linear trend (ΔT=0.005 K/yr) recovery from the Little Ice Age in addition to a few periodic variations determined from the data itself. In this model small temperature changes are expected this century (5).

The (lower) limit of 2.7 K warming by 2100 proposed as a goal for the COP 21 meeting, means a temperature increase of 0.025 K/yr. The 2.7 K limit is based on unvalidated climate models. It has been shown by professional forecasters (6), that a simple no change benchmark test is far better for a long range forecast than the unvalidated IPCC climate models. They found that the errors from the model projections were seven times larger than the benchmark method, and that the best estimate is that the temperature in 2100 will be within ±0.5 K of the temperature of 2008, which was their benchmark year.

Another concern is that the upward trend observed over the last century and half might be at least in part an artifact of measurement errors rather than a genuine global warming (7).

One may then ask why the modest warming observed the last century and the small warming forecasted for this century should lead to such expensive measures as proposed for the COP 21 climate conference in Paris.


  1. J. Hansen et al., 1988, Global Forecast by Goddard Institute for Space Studies Three-­‐Dimensional Model, Journal of Geophysical Research, 93,9341-­‐9364.
  1. J. Hansen: http://www.columbia.edu/~mhs119/UpdatedFigures/
  2. J. Hansen m. fl, 2013, Climate forcing growth rates: Doubling down on our Faustian bargin, Environ. Res. Lett. 8 011006.
  1. S.-­‐I. Akasofu, 2010, On the recovery from the Little Ice Age, Natural Science, 2,11, 1211-­‐1224.
  1. J.-­‐E. Solheim, 2013, Signals from the planets, via the Sun to the Earth, Pattern. Recog. Phys, 1, 177-­‐184.
  1. K. C. Green, J. Scott Armstrong, W. Soon, 2009, Validity of Climate Change Forecasting for Public Policy Decision Making, International Journal of Forecasting, 25, 826-­‐832.
  1. R. McKitrick, & P. J. Michaels, 2007. Quantifying the influence ofanthropogenic surface processes and inhomogenities on gridded global climate data. Journal of Geophysical Research, 112, doi:10.1029/2007JD008465.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: