The International Panel on Climate Change (IPCC) has concluded that it is 90% probable that global warming is caused by manmade greenhouse gases largely emitted since industrialisation. The red curves in Figure 1 show how the average surface temperature of the Earth has risen as time has passed since 1850.

Much effort has gone into modelling the average surface temperature of the Earth: the results of the calculations are shown by the grey curves in figure 1. The models represent well the observed rise in temperature if the effects of greenhouse gases are included (figure 1c). But the observed rise in temperature cannot be reproduced without greenhouse gases (figure 1a). So either global warming is man-made or these models, and with them the IPCC conclusions, are wrong.

For the IPCC to be wrong, these models must be incorrect, despite the many years of research effort spent developing them. Those of us accustomed to dealing with complicated computer models know that sometimes they can arrive at wrong answers so let's suppose that the models are indeed wrong. The next question is what can be causing the global warming? That is, we need a new effect which has not yet been discovered to account for the observed rise in temperature. So we need two things to occur before we can declare that the IPCC is wrong. Firstly, the modelling has to be incorrect and secondly, we need an as-yet undiscovered effect to account for the observed global warming. The IPCC put the probability of such a scenario at about 10% so that there is a 90% probability that global warming is man made.

Are cosmic rays the culprit?

In recent years much publicity has been given to the work of two groups of scientists - E Palle Bago and CJ Butler from the Armagh Observatory in Northern Ireland and N Marsh and H Svensmark from the Danish Space Institute - who observed that the amount of low altitude cloud seems to follow changes in the rate at which cosmic rays fall on Earth. The blue curve in figure 2 shows the seasonally-adjusted average fraction of the Earth covered in low cloud plotted against time since 1983, when satellite measurements began. The red curve shows the rate at which cosmic rays fell on Earth over the same period. The close correspondence between these two curves led the groups to propose that cosmic ray particles falling on the Earth's atmosphere influence cloud cover. Such particles produce ionizing radiation in the atmosphere; under certain conditions water droplets condense on the ions formed, a discovery for which CTR Wilson was awarded the Nobel prize for Physics in 1927.

Is this the "New Effect" which is needed to produce global warming if we assume that the models of the effects of greenhouse gases are wrong?

Clouding the issue

The observation that low cloud cover decreases as the cosmic ray rate drops led to the proposal that ionization from cosmic rays is responsible for the production of clouds i.e. less ionization leads to less cloud. The rate at which cosmic rays fall on the Earth has been observed to decrease over the past century or so. This has been discovered by various methods, for example examining polar ice cores to measure how the radioactivity produced by cosmic rays has changed with time. So if cosmic rays produce clouds and there are fewer cosmic rays nowadays we will have less cloud cover than we once had, allowing more of the sun's heat to penetrate to the Earth's surface and produce the observed global warming. If the proposal is true, it implies that we are wasting our time cutting greenhouse gas emissions to combat global warming. We could simply carry on with business as usual until the cosmic ray rate returns to normal. That makes the proposal very significant.

Before we accept this far-reaching conclusion we must apply some caution. We cannot accept that one effect causes another just because they occur at the same time. The number of cases of sunburn and the consumption of ice cream both increase in summer but you cannot conclude that eating ice cream causes sunburn. For this reason we set out to try to prove one way or the other whether changes in the rate at which cosmic rays fall on the Earth could significantly influence cloud cover and so provide the undiscovered effect needed in the event that computer climate models are wrong.

Changing times

Why did the rate at which cosmic rays fall on the Earth change from 1985 to 1990 (see figure 2)? The sun spews out streams of charged particles, as well as heat and light. The magnetic fields from these charged particles deflect some of the cosmic rays entering the solar system from surrounding space, i.e. galactic cosmic rays. During its periods of high activity the sun emits more such particles. So when the sun is highly active, more galactic cosmic rays are deflected and there is a decrease in the rate of cosmic rays reaching Earth.

From 1985 to 1990 the sun passed from a period of low activity to a period of high activity causing the decrease in cosmic ray rate shown by the red curve in figure 2. By 1995 the sun had relaxed back to a state of low activity and the cosmic ray rate returned to its former level.

In addition, the Earth has a magnetic field that also deflects cosmic rays. These deflections become progressively more severe moving away from the Equator. As a result, the dip in the cosmic ray rate becomes deeper further away from the Equator.

The first test of the potential "New Effect" that we did was to see if the decrease in cloud cover at the cosmic ray minimum in 1990 became more pronounced away from the Equator. This should happen if cloud cover is influenced by ionization from cosmic rays. We found that the dip in cloud cover does not change significantly in moving from the Magnetic Equator (figure 3). The line labelled NM, from neutron monitors used to measure the cosmic ray rate, shows how the depth of the cosmic ray dip increases on moving away from the Magnetic Equator (i.e. moving from right to left) . The dip in cloud cover should follow this line if cosmic rays influence cloud cover. So this observation does not corroborate the conclusion that changing cosmic rays are a significant cause of cloud cover. From a statistical analysis of figure 3 we showed that no more than 23% of the dip in the cloud cover in figure 2 can be caused by the decrease in the cosmic ray rate. Climatologists will be able to use this limit to constrain the contribution to global warming from changes in the cosmic ray rate.

The second test we performed was to study short-term changes in the cosmic ray rate to see if they could produce short-term changes in cloud cover. None were observed and this again provided no evidence to corroborate the proposal that cosmic rays are connected with cloud formation, implying that cosmic rays do not contribute significantly to global warming.

To conclude, we could find no evidence that changing cosmic ray rates contribute significantly to changing cloud cover and thereby to global warming. Therefore, cosmic rays are not the new effect that would allow us to declare that the IPCC is wrong. So it is back to cutting our greenhouse gas emissions if we are to avoid the future problems foreseen from climate change.