There are two main theories that attempt to explain recent changes in climate: the first states that changes in atmospheric carbon dioxide (CO2) are the main driver, and the second claims that varying solar activity, amplified by corresponding changes in cloud cover, are most important. The first is promoted heavily by politicians and activists, based on computer modelling; as for the second, the most interesting work is quietly being pursued at CERN, the European Centre for Nuclear Research.
The solar amplification theory depends on the seeding of clouds by galactic cosmic rays (GCR), the numbers of which are in turn controlled partially by the strength of solar activity. When that activity is low, the solar magnetic field strength and solar wind are low, allowing more GCRs into the Earth's atmosphere, thereby creating more clouds, which reflect solar energy back into space. Low solar activity means slightly lower energy from the Sun, and less of this gets through to the Earth's surface - the cooling effect is thereby amplified. The reverse is the case when activity is high.
The challenge is to identify and describe the mechanism that allows GCRs to create more clouds. A first step in this direction was taken with the successful SKY experiment by Henrik Svensmark in Copenhagen, and then expanded three years ago with the CLOUD experiment at CERN - a collaboration between scientists from 19 institutes in ten countries.
The intention is to use particle beams from the accelerators to simulate cosmic rays entering the lower atmosphere under various conditions - all carefully controlled and monitored. Initial results have proved encouraging, and so the experiment is being ramped up, and the new specifications are labelled CLOUD09. The current situation has recently been described by CERN's Jasper Kirkby at a CERN Colloquium on 4 June 2009. A pdf of notes by Kirkby from the colloquium is available here.