• Tectonic activity: The arrangement of continents plays an important role in determining the climate, and if the continents move, the climate may very well change.
• Orbital variations: The ice age cycles of the past few million years are driven by changes in the orbit of the earth about the sun. The earth's orbit has important variations, with time periods of approximately 25,000, 40,000, and 100,000 years.
• Solar variations: The sun is the primary energy source for our climate. As the output of the sun changes, so does the climate.
• Volcanoes: They inject ash and aerosols into the atmosphere, which reflect incoming sunlight. A strong eruption can cool the Earth for several years.
• Internal variability: The climate system is complicated, and internal modes of variability exist. The most well known one is the El Nino/Southern Oscillation (ENSO). During the El Nino phase, the earth is much warmer than during the opposite phase, the La Nina.
• Finally, there is a new player in the climate game: human-emitted greenhouse gases. These gases trap upwelling infrared radiation, thereby causing increases in the temperature of the surface.

If we look at the warming of the last few decades, we can immediately rule out tectonic activity and orbital variations -- they are much much too slow to account for warming over mere decades. We can rule out volcanic eruptions for a similar reason -- they affect the climate for only a few years. Thus, volcanic eruptions are also likely unrelated to the several-decades-long temperature increase we are experiencing.

We can rule out solar variability because we have high-accuracy measurements of the output of the sun from satellites since the mid-1970s, and we have not seen the increase in solar output necessary to explain the temperature increase. This is not to say that solar is playing no role, just that it cannot explain the majority of the observed warming.

Internal variability is the hardest to evaluate. We know that ENSO significantly changes the Earth's temperature, and so long-term ENSO-like variation is something we have to consider. However, nobody has yet put forth a viable mechanism or shown data that such a long-term cycle exists. In the absence of any evidence supporting it, we conclude that it's likely internal variability is playing a minor role in today's warming. Clearly, future research might cause us to re-examine this conclusion.

Finally, we have greenhouse gases. In this case, things work out well. Both the timing and magnitude of today's warming are well-explained by greenhouse gases.

This is why scientists conclude that humans are likely responsible for most of the warming of the last few decades. Greenhouse gases provide a reasonable explanation for the warming, while no other factor can explain the entire warming (though other factors, such as solar, might be playing a minor role). In the IPCC report, they attach the word "likely" to the statement about the importance of greenhouse gases, which denotes about 75% confidence that the statement is true. This takes into account our imperfect knowledge of the atmosphere, in particular with regard to internal variability, and that future work might lead to revisions of our views.

Finally, note that this conclusion does not come solely from GCMs. It sits on a firm foundation of peer-reviewed studies using a wide range of techniques. One of the things that gives us confidence is that the studies all paint a consistent picture of today's warming.