Since the past century, global mean temperatures have increased. Extreme hot temperatures have become more frequent, precipitation patterns have changed and several other variables indicate that climate change is happening. Are we responsible for these changes? How can we tell whether these changes are due to the human impact or a result of natural variations?
Tett et al. (1999) answer these questions on the example of the observed global temperature change. Possible causes for changes in global mean temperature are changes in greenhouse gas concentrations, man-made aerosols, volcanic aerosols and solar irradiance. Climate model simulations can estimate how the climate would be with and without the effects of the different possible causes. Global temperatures from model simulations where the possible causes vary realistically over time (e.g. increasing greenhouse gas concentrations) are the responses to the possible causes (signals).
In the method used by Tett et al. – called optimal fingerprinting – observations of global temperatures are assumed to be a linear combination of the responses to the possible causes and the internal variability of the climate system. In a linear regression framework, the authors test whether the signal is present in the observations. If this is the case, the respective cause contributes to the observed change. The results show that changes in volcanic aerosols and solar irradiance alone are unable to explain the temperature changes over the twentieth century. Anthropogenic changes in greenhouse gas concentrations contribute to the observed warming.
The analyses of Tett et al. are based on a method that has been applied to detect a human influence on many other variables of the climate system. Due to its wide application, the method is constantly under improvement. The improvements include the statistical methods and the use of longer timeseries and larger samples of climate models, which reduces possible errors in model simulations. However, detection of a human influence on the climate system is not a purely statistical task. The effects can be explained by a physical mechanism: The higher greenhouse gas concentrations are, the more infrared radiation is absorbed by the atmosphere. This leads to a warming. This effect has already been described over 100 years ago by Arrhenius, who also suggested that emissions of greenhouse gases could cause global warming (Arrhenius, 1896). In the last fifty years, we all have actively participated in the large-scale experiment to validate his hypothesis.
Tett, S.F.B., Stott, P.A., Allen, M.R., Ingram W.J., and Mitchell, J.F.B. (1999): Causes of twentieth-century temperature change near the Earth’s surface. Nature, 399.
Arrhenius, S. (1896): On the influence of Carbonic Acid in the Air upon the Temperature of the Ground. London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science (fifth series), April 1896. vol 41, pages 237–275.
Picture: Butterfly effect as a symbol for natural internal climate variability (see, e.g., Ch. 11 of IPCC AR5)