Patterns of change: whose fingerprint is seen in global warming?

Attributing observed climate change to causes is challenging. This letter communicates the physical arguments used in attribution, and the statistical methods applied to explore to what extent different possible causes can be used to explain the recent climate records. The methods use fingerprints of climate change that are identified on the basis of the physics governing our climate system, and through the use of climate model experiments. These fingerprints characterize the geographical and vertical pattern of the expected changes caused by external influences, for example, greenhouse gas increases and changes in solar radiation, taking also into account how these forcings and their effects vary over time. These time-space fingerprints can be used to discriminate between observed climate changes caused by different external factors. Attribution assessments necessarily take the natural variability of the climate system into account as well, evaluating whether an observed change can be explained in terms of this internal variability alone, and estimating the contribution of this source of variability to the observed change. Hence the assessment that a large part of the observed recent warming is anthropogenic is based on a rigorous quantitative analysis of these joint drivers and their effects, and proceeds through a much more comprehensive and layered analysis than a comparison at face value of model simulations with observations.