To the Editor

Parmesan and colleagues1 claim that the Intergovernmental Panel on Climate Change (IPCC) “advocates an ever-more-detailed approach to attribution” in a guidance paper for its fifth assessment report, and that applying this approach to biological systems “effectively yields to the contrarians' inexhaustible demands for more 'proof' [of human-induced climate change], rather than advancing the most pressing and practical scientific questions.” Although we welcome the scientific debate on this issue, we provide some background to this topic from the IPCC Working Group I to address these assertions.

The question of whether an observed change is caused by human activities is among the most often asked by the public, and is of immediate relevance to policymakers: planning for the future requires consideration of the climate forcing due to human-induced factors and the impacts associated with it. Therefore, scientists are called on to investigate this issue with all the tools available.

Detection and attribution — enabling the quantitative distinction between anthropogenic climate change and natural climate variability — has become a robust and well-tested methodology in climate science with a growing body of relevant peer-reviewed literature2. Some major statements in the Working Group I contributions to the third and the fourth assessment reports of the IPCC were based on it, and it will also be emphasized in the IPCC's fifth assessment report in both Working Groups I and II, with Working Group I devoting an entire chapter to detection and attribution at global to regional scales.

To support the scientific community engaged in the assessment process, IPCC Working Groups I and II jointly held an IPCC expert meeting on 'Detection and Attribution Related to Anthropogenic Climate Change' in September 2009. The meeting goal was to develop consistency and coherence of terminology used in detection and attribution studies, in particular where they extend to impact-relevant climate change, for example, detection and attribution of extreme events or changes in the carbon cycle and in ecosystems.

The product of this meeting was a 'Good Practice Guidance Paper'3, summarizing the discussions and clarifying methods, definitions and terminology across the IPCC Working Groups. It is a carefully formulated document, jointly authored by scientists from the physical sciences and from the impacts and ecosystem research communities, and in no way makes recommendations on research needs. It is fully in line with the mandate of the IPCC, which is to comprehensively assess the available science while not performing or promoting specific science.