date: Thu, 9 Feb 2006 16:32:22 +0000 (GMT) from: Martin Juckes subject: Draft conclusions for report to Netherlands Environment Assessment to: mitrie -- Anders Moberg , Eduardo Zorita , , Jan Esper , Keith Briffa , Martin Juckes , Myles Allen , Nanne Weber , Hello, I need to send in a draft report to RIVM soon. The summary should lay out what we believe to be the state of knowledge on temperatures in the last millenium. I would be grateful for feedback on the text below. regards, Martin Summary IPCC (2001) concluded that ``The 1990s are likely to have been the warmest decade of the millennium in the Northern Hemisphere, and 1998 is likely to have been the warmest year," where ``likely'' implies a greater than 66\% probability (this conclusion will be referred to below as ``C1''). The Northern Hemisphere temperatures are believed to have shown a gradual cooling trend from the start of the millenium until the mid 19th century, and a warming trend since then. Substantial interannual, decadal and centennial scale variability was superimposed on these trends. The warming trend contains a signifcant natural component, but an anthropogenic contribution was clearly detectable towards the end of the 20th century. This conclusion was based on a wide range of results, including that of Mann et al., (1999). Since publication of the IPCC (2001) report there has been much criticism of the techniques used to estimate temperatures, particularly those used by Mann et al. The criticism of the latter work has drwan attention to incomplete documentation of the wide range of data sources used and to incomplete description of some aspects of the analysis algorithm. The debate has attracted much public interest and generated considerable confusion. (C1) is sometimes paraphrased as ``there was no hemispheric wide Medieval Warm Period'', but this terminology leads to confusion: there is no agreed definition of what would constitute a `` Medieval Warm Period''. A second conclusion of the IPCC report, which is related to but distinct from (C1), is that current temperature trends have a signifcant anthropogenic component (referred to as ``C2'' below). Conclusion (C2) is based mainly on GCM simulations and is not directly addressed in this study. Conclusion (C1) is based mainly on the interpretation of proxy climate records: this is the specific issue addressed here. Reconstructions of past climates are also used to evaluate GCM simulations of those climates and hence to evaluate the GCMs: this provides some indirect input into conclusion (C2). The following concpetual model can help us to understand how studies of the past millenium can contribute to discussion of future climate change: Temperature anomaly- = [ ( climate sensitivity-) times ( sum of forcings-) ] plus ( natural variability-) This is a drastic simplification: the different ``forcings'' (solar variability, volcanic and other natural changes to atmospheric composition, anthropogenic changes to atmospheric composition) can not be wholly characterised by a single number: their influcence on the climate system is extremely complex and the response of the climate is neither instantaneous nor uniform. Nevertheless, scientists have found this simple conceptual model to be a useful basis for discussion. By testing the models against observed climate variability it can be dtermiend whether they have a climate sensitivity which is realistic. The problem is that the period of reliable, global measurements is too short to carry out this exercise comprehensively. In the last 5 years a number of studies using different techniques and different, though overlapping, data collections have re-inforced (C1), though they disagree, both with Mann et al. and among themselves, on other issues. In particular, there is a relatively wide range of estimates as to the magnitude of the cold anomaly in the 18th century (during the ``Little Ice Age''). It is clear that regional temperature anomalies can be much larger than those on the hemispheric scale. IPCC (2001) did not suggest that current temperattures are above the extremes experienced by any region in the past thousand years. Recent modelling work has led to greater understanding of climate variability on different scales. A lot of discussion in the popular and electronic media, and also, to a limited extent, in the peer reviewed literature, neglects this crucial distinction between what is happening on the global and regional scales. Data centres have improved the transparency with which data is available and the quality of the information accompanying the data, recording its provenance has also improved. The use of a wide range of different data sources and different analysis techniques makes evaluation of the differences among published results difficult. Within this project we have subjected data collections from a variety of authors to several analysis techniques. It is found that the range of different results is still spanned by the results when a single analysis technique is used. This suggests that a priority for further work to reduce the uncertainty will be to improve understanding of the data.