proxy climate information available (Table 3-1). In a classic early study, Ladurie (1972) used farming and phenological¹ records to document times of feast and famine in western Europe during the Little Ice Age (roughly 1500–1850). Logbooks and diaries, such as the diary kept by Benjamin Franklin when he was American ambassador in Paris during the 1780s, provide another, complementary source of data. Franklin reported a “constant dry fog on which the rays of the sun seemed to have little effect” along with severe late frosts, which we now attribute to the Laki volcanic fissure eruption in Iceland (Grattan and Brayshay 1995).

Many historical documents, rather than recording weather per se, provide indirect evidence of past climatic conditions. Historical paintings of alpine landscapes, for example, allow us to pinpoint the former extent of glaciers at precise moments in time, thus contributing to the temperature reconstructions derived from glacier length records discussed in Chapter 7. Similar, but potentially more continuous, time series of sea ice cover have been derived from Antarctic whaling records and from observations of drift ice around the coast of Iceland (e.g., Ogilvie 1992, de la Mare 1997). In the tropics and in dryland regions, periods of drought and flood are most frequently reported; Endfield et al. (2004), for instance, used archival sources to reconstruct rainfall fluctuations in Spanish colonial Mexico. To quantify long series of documentary data such as these in climatic terms, they, like other proxies, need to be calibrated against instrumental measurements. Brázdil et al. (2005) provide a comprehensive review of the methodological framework within which historical archives and documents are currently utilized.

All historical sources need to be evaluated critically, even for relatively recent times. For example, frost fairs were routinely held on the iced-over surface of the River