The Little Ice Age: How Climate Made History 1300-1850 by Brian Fagan

are another barometer of changing temperatures, in the sense that they can be used to identify cycles of unusually wet or dry weather that brought poor harvest in their train. Economic
historians like W. G. Hoskins have tracked grain prices over many centuries and chronicled rises of 55 percent to as much as 88 percent above
normal at times of scarcity, when hoarders and merchants stockpiled
grain with an eye to a windfall profit or cereals were just in short supply.
In countries like Britain or France, where bread was a staple, such rises
could be catastrophic, especially for the poor. In the social disorders that
usually followed, farmers lived in fear of their crops being pillaged before
they were ripe, and mobs descended on markets to force bakers to sell
bread at what they considered to be fair prices. Monastery records and the
archives of large estates are a mine of information about harvests good
and bad, about prices and yields, but, like most early historical sources,
they lack the precision of a tree-ring sequence or an ice core. Annalists
would write of heavy rain storms that "in many places, as happens in a
flood, buildings, walls, and keeps were undermined," but such vivid descriptions are no substitute for reliable daily temperature readings.3

    The dates of wine harvests, derived from municipal and tithe records,
also vineyard archives, provide a general impression of cooler and warmer
summers, with the best results coming from linking such information
with readings from tree rings and other scientific sources. The climate
historian Christian Pfister focuses on two crucial months that stand out
in colder periods: cold Marches and cool and wet Julys. Such conditions
marked 1570 to 1600, the 1690s, and the 1810s, probably the coldest
decades of the Little Ice Age.4
    Climatic historians are ingenious scholars. For instance the Hudson's
Bay Company insisted that its captains and factors in the Canadian Arctic keep weather records on a daily basis even at its most remote stations.
Since the same employees often worked for the company for many years,
the records of ice conditions and of the first thaw and snowfall are remarkably accurate for the late eighteenth and early nineteenth centuries,
to the point that you can track annual fluctuations in first snowfall and
the beginning of the spring thaw to within a week, even a few days, over
long periods of years. Spanish scholars are using the records of rogation
ceremonies performed to pray for rain or an end to a deluge. Rogation
rituals were rigorously controlled by the church and unfolded at various
levels, culminating in years of crisis in formal processions and pilgrimages. Thus, they provide a crude barometer of climatic fluctuations.

    Historical records like these clearly display minor fluctuations between
decades, but how they relate to broader climate change is a matter for future research. In recent years, statistical methods are being used to test indices developed from historical sources against tree-ring and other scientific climatic data. From such tests we learn, for example, that
sixteenth-century central Europe was cooler at all seasons than the period
1901 to 1960, and that winters and springs were about 0.5° cooler, with
autumn rainfall about 5 percent higher. Almost uninterrupted cold winters settled over the area between 1586 and 1595, with temperatures
about 2°C cooler than the early twentieth-century mean. The same indices proclaim 1691-1700 and 1886-95 as the coldest decades in
Switzerland over the past five centuries.
    For all the richness and diversity of archival records, we have to rely in
large part on scientific sources for year-by-year climatic information on
the Little Ice Age. This record comes in part from ice cores, sunk deep
into the Greenland ice cap, into Antarctic ice sheets, including one at the
South Pole, and into mountain glaciers like that at Quelccaya in the
southern Andes of Peru.