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

Ice-core research bristles with highly technical
difficulties, many of them resulting from the complex processes by which
annual snowfall layers are buried deeper and deeper in a glacier until they
are finally compressed into ice. Scientists have had to learn the different
textures characteristic of summer and winter ice, so they can assemble a
long record of precipitation that goes back deep into the past. Snowfall
changes are especially important because they provide vital evidence on
the rate of warming and cooling during sudden climatic changes.
    Two cores from the Greenland ice cap, known as GISP-1 and 2, are of
particular interest for the Little Ice Age. GISP-2 has an accuracy in calendar years of ±1 percent, which makes it exceptionally useful for dating
temperature changes, themselves identified by changes in the isotopic signal of deuterium (D) from year to year, even season to season. Lower isotopic excursions signal low temperatures, such as those in Greenland during the fourteenth century, where winters were the coldest they have been
over the past seven hundred years. Ice-core climatic reconstructions offer
great promise for studying the short-term cycles of warmer and colder
conditions that affected the medieval Norse settlements in Greenland.
    Until the 1960s, tree-ring research was largely confined to the Southwestern United States, where astronomer Andrew Douglass achieved sci entific immortality by dating ancient Indian pueblos from the annual
growth rings in desiccated wooden lintel beams. Since then, thousands of
tree-ring sequences have come from the Southwest, to the point that experts can trace the year-by-year progression of serious droughts across the
region 1,000 years ago. Originally, tree-ring dating was applied only in
areas with markedly seasonal rainfall, but the science is now so refined
that we have highly accurate sequences from German and Irish oaks going back at least 8,000 years.

    Tree-ring temperature reconstructions now span the entire Northern
Hemisphere and come from over 380 locations. We have the first interannual and interdecadal temperature variability curves as far back as A.D.
1400 or earlier, with very reliable data for the years after 1600.5 Such
temperature estimates, acquired by statistical regression analyses from
modern instrument records or by proxies from historical records and
other sources, are vital to establishing just how warm the late twentieth
century has been in comparison with earlier times.
    Major volcanic eruptions, like that which destroyed Roman Herculaneum and Pompeii in A.D. 79, are spectacular, often catastrophic events.
The greatest of them can be detected in tree-ring sequences and through
fine dust in ice cores. Eruptions have important climatic consequences
because of the fine dust they throw out, which can linger in the atmosphere for years on end. Hypotheses linking eruptions and weather have
been around a long time. Benjamin Franklin theorized that volcanic dust
could lower temperatures on earth. In 1913, a U.S. Weather Bureau scientist named William Humphreys used data from the spectacular
Krakatau eruption in Southeast Asia in 1883 to document the correlation
between historic volcanic eruptions and worldwide temperature changes.
Volcanic dust is some thirty times more effective in shielding the earth
from solar radiation than it is in preventing the earth's heat from escaping. During the three years it may take for the dust from a large eruption
to settle out, the average temperature of much of the globe may drop as
much as a degree, perhaps even more. The effects tend to be most marked
during the summer following a major volcanic event.
    The provisional temperature curves for the Little Ice Age display some
conspicuous downward spikes, when a single year was unusually cold. Almost invariably, these are associated with major eruptions, such as that of Mount Tambora in southeast Asia in 1815, the most