UCAR Office of Education and Outreach
Although we don't have any direct means for measuring, for example, the mean annual temperatures at various places in North America two million years ago, we do have indirect methods for estimating climatic conditions of the distant past. Such indirect clues about past climates are know as paleoclimate proxies, or proxy records. Just as Sherlock Holmes might infer the height, weight, and other telltale features of a suspect from a series of footprints, paleoclimatologists infer the climatic conditions of the past from tree rings, ice cores, layers of ocean sediments, and similar proxy evidence.
This course has several pages devoted to most of the major types of proxies
commonly studied. As you read these pages, don't worry too much about specific
details of each of the proxies. Instead, look for overarching trends
and commonalities between the different types of records. Here are a few "study guide" pointers
to keep in mind as you survey the types of proxy records:
- What is actually being measured, and how is it related to
climate? Which aspect(s) of climate (precipitation, temperature, etc.) does
the proxy provide information about? What uncertainties are involved in inferring
knowledge of climate from the type of data collected?
- For example, we look
at the thicknesses of tree rings and presume that elements of the climate
in the tree's locale, such as temperature and moisture, influenced
the tree's rate of growth. Other factors besides climate, such as disease,
could alter a tree's growth rate and diminish the accuracy of this proxy
as a climate indicator.
- Each type of proxy has a "span" and a "resolution". Span influences how far into the past
we can "peer" using a particular proxy. Resolution describes
how "fine-grained", in terms of time spans, the climate information
is, and thus to what extent we can detect short-term
changes in climate.
- The "coverage" of some proxies overlap. Some proxies cover, at least partially,
the same time spans. We can compare the "stories" told by multiple proxies
to see how well they provide a coherent picture of climate. We can calibrate
one proxy against another. Overlap in time coverage can provide a more complete
picture if proxies cover different geographic areas or "measure" different
quantities (snowfall vs. sea surface temperature, etc.).
- Growth bands in coral reefs and ring patterns in trees are
two proxies that complement each other. Both provide records going back
years. Corals grow only in the tropics, and tell us about sea surface
temperatures. Trees that have annual growth rings live in mid- to high-latitude
regions, and tell us about air temperature and precipitation.
- Several types of proxies have "layers" or "bands" or "rings". Each such
layer represents a fixed amount of time, often a year of seasons, and tells
us about the time when it was created or laid down. Layers vary in thickness,
color, chemical composition, and other traits which indicate the climate
at the time of formation of each layer. Tree rings, ice cores, lake and ocean
sediments, and several other proxies display such layering.
- We can match up layers from multiple records, if the records overlap
in time at least a bit, to build up a chronology that spans a longer
time period than does any individual piece of data. By combining tree
ring data from a tree that sprouted in the 1850s and was recently
cut down with the ring pattern in a beam from a log cabin built in the
1870s, we might establish a proxy climate chronology that runs from
the 1700s to the present day. This sort of splicing together of records
is done with layers in ice cores, coral bands, and other proxies as well
as tree ring data.
The main types of proxies used to "measure" past climates are briefly
described below. Click on each proxy to go to a page with a more in-depth treatment.
- Tree Rings & Dendrochronology -
the annual growth rings of trees from mid-latitude climates, where a warm
growing season alternates with a winter of dormancy, tell us about temperature,
precipitation, and other factors that affect plant growth rates.
- Ice Cores - long,
cylindrical cores of ice drilled from glaciers and the ice caps of Greenland
and Antarctica have annual bands. The layers in the ice tell us about temperature
and precipitation. Gas bubbles trapped in the ice provide data about atmospheric
composition in the past.
from Lakes Beds and Seafloors - layers
of silt, sand, and other sediments are deposited on the bottoms of lakes and
oceans. Layer thickness can indicate amounts of precipitation or meltwater
runoff as rivers carry sediments downstream. Chemical and isotopic composition
of the shells of microorganisms embedded in sediments can tell us about the
temperatures when those shells were made.
reefs - corals form annual
growth bands. Analysis of ratios of different types of oxygen in minerals in
corals tells us about sea temperatures, although variations in salinity can
make these records more difficult to interpret.
- Pollen - the amounts
and types of pollens found in sedimentary deposits tell us about the types
of vegetation found in a certain area at a given time. The vegetation that
grows in a place indicates the climate of that place, since different plants
grow in differing conditions of temperature, moisture, etc.
- Fossils -
knowledge of the conditions in which various animals and plants thrive,
or at least survive, allows us to use fossils from sedimentary rocks as
Middens - colonies of
packrats generate clusters of plant and animal remains in the caves in
which they take refuge. The remains, called "middens", provide clues about
the climate of the immediate vicinity, especially in terms of the plant
species found there.
- Speleothems -
as a group, stalactites, stalagmites, and flowstones in caves are dubbed
"speleothems". Their rate of growth depends upon water percolating into
a cave through the ground, and thus indicates long-term rainfall rates
in the region near the cave.
Last modified: 11 October 2010
Created: 6 October 2010