http://en.wikipedia.org/wiki/Proxy_(clim...
http://serc.carleton.edu/microbelife/top...
http://www.planetseed.com/relatedarticle...
NOAA also has a good website on climate proxies but the science deniers have managed to get it shut down: http://governmentshutdown.noaa.gov/
Cores are also taken from lakebeds and ocean floors. The types of sediment, content of the sediment thickness of each layer etc all provide some data that can be useful. You should never get tunnel vision and allow and one dataset to weight to heavily against the total. Only when multiple datasets like these agree can you begin to say that the identified trends are much more than random fluctuations.
As a general rule, how *all* of these sampling methods work is that you look at a sample (for example, tree rings), you figure out what conditions would have produced that sample (for example, trees tend to grow faster in warm years), and you figure out how long ago a particular sample was (for example, you count tree rings).
It's best to compare many data sets, and many different measurement methods, because all of them have weaknesses, many of them different from one another. For example, tree rings can be skewed by high or low rainfall, ice cores can lose years when there is a lot of melting, and so on.
Tell lies and more lies, when found out scream and shout” Everyone else is a moron, the science is settled.”
For example, historical temperature patterns extracted from tree rings in Siberia’s Yamal Peninsula look increasingly dubious. They indicate warming, but were taken from a sample of only 12 trees. A larger sample (34 trees) in the same area shows no warming.
The scientists who compiled the record using the smaller sample are rightfully being accused of cherry-picking their data.
http://bijenkorf.wordpress.com/2009/10/0...
http://www.theresilientearth.com/?q=cont...
http://himalayanenvironment.org/newslett...
