The first thing we have to acknowledge is that paleoclimatology yields century-term trends; it's not able to tell us the precise local temperatures of a given day 500 years ago, for example.
So what climate scientists do is establish "ranges of uncertainty"; their graphs and data show that, given the raw data, the reconstructed temperature was somewhere between Y1 and Y2 for a particular hundred-year period.
After comparing the reconstructions, going as far back as a.d. 1000, scientists are finding that every source of paleoclimatic data that they can come up with is showing that current temperatures are well above historical temperatures, even after taking the uncertainty into account.
That is, this century is the warmest global century since a.d. 1000, even if all of the paleoclimatic data were interpreted as warmly as possible.
They don't agree precisely in the shape of the graph -- glacial studies show a slightly different-shaped graph than borehole studies do, for example -- but the aggregate graphs do all agree on a warming trend over the last century.
However, in geological terms, a warming trend over a thousand-year period is not that big of an indicator. So what happens if we go back further?
There is a paper [1] which uses data from several other studies to reconstruct temperature deviations at 20 different global locations over a 2000-year period. It argues that the statistical methods used in other paleoclimate models are flawed, and uses a different statistical method which concludes that, in the absence of tree ring data, the Medieval Warm Period (approx. 950 a.d. to 1250 a.d.) was warmer than the current century.
I don't have the necessary background in statistics to be able to analyze the two different methods. However, even if we take that paper at face value, 2000-year data is still not geologically significant.
So, we turn to Antarctic ice cores. Specifically, data from the Vostok ice core sample shows a clear 150,000 year warming and cooling cycle, going back about 400,000 years ago (with increasing uncertainty).
At first glance, someone might think that this is damning evidence in the case for AGW. However, the recent thousand year period is a mere blip on a graph stretching back 400,000 years, so it's difficult to see the part that has climate scientists really concerned:
The amount of CO2 currently in the atmosphere, and reconstructed over the last thousand years, is dramatically greater than at any other point in the last 400,000 years. Specifically, CO2 concentrations have peaked at around 300 ppmv during the 150,000 year cycle, but are currently at around 383 ppmv.
To put this in perspective, the difference between the amount of CO2 in the atmosphere now, versus the amount in the peaks of the last 400,000 years, is close to the same as the difference between the peaks and the troughs.
So while the Earth clearly does have its own warming and cooling cycle, we have induced a dramatic change in the cycle, which should lead to some interesting results.
The first thing we have to acknowledge is that paleoclimatology yields century-term trends; it's not able to tell us the precise local temperatures of a given day 500 years ago, for example.
So what climate scientists do is establish "ranges of uncertainty"; their graphs and data show that, given the raw data, the reconstructed temperature was somewhere between Y1 and Y2 for a particular hundred-year period.
After comparing the reconstructions, going as far back as a.d. 1000, scientists are finding that every source of paleoclimatic data that they can come up with is showing that current temperatures are well above historical temperatures, even after taking the uncertainty into account.
That is, this century is the warmest global century since a.d. 1000, even if all of the paleoclimatic data were interpreted as warmly as possible.
They don't agree precisely in the shape of the graph -- glacial studies show a slightly different-shaped graph than borehole studies do, for example -- but the aggregate graphs do all agree on a warming trend over the last century.
However, in geological terms, a warming trend over a thousand-year period is not that big of an indicator. So what happens if we go back further?
There is a paper [1] which uses data from several other studies to reconstruct temperature deviations at 20 different global locations over a 2000-year period. It argues that the statistical methods used in other paleoclimate models are flawed, and uses a different statistical method which concludes that, in the absence of tree ring data, the Medieval Warm Period (approx. 950 a.d. to 1250 a.d.) was warmer than the current century.
I don't have the necessary background in statistics to be able to analyze the two different methods. However, even if we take that paper at face value, 2000-year data is still not geologically significant.
So, we turn to Antarctic ice cores. Specifically, data from the Vostok ice core sample shows a clear 150,000 year warming and cooling cycle, going back about 400,000 years ago (with increasing uncertainty).
At first glance, someone might think that this is damning evidence in the case for AGW. However, the recent thousand year period is a mere blip on a graph stretching back 400,000 years, so it's difficult to see the part that has climate scientists really concerned:
The amount of CO2 currently in the atmosphere, and reconstructed over the last thousand years, is dramatically greater than at any other point in the last 400,000 years. Specifically, CO2 concentrations have peaked at around 300 ppmv during the 150,000 year cycle, but are currently at around 383 ppmv.
To put this in perspective, the difference between the amount of CO2 in the atmosphere now, versus the amount in the peaks of the last 400,000 years, is close to the same as the difference between the peaks and the troughs.
So while the Earth clearly does have its own warming and cooling cycle, we have induced a dramatic change in the cycle, which should lead to some interesting results.
(Further reading: "Surface Temperature Reconstructions for the Last 2,000 Years", free download at http://www.nap.edu/catalog.php?record_id=11676)
[1]: http://www.ncasi.org/publications/Detail.aspx?id=3025