Yet another use of ice record

Today's ScienceDaily [reported](http://www.sciencedaily.com/releases/2008/12/081231131813.htm) a new possibility of ice core sample analysis, this time using bacteria frozen in the ice.

It was found that in ice core sample collected from Mountain Geladaindong, the abundance of bacteria is in accordance with $$\delta ^{18}\text{O}$$, which can serve as an indicator of temperature. It is already known that higher $$\delta ^{18}\text{O}$$ level implies higher temperature, now it seems bacteria abundance also increases with $$\delta ^{18}\text{O}$$, therefore it is possible to use bacteria numbers as an indicator of historical temperature change.

However, $$\delta ^{18}\text{O}$$ can also serve as an indicator of precipitation, with higher level in low-precipitation years. It has also been noted that in the ice core higher bacteria level concurs with higher amount of dust. So here's my speculation: it may well be that in dry years the dust activity is more pronounced, leading to more dust and bacteria precipitate on the glacier. In that case, bacteria itself is not a useful indicator of either temperature nor precipitation.

It is also interesting to note that the lowest bacteria level was found in 1938 and the highest in 1997. According to [Desert Research Institute](http://www.wrcc.dri.edu/enso/ensodef.html), 1938 was a strong La Niña year, while 1997 was a strong El Niño year. So there is possibility that the effect of El Niño/La Niña may have so strong signals that the actual trend due to climate change is harder to observe. Moreover, if the bacteria abundance is proved more correlated to El Niño/La Niña, it suggests the increased possibility that bacteria level is linked with precipitation rather than temperature, as [earlier research](http://ecosystems.wcp.muohio.edu/studentresearch/climatechange03/elnino/Ice%20core%20records.pdf) argued that precipitation is more correlated to El Niño than $$\delta ^{18}\text{O}$$ does.

On the other hand, if this method is proved feasible, it would mean that scientists have get hold of another relatively cheap way of deriving historical climate data, and do not have to deal with expensive isotope analysis. Similar method may also be used in lake sediment or peat analysis. But to validate this method, "we need more data".