Principal indicators of past climates - Climate Change: Biological and Human Aspects - page 60

that there are others. Planktonic species abundance in oceanic cores (analogous to
pollen analysis in peat bogs), lake-level reconstructions, lake sediments and glacial
moraines are among a number of other palaeo-indicators of climate. Together all
indicators provide a valuable insight into past climates, over a variety of scales, both
spatially and temporally.
2.5 Interpretingindicators
Because proxy indicators are not actual instrumental measurements of climate but a
reflection of climate it is important to recognise their limitations. One might consider
the bringing together of several proxy data sets to form a record of a region's climate
as superimposing photocopies of different parts of a picture, with some pictures
having parts that are different to others and others the same. Such superimpositions
are perfectly readable but they are still blurred. Yet even statistical analyses of palaeo-
records, even if they reflect when times were warmer or cooler, may not necessarily
accurately reflect by exactly how much warmer or cooler past times actually were.
Palaeo-indicators such as tree rings are, as previously noted, themselves subject to a
number of non-climatic factors such as whether the tree grew on the north or southern
side of a hill, soil suitability and so forth. The data from such proxies are therefore
said to be noisy.
In 2004 a European team, led by Hans von Storch, attempted to see whether past
climate reconstructions from noisy proxies reflected climate by working backwards
from a climate model that had already been shown to reasonably accurately reflect
the northern hemisphere's climate for the past 1000 years. They used this model to
generate data as if from climate proxies and then added noise. They then analysed
the resulting proxy data to reconstruct the northern hemisphere's climate just as had
already been done with real proxy data. They found that their simulated noisy proxy
data did not accurately reflect the magnitude of the (simulated) past climate change.
This suggests that some (actual) palaeoclimate reconstructions (including those used
by the UN Intergovernmental Panel on Climate Change, IPCC) may only reflect half
the actual change that has historically taken place since the Industrial Revolution (von
Storch et al., 2004). If this were so it means that the IPCC's forecasts up to 2007 may
not have fully reflected the degree of climatic change we may actually experience
(see Chapter 5 and the subsection on uncertainties and the IPCC's conclusions).
2.6 Conclusions
Each technique for ascertaining an aspect of past climates has its limitations in terms
of: climatic dimension being indicated (or reflected), be it temperature or aridity;
resolution in time and space; and finally the sensitivity of the proxy. These lim-
itations have to be borne in mind when attempting to reconstruct past climates
and any interpretation made with reference to other climatic proxies. However,