Despite the widespread use of methods to reconstruct past environmental conditions from microfossil assemblages, it has been difficult to assess how robust the reconstructions are. This has potentially allowed some weak or erroneous reconstructions to be published. A paper published in Quaternary Science Reviews by Bjerknes scientists Richard Telford and John Birks develops a method that should solve this problem by testing the statistical significance of palaeoenvironmental reconstructions.
 |
| Fig 1: Light-microscope picture of the diatom Cymbella lanceolata, a species found mainly in nutrient-poor lakes. Credit: RBGE |
Palaeoecologists use a wide range of microfossil assemblages to reconstruct past ecological conditions. Reconstructions of lake-pH from diatoms were central to the acid rain debate in the 1990s, showing that lakes had acidified recently, at the same time as atmospheric pollution from industrial had increased. Reconstruction of nutrient concentrations from diatoms and other microfossils has allowed us to determine if eutrophic lakes and fjords are naturally nutrient rich, or have become eutrophic recently in response to human activity. Reconstructions of past climate conditions, using, for example, pollen preserved in lake sediments to infer air temperatures, or foraminifera preserved in ocean sediments to infer sea surface temperatures, have been important tools to help understand the variability of climate before it was perturbed by anthropogenic emissions of carbon dioxide.
The new method works by testing if the reconstruction of the environmental variable explains more of the variance in the fossil data than most reconstructions of random environmental variables. This method can also determine which environmental variable can be best reconstructed from the fossil data, which will help to make palaeoecology a more quantitative, rigorous and robust science.