In this approach, paleoecologists are interested in the causal underlying processes or 'forcing functions' for the observed stratigraphical patterns. For example, what factors caused the observed changes in pollen stratigraphy at a site over the last 11 000 years? Are the biotic changes responses to changes in climate, soils, biotic interactions, pathogens, disturbance regimes, land-use, etc..? It is essential to interpret paleoecological data in terms of underlying causal factors if the paleoecological record is to be used as a long-term record of ecological dynamics that can help understand present-day systems.
In causal interpretations, there may be two or more competing hypotheses to explain the observed patterns.
At least three independent proxies are needed to test two hypotheses. A major development in Quaternary paleoecology has been multiproxy studies where several proxies are studied on the same sediment core. To test competing hypotheses, one or more proxy is used to reconstruct the past environment. These reconstructions and other independent paleoenvironmental variables are then used as predictor variables to test hypotheses about the causes of change in the other proxies when they are considered as response variables. A range of statistical regression-modeling techniques can be used to test different hypotheses about the causes of the observed changes in the response variable in relation to the predictor variables. Statistical significance can be assessed by permutation tests that take into account the numerical properties of time-ordered paleoecological data.
Was this article helpful?