Paleoclimatic Reconstructions and Qualitative Models
Paleoclimatic indicators, including the economic deposits of interest to this course, range from the very specific to the very general. Latentes, for example, seem to indicate very specific conditions of temperature and precipitation. In contrast, biogeographic patterns are very general indicators, which usually cannot reveal specific climatic parameters but which are excellent for understanding gradients, particularly in temperature. Only a few quantitative methods exist for empirically determining paleoclimates. They are stable isotopes of oxygen (e.g., Savin, 1977), angiosperm leaf morphology (e.g., Wolfe, 1979), and biogeographical patterns in very young rocks (e.g., CLIMAP, 1976). Although certain other indicators, such as latentes, are thought to have very specific climatic requirements, they are so sparsely distributed in the rock record as to be of limited use in paleoclimatic recontructions.
The biogeographic patterns that can be used to quantitatively determine ocean temperature are those on which transfer functions have been performed. The use of transfer functions is a numerical method that infers the temperature requirements of ancient biotic assemblages from the degree of taxonomic similarity to modern assemblages, whose temperature requirements are known (Sachs et al., 1977). The problem is that, as fewer modern forms are found in the successively older sediments, the less reliable is the paleotemperature determination. Therefore, although the technique may be useful for the late
Pleistocene (CLIMAP, 1976), it is virtually useless for older rocks. An informal version of this approach is to use individual taxa, for example, crocodilians, as paleoclimatic indicators (Colbert, 1964). However, this technique presupposes similar environmental