Estimation of environmental paleogradient values based on presence/absence data: a case study using benthic foraminifera for paleodepth estimation

A two-step procedure for estimating paleogradient values based on the presence/absence of variates (e.g, species, chemical elements, and traits) is presented. The first step requires information about the general (basic) distributions of variates along the gradient. Using species as variates, fundamental ecological niches corresponding to the gradient represent these general distributions. The second step involves a transfer equation, which is the arithmetic mean of location parimeters of variates weighted by their dispersion parameters. This procedure strengthens variates with narrow dispersions, yielding more information about the location at the gradient than variates with wide dispersion. The method proposed here does not require information about frequencies; solely the presence of variates is sufficient. Objective criteria are used to eliminate non-overlapping variates along the gradient (e.g., species whose, distributions do not overlap in living individuals but are mixed in samples of empty tests). This is important in estimating paleogradients because mixing of non-overlapping variates is a common effect due to post-mortem transport, bioturbation, or sediment reworking. The new method is calibrated for depth estimation using living symbiont-bearing benthic foraminifera and empty tests from the NW Pacific, and living smaller benthic, foraminifera from the shelf and continental slope of Senegal. This approach is then successfully applied for paleodepth estimation of a lower to middle Miocene transect in the Styrian Basin (Austria), where plankton/benthos ratios of foraminifera yield poor depth estimations. (C) 2004 Elsevier B.V. All rights reserved.