GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 63-3
Presentation Time: 2:00 PM

INTERPRETING TRACE FOSSILS PRODUCED BY PREDATORS: THE EFFECT OF BIOLOGICAL FILTERS ON ESTIMATES OF DRILLING PREDATION FREQUENCY


KOWALEWSKI, Michał, Florida Museum of Natural History, University of Florida, Gainesville, FL 32611, Chile and NAWROT, Rafal, Florida Museum of Natural History, University of Florida, 1659 Museum Rd, Gainesville, FL 32611

Various types of paleontological data carry information about the fossil record of biotic interactions. However, only few types of data are quantifiable and sufficiently widespread to be applicable in quantitative paleoecological analyses. Drill holes produced by predatory organisms are one of the most frequent targets of analyses that attempt to quantify the intensity of biotic interactions. Here, using drill holes as a model system, a simple modeling exercise serves to illustrate sources of uncertainty associated with interpreting frequency of drill holes.

Drilling frequency may provide insights into two parameters of interest: drilling predation mortality [DPM] of a given prey (what proportion of prey was successfully killed by drilling organisms?), and “overall predation mortality” [OPM] of a given prey (what proportion of prey was killed by any predators?). Two consecutive filters (biological and taphonomic) affect paleontological estimates of drilling frequency [DPM* and OPM*]. Here the focus is on modeling obvious biological filters that affect drilling frequency estimates. These include: (1) facultative predator filter, represented by drilling predators that occasionally kill without drilling (DPM* < DPM and OPM* < OPM); (2) durophagous predator filter represented by other predators, which killed some prey and also destroyed their skeletons (DPM* > DPM and OPM* > OPM); and (3) invisible predator filter represented by other predators that killed some prey without leaving any evidence (DPM* = DPM and OPM* < OPM).

A simple 3-parameter model that jointly considers the three filters suggests that, except for unrealistic parameterization, the DPM* and OPM* estimates underestimate the actual value of DPM and OPM. Because drilling frequency estimates are used nearly exclusively to study relative changes, the filters are inconsequential unless they vary notably in space, through time, or across taxa.