2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 14
Presentation Time: 11:35 AM

COMPUTER-ASSISTED SEQUENCING OF LARGE NUMBERS OF EVENTS FROM THE GEOLOGIC RECORD – ESCAPING THE CONSTRAINTS ON RESOLVING POWER IMPOSED BY TRADITIONAL BIOZONES


SADLER, Peter M., Department of Earth Sciences, Univ of California, Riverside, CA 92521 and SABADO, Jennifer A., Dept. of Earth Sciences, Univ of California, Geology Building 1448, Riverside, CA 92521, peter.sadler@ucr.edu

The CHRONOS database hub is designed to retrieve information concerning the relative age of large numbers of geologic events, including ash falls, geomagnetic pole reversals, excursions in ocean composition, and the origination and extinction of species. The information may take many forms, including numerical age estimates (e.g. from EARTHTIME), superpositional relationships in measured stratigraphic sections (e.g. PALEOSTRAT, NEPTUNE), and taxon coexistences (e.g. PALEOBIOLOGY DATABASE). Several software tools (e.g. BIOGRAPH, CONJUNCT, CONOP, GRAPHCOR, PAST, PC-SLOT, RASC) already use such information to estimate the sequence of events according to geologic optimization criteria. Together, the databases and tools can build an electronic stratigraphic record, with a grand composite time line of geologic events, complete with confidence intervals.

The coupling of community databases with software tools enables a third generation approach to geologic time correlation. In the first generation, individuals built sets of biozones manually from a few stratigraphic sections, using a tiny fraction of available origination and extinction events. The approximately 80 million year interval of the Ordovician and Silurian, for example, has been divided into 50-60 zones and sub-zones for graptolites and 30-40 for conodonts. In the second generation, individuals compiled all available events for one or two clades from all published stratigraphic sections and sequenced them with computer assistance. For Ordovician through Silurian time, we used CONOP9 to sequence the first and last occurrences of 1429 graptolite taxa from 262 sections and 1027 conodont taxa from 208 sections (more than 21,000 locally observed range-end and other events). After allowing for overlapping confidence intervals on the placement of events, more than an order of magnitude improvement remains relative to the resolving power of traditional zones and subzones. The third generation is now at hand, in which the process can involve the whole community, via shared databases, making feasible the compilation and quality control of much larger datasets and the sequencing of multi-clade geologic histories. The second generation broke the constraints of manual computation; the third generation breaks the constraint of individual expertise.