CARBONATE SHRUBS IN TRAVERTINES OF TIVOLI, ITALY: CAN MORPHOLOGY BE USED TO DISTINGUISH BETWEEN BIOTIC AND ABIOTIC PRECIPITATES?

Travertine successions from the Pleistocene of Tivoli, Italy, are commonly composed of mm- to cm-thick layers of ‘shrub-like' calcite interbedded with wavy, stromatolitic laminae. Each ‘shrub' is composed of an intricately branched aggregate of crystals that radiate upwards from a basal surface. Shrub layers are traceable across tens of meters, and are encased in calcite cement. Although the significance of dendritic morphologies is poorly understood, it has been suggested that they reflect the growth habit of bacterial colonies (Chafetz and Folk 1984). Guidry and Chafetz (2003) noted the morphological similarity between siliceous shrubs and bacterial colonies grown by Ben-Jacob (1997) as evidence of bacterial influence. By contrast, Pentecost (1990) concluded, from the examination of both natural and synthetic travertines, that shrubs formed abiotically by rapid CO₂ degassing and preferential crystal growth on substrate protuberances. Whereas microorganisms are undoubtedly abundant in modern springs, these studies suggest that their role in the formation of shrubs remains unclear.

In this study, shrubs similar in morphology to those observed in natural travertines were precipitated abiotically from a solution of oxalic acid dihydrate and acetone. Photomicrographs of the precipitates and bacterial colonies from Ben-Jacob (1997) were compared using a model that parameterizes morphology via link-length ratio (R₁), bifurcation angle (β), and branching ratio (R_B) (Cheetham et al. 1980). Whereas R₁ and β were unsuccessful in separating biotic from abiotic shrubs, R_B proved consistent in making this distinction. Data were used to establish fields of biogenicity and abiogenecity within the range of shrub morphospace, and the model was then applied to travertine shrubs from Italy. Natural travertine shrubs plotted entirely within the field of abiotic precipitates. These preliminary data suggest that travertine shrubs from Italy are abiogenic, and that morphometric classification may be an effective method of approaching questions of biogenicity, especially in Martian deposits where the only data available may be morphological in nature. The model will be refined by applying kernel density and cluster analyses to establish a confidence interval for biogenicity.