PALYNOSTRATIGRAPHY OF THE EOCENE FLUVIAL ESMERALDAS FORMATION, COLOMBIA

Most of the oil production in Colombia is derived from the Cenozoic continental stratigraphic record in a major foreland basin. The evolution of the northern Andes has fragmented the basin, creating many structural complexities and stratigraphic asymmetry related to syntectonic deposition. The Middle Magdalena Valley basin located between the Central and Eastern Cordilleras is one of the products of this fragmentation. The alluvial plain rock sequence of the Esmeraldas Formation is an important reservoir in the basin. Because of the complex and continental nature of the basin, the lateral continuity of the Esmeraldas is not clearly understood. Additionally, there are very few reliable correlation elements. The Los Corros fossil horizon, a mollusk horizon in the upper part, is the most traditional correlation component used to differentiate the Esmeraldas. This is the only paleontological unit formally associated with the Esmeraldas. This horizon, however, is not continuous over the entire basin and its age is still debatable, introducing uncertainty to existing stratigraphic models. Furthermore, the uncertainty in the age of the Esmeraldas makes temporal correlations of this Formation with units in adjacent basins unclear. Palynology appears to be the best paleontological tool for correlation in the basin. Data from an outcrop section and seven wells have been integrated with a palynostratigraphical zonation previously proposed for northern South America in order to obtain the age for the Esmeraldas Formation. We infer that the lower part of the Formation has a Middle Eocene age, mainly based on the First Appearance Datum (FAD) of Rhoipites guianensis and Cicatricosisporites dorogensis. The upper part (including the Los Corros) is Late Eocene, based on the Last Appearance Datum (LAD) of Racemonocolpites facilis and the presence of Luminidites colombianensis. The palynostratigraphy of the unit will serve as a reference for accurate subsurface correlations in the basin. It also will be useful when integrated with non-paleontological tools to better constrain the lateral extent of the Esmeraldas and its relationship with lithostratigraphic units in adjacent basins in Colombia. In addition, the age of the Esmeraldas will contribute to a better understanding of the overall evolution of the basin.