.

Seismic data acquired from “Finima” Field of the Niger Delta was subjected to chimney cube analytical technique via the artificial neural networks (ANN) with the aim of identifying direct hydrocarbon indicators (DHIs) such as bright spot, dim spot, flat spot and possible geologic structures that could indicate the presence of gas within the subsurface. A number of inlines, cross-lines and z-slice were employed while few that gave high reflectivity were selected. Chimney YES and chimney NO algorithm were generated via the artificial neural networks template within the Opendtect environment. The result showed that the density of chimneys was concentrated around bright spot at time 3000 msec on inline 5800 and dim spot at time 1600 on inline 6000. At time 2000 to 3800 msec on inline 5800 and at time 1800 to 4400 msec on inline 6000, prominent faults were detected. The amplitude of the chimney cube was high on bright spots, dim spots, flat spots and faulted zones. The concentration of chimney cube density around the bright spots revealed that the zone is gas rich while those around the dim spots were oil rich. Around the flat spots were evidences of fluid contact. However, the density of the chimney on the bright spot diminishes upwards until it fizzles out, indicating that other zones with different direct hydrocarbon indicator such as dim spot are present.

This study concluded that artificially generated chimney cube, when applied on seismic data, provides a clearer result than when viewed under conventional seismic data without an applied chimney.

Keywords:  Seismic, Finima, Chimney, Neural, Networks, Training, Delta