Characterisation of Hydrocarbon Reservoirs in Bonny Area of Niger- Delta, Using Probabilistic Approach

This work proposed a joint estimation of petro-physical properties which combines statistical rock physics and Bayesian seismic formulation. This work is to present a strategy for estimating the probability distributions of petro-physical parameters and litho-fluid classes from seismic. The estimation of the hydrocarbon reservoir properties and the associated uncertainty is performed in three steps: firstly, a rock physics model is established using well log data and seismic data to predict elastic attributes (velocities or impedances) from petro-physical properties, after which an elastic property from partially stacked seismic angle gather is estimated. Finally, the conditional probability of petro-physical variables and litho- fluid classes is calculated. The application of this reservoir study included two well data (well A and well B) and partially stacked seismic volume which provided the probability density of petro-physical properties and litho-fluid classes. This clearly demonstrates the applicability of the proposed statistical method. Results obtained showed that well A had effective porosity between 40% - 55%, clay content 34cm3 - 48cm3 and water saturation between 10cm3 - 20cm3 and well B had effective porosity between 20% - 25%, clay content between 54cm3 - 60cm3 and water saturation between 25cm3 - 40cm3. Due to this result, well A will have greater production that would last for years at a true vertical depth (TVD) of between 5720ft - 5855ft. It is therefore concluded that probabilistic method of estimating hydrocarbon reservoir works better in the near surface layer of the overburden, where the signal to noise ratio is high, rather than the lower layer where the signal to noise ratio is low.