Thin reservoir prediction by prestack inversion combined with geostatistic simulation

In JD Oilfield of Bohai,China,delta fronts and distributary channel sands characterized by rapid lateral variations dominate reservoirs.Both extrusive(lava flow) and intrusive igneous rocks are distributed randomly in the target,and the resolution of seismic data is poor.Therefore it is difficult to predict reservoir with the thickness less than 15 m.Based on Greebert-castagna model,we propose a new approach for thin reservoir prediction by prestack inversion combined with geostatistic simulation.Through rock-physic modeling,a template for oil sands,oil-show sand,wet sands and non reservoir rocks in the area is established.A new attribute,which indicates variation of a sand reservoir and fluids in it,was found through elastic parameters combination.With this new attribute,we combined prestack inversion and geostatistcal simulation to make full use of both information to predict thin reservoir in the area and results are satisfied.