Abstract
Abstract
One of the proven technologies for reservoir monitoring is 4D (time-lapse) seismic. It allows tracking the fluid front for water & gas and detecting pressure & density changes. Although 4D seismic monitoring is challenging in the Middle Eastern carbonate reservoirs, dealing with non-repeated acquisition geometries pose additional challenges.
This paper will present how the 4D seismic processing methods and techniques were applied to overcome the following challenges from an offshore Jurassic carbonate oil and gas field: Seismic acquisition geometry differences between the two surveys leading to strong 4D noise related to repeatability issuesMultiple contamination especially in shallow water environment (water depth ranges from 15 to 25m)
Cascaded application of denoise and demultiple procedures were performed. Different types of migration were conducted too. In order to overcome specifically acquisition geometry differences between legacy and monitor surveys, Least-Squares Migration algorithm was applied to 4D datasets.
4D Least-Squares migration showed better 4D inversion results compared with the ones from Kirchhoff migration. As 4D inversion signal has the same order of magnitude as 4D noise, additional information such as geology and well production data were required to discriminate 4D signal from noise. Nevertheless, some areas still suffer from residual multiples that generate strong and low frequency noise in 4D inversion results. However, many 4D seismic anomalies were validated as 4D signal mainly corresponding to 4D P-impedance increase which were nicely matched with water cut observed at wells and with water saturation from reservoir simulation. As a general observation, the water rise observed in 4D seismic inversion results was higher (shallower) than dynamic model water rise. In addition, 4D water rise showed heterogeneous water paths compared to the uniform one in the dynamic model. 4D interpretation results were integrated in the reservoir model.