Seismic Virtual Reality - Finite Difference Modelling of a Giant Oil Field in the Caspian

In this paper, we discuss the development of a model of the giant Azeri-Chirag-Gunashli (ACG) field offshore Azerbaijan for the purpose of simulating future seismic acquisition. Historically, seismic models have either been constructed manually or derive their properties from existing seismic images. In the first case, reflectivity is often too simplistic; whilst in the second case artefacts and bandwidth limitations are introduced as truth into the model. Here we present an alternative. We first use an automatic, probabilistic method to generate synthetic stratigraphic perturbations based on a set of facies templates and rock properties trends, and then morph the resultant layering to structure. These high-frequency perturbations are added to a low-frequency regional velocity model. Finally, mud volcanoes, anhydrite layers and pseudo-random attenuating shallow gas bodies are inserted. This combination of stochastic stratigraphy and semi-automatic definition of shallow hazard features has allowed the rapid development of a highly detailed 3D computational model. Preliminary images from seismic finite-difference simulations are realistic in terms of detail, bandwidth, resolution and signal to noise. This model will be used to assess the impact of different acquisition configurations and processing sequences on image fidelity, and can be further adapted to test 4D feasibility.