The 1st AAPG/EAGE PNG Geosciences Conference, PNG’s Oil and Gas Industry:
Maturing Through Exploration and Production

Datapages, Inc.Print this page

Integrated PSDM Transect Modelling and Imaging across PNG


Seismic imaging across Papua New Guinea is routinely challenging. Poor signal penetration, very complex raypaths and inadequate wavefield sampling due to only having a 2D recording geometry with limited offsets, frequently conspire to prevent the generation of adequately constrained depth models. Further, historical seismic datasets are often prospect/field targeted, making regional interpretations challenging due to significant variations in datum, acquisition geometry and phase/amplitude scaling between vintages. With such limitations, it becomes increasingly necessary to invoke a comparatively greater level of integration with other data types and adjacent datasets, in order to reduce the range of plausible model solutions. Regionally, PNG has a long petroleum exploration history which has heavily relied upon field sampling and aerial surface structure mapping as key inputs to the prospect evaluation process. Importantly, recent efforts have shown that, when these datatypes are optimally integrated, it is possible to derive meaningful, lithologically and tectonically diagnostic information across the Digital Elevation Model (DEM). This information may be further utilised to optimise the seismic imaging process. This paper examines how the incorporation of high density lidar elevation data and surface sampled geological information, such as strike/dip and lithology, into the 2D velocity modelling process, can help with the creation of improved regional, geologically consistent subsurface images. Additional reduction of velocity model uncertainty is achieved, particularly in the deeper basement section, by extending the final transects to include adjacent lines in the pre-stack binning and depth imaging (PDSM) process. Two example transects are reviewed which have utilised this approach: The first example is taken through the Western highlands where the dominating geophysical constraint is the surface karstified limestone. Here, the source signal is hampered by significant scattering and attenuation. The second example crosses the lower Eastern Fold Belt toward the Gulf of Papua. Here, the surface conditions are highly variable, ranging from anisotropic to highly attenuative, resulting in a more complex geophysical problem where wavefield distortion needs to be integrated into the velocity modelling. Although the surface geology and tectonics from each of the regions vary, the iterative method behind the modelling process remains very similar. In both cases, the derived, final merged images are more geologically plausible and show improved focussing throughout the entire section, in comparison with results generated from using the inputs in isolation. Due to the compounding benefits of this approach the integrated results are better than sum of the individual images. Ultimately, the technique can be applied to several cross-cutting regional transects which could be used to underpin the formulation of regional scale basin models and provide a macro framework for prospect scale exploration and development efforts.