Automated reconstruction of the tectonic and thermal history of the Northern Viking Graben using two-dimensional tectonostratigraphic forward models

Lars H. Rüpke¹, Stefan M. Schmalholz², Dani Schmid¹, and Yuri Y. Podladchikov^1
1Physics of Geological Processes, University of Oslo, PO Box 1048 Blindern, 0316 Oslo, Norway
e-mail: [email protected]
²Geological Institute, Leonhardstrasse 19, ETH Zurich, 8092 Zurich, Switzerland

Numerical basin models are a quantitative tool for understanding the formation and evolution of sedimentary basins. They have proven to be particularly useful in understanding a basin's thermal evolution and thereby hydrocarbon prospect. Two fundamentally different approaches exist in basin modeling: inverse (a.k.a. backstripping) and forward modeling. Both approaches have their advantages and restrictions but forward modeling has the virtue of yielding self-consistent structural and thermal solutions. Regardless of which modeling approach one follows, basin modeling should ideally integrate seamlessly into the work-flow from data acquisition to geometric model building. This integration is sometimes difficult with basin models being complex to use and requiring 'non-standard' input data that is not routinely acquired.

We try to overcome some of these restrictions by partially automatizing the basin reconstruction process. This is achieved through the use of a tectonostratigraphic forward model which is coupled to a constrained optimization algorithm which automatically refines the model parameters. The two-dimensional forward model is based on pure-shear kinematics and solves simultaneously for lithospheric scale (e.g. thinning, flexure, temperature) and sedimentary basin scale processes (e.g. sedimentation, compaction, maturation). Automated reconstruction is performed using an inversion method. The inversion is treated as a constrained optimization problem and consists in the iterative search for the optimal set of crustal (d) as well as lithospheric (b) stretching factors, and palaeo-water depth values, which yield the best fit between the observed and modeled stratigraphies. The seismic stratigraphy is therefore the main parameter to fit in the modeling process. Using this modeling approach, a first order basin reconstruction can be done by only using the seismic stratigraphy and information on the timing and duration of rifting events. The model can, of course, be improved by including additional data into the reconstruction process such as chrono-stratigraphic information on deposited rock type, temperature and vitrinite reflectance data.

In a test of our modeling approach we have reconstructed the thermal and structural evolution of the Northern Viking Graben. This reconstruction yields a present day stratigraphy, temperature field, vitrinite reflectance data and maturity prediction that is quality checked against regional observables and well data. This demonstrates that our model captures the first order physical processes involved in basin formation and evolution and that basin reconstruction can be largely automated and thereby more easily integrated into the normal work-flow of target area evaluation.

 

AAPG Search and Discover Article #90066©2007 AAPG Hedberg Conference, The Hague, The Netherlands