Abstract. Five seismic interpretation experiments were conducted on an area of
interest containing a fault relay in the Snøhvit field, Barents Sea,
Norway, to understand how the interpretation method impacts the analysis of
fault and horizon morphologies, fault lengths, and throw. The resulting
horizon and fault interpretations from the least and most successful
interpretation methods were further analysed to understand their impact on
geological modelling and hydrocarbon volume calculation. Generally, the
least dense manual interpretation method of horizons (32 inlines and 32 crosslines; 32 ILs × 32 XLs, 400 m) and faults (32 ILs, 400 m) resulted in inaccurate
fault and horizon interpretations and underdeveloped relay morphologies and
throw, which are inadequate for any detailed geological analysis. The
densest fault interpretations (4 ILs, 50 m) and 3D auto-tracked horizons (all
ILs and XLs spaced 12.5 m) provided the most detailed interpretations, most
developed relay and fault morphologies, and geologically realistic throw
distributions. Sparse interpretation grids generate significant issues in
the model itself, which make it geologically inaccurate and lead to
misunderstanding of the structural evolution of the relay. Despite
significant differences between the two models, the calculated in-place
petroleum reserves are broadly similar in the least and most dense
experiments. However, when considered at field scale, the differences in
volumes that are generated by the contrasting interpretation methodologies
clearly demonstrate the importance of applying accurate interpretation
strategies.