GEOSCAN, résultats de la recherche


TitreImproving seismic velocity estimation for 2-D poststack time migration of regional seismic data using kriging with an external drift
AuteurDuchesne, M J; Claprood, M; Gloaguen, E
SourceThe Leading Edge vol. 31, no. 10, 2012 p. 1156-1166,
Séries alt.Secteur des sciences de la Terre, Contribution externe 20110308
ÉditeurSociety of Exploration Geophysicists
Documentpublication en série
Mediapapier; en ligne; numérique
Sujetsdonnées sismiques; interpretations sismiques; vitesse des ondes sismiques; modèles; établissement de modèles; géostatistiques; géophysique
Illustrationsseismic profiles; plots; sections
ProgrammeBassin sédimentaire Sverdrup, GEM : La géocartographie de l'énergie et des minéraux
Résumé(disponible en anglais seulement)
Kriging with an external drift (KED) is used to estimate lateral velocity variations within geological layers from well data and seismic horizons. Alternatively velocity estimations are also obtained from a commercial processing package using linear interpolation. Comparisons between velocity estimations derived from linear interpolation and KED show that as expected the first method fails to accurately picture lateral velocity changes within each geological layer. To test the reliability of the estimations provided by both methods, velocities are then used to migrate vintage stacked seismic section. Both migrations have successfully collapsed hyperbolic reflections observed on the section. However the image migrated with the velocity estimated by linear interpolation presents numerous migration ¿smiles¿ that are diagnostic of using inadequately high seismic velocities. Those migration artefacts are not present on the seismic image obtained with the KED velocity estimations. The good migration results obtained using the KED estimations show that this method is reliable. Thus, it can be use to improve seismic imaging where information on lateral velocity changes is sparse or for seismic lines covering long distances where such changes are expected to be important even within a single geological layer. Additionally, including and adapting such a simple geostatistical method for estimating velocities in the re-processing workflow of vintage seismic data sets can provide additional insights for the efficient planning of new surveys and on decision making for prospects located in frontier basins where the risk is generally higher.