Application of geostatistical simulation models in the characterization of complex geological structures

Δ. Σιδέρη, Κώστας Π. Μόδης, Δημήτριος Ε. Ρόζος


Geostatistical simulation methods are able to generate numerical models or relations of the spatial distribution of a continuous geologic variable (grade, thickness, density, etc.) or a categorical variable (geological units and lithofacies or rock types). In this work, a review of traditional simulation techniques, as the Sequential Indicator Simulation (SIS), reveals a major pitfall that comes from theoretical difficulties in the development of a valid cross covariance model. On the contrary, a valid indicator cross covariance model is automatically defined in the framework of the Truncated Gaussian Simulation Method (TGS). This method is based on the concept that the categorical variables are obtained by truncating one standard multigaussian random variable at different thresholds. Plurigaussian Simulation Method (PGS) is an extension of the TGS Method but based on the simultaneous truncation of several multigaussian variables. An application of Plurigaussian method to simulate the lithofacies in the alluvial formations of the West Thessaly Basin is finally presented. This method was shown to be effective in reproducing the spatial characteristics of the different lithofacies and their distribution across the studied area.

Λέξεις κλειδιά

Διεθνές Συνέδριο της Ελληνικής Γεωλογικής Εταιρίας (13ο : 5-8 Σεπτεμβρίου 2013 : Χανιά, Ελλάδα); International Congress of the Geological Society of Greece (13th : 2013 September 5-8 : Chania, Greece); plurigaussian; lithofacies modelling; categorical variables; λιθολογικές φάσεις; κατηγορικές μεταβλητές

Πλήρες Κείμενο:


Εισερχόμενη Αναφορά

  • Δεν υπάρχουν προς το παρόν εισερχόμενες αναφορές.

©1999-2001 Βιβλιοθήκη "Θεόφραστος", Με την επιφύλαξη παντός νομίμου δικαιώματος.
Εφαρμογή / Διαχείριση Ιστοχώρου: Μηντζαρίδης Στέργιος
Η ανάπτυξη του site πραγματοποιήθηκε με την αποκλειστική χρήση Λογισμικού Ανοιχτού Κώδικα