GEOSCAN, résultats de la recherche

Menu GEOSCAN


TitreGeostatistical modeling of regionalized grain-size distributions using Min/Max Autocorrelation Factors
AuteurDesbarats, A J
SourcegeoENV III - geostatistics for environmental applications: proceedings of the Third European Conference on Geostatistics for Environmental Applications, Avignon, 2000 ; par Monestiez, P (éd.); Allard, D (éd.); Froidevaux, R (éd.); 2001 p. 441-452
Année2001
Séries alt.Commission géologique du Canada, Contributions aux publications extérieures 2000184
ÉditeurKluwer Academic Publishers |a Dordrecht, The Netherlands (Dordrecht, The Netherlands)
RéuniongeoENV III: Third European Conference on Geostatistics for Environmental Applications; Avignon; FR; Novembre 22-24, 2000
Documentlivre
Lang.anglais
Mediapapier
Sujetsanalyses granulométriques; repartition granulométrique; distribution des sédiments; milieu sédimentaire; faciès sédimentaires; roches sédimentaires; applications par ordinateur; logiciel; analyse statistique; établissement de modèles; géomathématique
Illustrationsformulae; tables; bar graphs; graphs
Résumé(disponible en anglais seulement)
Sediment grain-size distributions can be represented in a discrete form, as vectors of coregionalized size-class abundances. Since the number of classes may be large and abundances in adjacent classes may be highly cross-correlated, practical simulation of regionalized grain-size distributions requires an efficient method for de-correlating and reducing the number of variables. Using an example based on detailed grain-size measurements from the Las Cruces Trench Site, this study demonstrates features of Min/Max Autocorrelation Factors that make them a superior alternative to Principal Components for dimensional reduction and decorrelation in simulations of coregionalized variables.
GEOSCAN ID211774