MULTIPLE-POINT GEOSTATISTICS

Jef Caers. ( 1999 )
in: 19th gOcad Meeting, ASGA

Abstract

The current practice of geostatistics for reservoir characterization is limited by the variogram which, as the fundamental geostatistical measure of geological continuit y, can capture only two-point statistics. Important geological information can be depicted, hence quantified in so-called training images which act as an interface between descriptive geology and geostatistics. These training sets can be used to extract multiple-point statistics (pattern extraction and recognition). Stochastic reservoir simulation then consists of anchoring the borrowed geo-structures to the actual subsurface hard and soft data (pattern reconstruction, stochastic inversion).

Download / Links

    BibTeX Reference

    @inproceedings{CaersRM1999,
     abstract = { The current practice of geostatistics for reservoir characterization is limited by the variogram which, as the fundamental geostatistical measure of geological continuit y, can capture only two-point statistics. Important geological information can be depicted, hence quantified in so-called training images which act as an interface between descriptive geology and geostatistics. These training sets can be used to extract multiple-point statistics (pattern extraction and recognition). Stochastic reservoir simulation then consists of anchoring the borrowed geo-structures to the actual subsurface hard and soft data (pattern reconstruction, stochastic inversion). },
     author = { Caers, Jef },
     booktitle = { 19th gOcad Meeting },
     month = { "june" },
     publisher = { ASGA },
     title = { MULTIPLE-POINT GEOSTATISTICS },
     year = { 1999 }
    }