OBJECTIVES: to provide additional tools for the characterization and estimation of resources when there are several variables to consider simultaneously. It allows us to highlight spatial relationships between variables, to improve the estimation of a variable using secondary variables, collected at the same points [isotopic case] or not [heterotopic case], to improve consistency between several variables estimation, to simulate several variables together.


Several variables are investigated at the same time, in the case of polymetallic deposits, sedimentary geometry multi-surface environments or geochemical prospecting.

  • Basic statistics: reminders on linear regression;
  • Multivariate structural tools: cross covariance and cross variogram (these tools allow to quantify the spatial relationships between variables);
  • Intrinsic correlation model, linear model of coregionalization;
  • Cokriging: estimation method proposed by geostatistics for multivariate applications (general equations for isotopic or heterotopic cases) and cokriging simplifications for specific models or specific types of sampling (collocated cokriging, kriging with external drift...);
  • Factorial kriging analysis: to filter or map one or more of the spatial components of the model (used for mapping anomalies on small or large scale, as for geophysical and geochemical survey);
  • Multivariate statistical tools such as principal components analysis;
  • Elements for multivariate simulation (a simulation module is proposed later in the year).

Prerequisite: Proficiency in basic geostatistics is required see Linear Geostatistics training

Duration: 10 days
Dates: 06/11/2017 - 17/11/2017
Cost: 5 000 €
Including lunch
Language: English
an application for registration
with a CV

Training location : Mines ParisTech, 35 rue Saint Honoré, 77300  Fontainebleau  -  France

modules list                 previous module        next module

Catalog CFSG