M-GS

M-GS (Moving GeoStatistics) is a technology dedicated to the local optimization of geostatistical models parameters. It enables the generation of a set of optimized parameters, called M-parameters, that are directly usable in interpolation processes and spatial simulations (1D, 2D, 3D…) in mono- or multi-variable contexts.

The principle

Local optimization of geostatistical parameters : the M-Parameters

Dealing locally with varying structures Variographic models are the geostatistical models most often used in industry. These models involve a certain number of parameters (structural parameters and calculation parameters).

M-GS (Moving GeoStatistics) collects a set of local and automated processes for determining variographic model parameters. The optimized parameters obtained are called M-Parameters.

M-GS tools

Determining and using M-Parameters

  • Optimisation through local cross-validation (ESTIMAGES patent)
  • Morphological approaches (developed with Mines Paris Tech)
M-parameters determination
Structural characterization using ray tracing approach

Morphological approach developed by ESTIMAGES



M-GS Kriging and Simulations

  • Kriging – simple and ordinary kriging, kriging with external drift, co-kriging, factorial kriging
  • Simulations : SGS, SIS
TECHNICAL PAPERS

Upstream petroleum

• SBGF 2009

M-Factorial Kriging for seismic data noise attenuation

• IAMG 2009

Noise reduction by M-Factorial Kriging

• EAGE 2008

M-GS - New spatial estimation and simulation models for more precision and more realism


Precision agriculture

• EMS 2009 - A better taking into account of the climatic risk with new agrometeorological services and applications

Oceanography

• GEOENV 2009 - Integrating prior knowledge and locally varying parameters with Moving-GeoStatistics : methodology and application to bathymetric mapping

Meteorolgy

• Journées de Géostatistiques 2011 - Moving Geostatistics for automated interpolation - Application to environmental data