Geostatistics History

Geostatistics, initiated by Georges Matheron at the Mining Bureau of Algeria then at BRGM and at CEA, created the Center for Geostatistics and Mathematical Morphology at the École des Mines in Paris in 1967. Its strong development led it to split into two entities: the Center of Geostatistics and the Center of Mathematical Morphology .

From the begining the leading idea was to develop the theoretical and methodological elements making it possible to meet the needs of the industry. If initially the applications were devoted exclusively to the mine, they quickly widened to petroleum and other fields like meteorology, nautical cartography, fishery, air, water and soil pollution, etc.

From a theoretical point of view, until around 1985, Geostatistics was mainly interested in Gaussian random functions (or linear methods) which are placed in a much broader framework, but are known to have good properties in the Gaussian case. The work then turned to non-Gaussian random functions, in particular for the conditional simulation of random sets, to represent for example lithological facies.

For the past twenty years, work was mainly concerned in coupling with other approaches as well as application to new fields such as:

• Geostatistics and Data assimilation
• integration of digital simulations in the study of exposure to noise or radio waves
• application to pulluted soils, air quality or water quality (with the definition of random functions on graphs)
• stochastic genetic Models
• Extremal values in a spatial framework
• simulation of spatio-temporal processes on a sphere
• machine learning in spatial framework.

Some of our recent developments are presented on a dedicated page: New methodologies in Geostatistics