Standard Article

Conditional Simulation

Statistical and Numerical Computing

  1. John M. Shafer

Published Online: 15 SEP 2006

DOI: 10.1002/9780470057339.vac045

Encyclopedia of Environmetrics

Encyclopedia of Environmetrics

How to Cite

Shafer, J. M. 2006. Conditional Simulation. Encyclopedia of Environmetrics. 1.

Author Information

  1. University of South Carolina, SC, USA

Publication History

  1. Published Online: 15 SEP 2006


Conditional simulation, also referred to as stochastic imaging, is a stochastic simulation process whereby equally probable realizations of a spatial random function are created with each realization honoring discrete data values at specific locations or geometric patterns, i.e. ‘conditioned’ on the data. Conditional simulations of a random function satisfy two exactitude constraints, the local exactitude of honoring data values and the global exactitude of reproduction of a specified spatial correlation structure. The random variable may be categorical such as an indicator of the presence or absence of some feature (e.g. rock type), or spatially continuous such as porosity or ore grade. Conditional simulation is often used to address problems associated with the spatial distribution of physical properties relevant to the mining industry and several fields of earth sciences, especially hydrogeology, contaminant hydrology, meteorology, and gravimetry. Through Monte Carlo analysis, for example, conditional simulation can be used to evaluate the uncertainty in numerical model parameter values and the translated uncertainty in model response.