Hutton property modelling whitepaper
Geostatistical models comprise two elements: a geological prior (trend model) and a geostatistical residual. The prior is developed by interrogating the observed data for evidence of non-stationary effects (trends), defining these with mathematical functions, and subtracting them from the observed data...
First Break - October 2019 issue
The decision-oriented world: effective management of uncertainty in geomodelling workflows
Lucy MacGregor, Michael Stewart, Keegan Benallack and Luke Johnson present a machine learning approach to identify commercially significant models within a large ensemble and characterise the uncertainty in decisions based on these.
The 5th International Workshop on Data Reduction for Big Scientific Data (DRBSD-5)
Using machine learning to reduce ensembles of geological models for oil and gas exploration
...the result of this work is a series of lessons learnt and techniques that are not only applicable to oil and gas exploration, but also more generally to the HPC community as we as a community are forced to work with reduced data-sets due to the gathering of data growing so rapidly.