GASA Course in Advanced Multivariate Geostatistics – November 2016

The Geostatistical Association of Southern Africa in conjunction with the University of Alberta is pleased to present Dr. Clayton Deutsch’s short course on

Advanced Multivariate Geostatistics

Monday 7 – Friday 11 November 2016

Venue in Johannesburg, Gauteng (to be confirmed)


Clayton V. Deutsch and Ryan M. Barnett

Centre for Computational Geostatistics

School of Mining and Petroleum Engineering

University of Alberta • Edmonton, Alberta CANADA

Click here for the Announcement

Click here for the Registration form


This course will cover essential theory and practice of advanced multivariate geostatistical modeling. This course builds on existing courses to cover advanced and emerging techniques for multivariate geostatistics. Selected CCG software may be used for some hands-on exercises. The morning will be reserved for lecturing. The afternoons will be a combination of lectures, hands-on exercises and demonstrations.


Dr. Deutsch is Director and Professor in the School of Mining and Petroleum Engineering, Department of Civil & Environmental Engineering at the University of Alberta. He teaches and conducts research into better ways to model heterogeneity and uncertainty. Prior to joining the University of Alberta, Dr. Deutsch was an Associate Professor (Research) in the Department of Petroleum Engineering at Stanford University. His employment history also includes three years with Exxon Production Research Company and three years of experience with Placer Dome Inc. Dr. Deutsch has published six books and over 200 research papers. Dr. Deutsch holds the Alberta Chamber of Resources Industry Chair in Mining Engineering and the Canada Research Chair in Natural Resources Uncertainty Characterization.


Day One

  01 – Introduction and Overview

  02 – Multiple Variables and Transformations

03 – Multivariate Distributions Including the Gaussian

04 – Linear Model of Coregionalization and Cokriging

Day Two

 05 – Cokriging with Unequally Sampled Data

06 – Colocated Cokriging and Variants

 07 – Simulation with Colocated Cokriging and Variable Aggregation

 08 – Principal Component Analysis

Day Three

 09 – Sphering and Minimum/Maximum Autocorrelation Factors

 10 – Projection Pursuit Multivariate Transformation

 11 – Gaussian Mixture Models and Stepwise Conditional Transform

12 – Other Transforms and Guide to Modelling Technique Selection

Day Four

13 – Multivariate Data Imputation

 14 – Clustering and Classification

 15 – Response Surface Fitting and Sensitivity Analysis

 16 – Direct Multivariate Estimation and Kernel Density Estimation

Day Five

 17 – Issues of Scale and Averaging; Geometallurgy Case Study

18 – Trends, Universal Kriging and Kriging with External Drift

 19 – Post Processing and Checking

 20 – Special Topics from CCG

This schedule will be modified to meet the objectives and pace of the instructor/participants

*Copyright, 2016, Centre for Computational Geostatistics   All rights reserved.


The cost for the week’s course is R18 000-00 per delegate.

Please complete the registration form and return to

For any queries please contact the Course Secretariat.

Course Secretariat

Yolandé van den Berg


Cell: 082 323 3910

Fax: 086 260 7024



Geostatistics constitutes a globally accepted technical approach to mineral resource-reserve estimation and the basic
toolkit for mine evaluation practitioners.  The conference to be held in Johannesburg, South Africa, 19–20 August 2015 is  sponsored by  GASA and the SAIMM.


The conference provides authors who have recently published papers in the SAIMM’s Danie Krige Commemorative Volumes, a platform to present their research. In addition an invitation to geostatisticians, resource estimation practitioners, and those with an interest in geostatistics to present new papers for inclusion in the proceedings
is now open.


The theme of the conference is ‘Geostatistical Geovalue—Rewards and Returns for Spatial Modelling’, a theme which emphasizes the improvement in, or addition to, value that spatial modelling can bring to the process of mine evaluation and mineral resource and reserve estimation. Spatial modelling of earth related data to estimate or enhance attributed value is the principle domain of geostatistics, the broad content of the Danie Krige Commemorative Volumes, and the focus of this conference.

This is the only conference of its kind since 2004!

Additional information on topics can be found in the Announcement following the link below.

Danie Krige_Call_for_Papers

Call for Papers

This is an urgent call for papers.  All South African practitioners are encouraged to step up to the occasion and support this prestigious and memorable event. Note that this event is open to international authors as well.

Prospective authors are invited to submit titles and abstracts of their papers, in English. The abstracts should be no longer than 500 words and should be submitted to:

Yolanda Ramokgadi
Conference Co-ordinator, SAIMM
P.O. Box 61127, Marshalltown 2107, South Africa.
Telephonic enquiries may be made at
Tel: +27 (11) 834-1273/7
Facsimile +27 (11) 838-5923 or