Decision Support Modelling Notebooks

The hardest part of using programs of the PEST and PEST++ suites is preparation of their input datasets. Software assistance is required. The PyEMU suite provides this assistance.

PyEMU provides many services to simulator-based decision support. Among other things, it can define and manipulate spatial model parameters, and sample complex parameter probability distributions to hydrogeologically generate realistic stochastic parameter fields. It can also automate processing of many types of field measurements that are commonly used for model history-matching, together with complementary model outputs. It can then write a complete PEST++ input dataset based on all of these measurements.

PyEMU can also be used to visualize simulator outputs, as well as the outcomes of history-matching and uncertainty analysis such as graphs of model-to-measurement misfit, and histograms of parameter and predictive uncertainty.

As the name implies, PyEMU is written in Python.

A challenge to adopting PyEMU for PEST++ preprocessing is that the learning curve can get a little steep – especially if you are new to Python. Nevertheless, it is worth the learning effort. The nice thing about using Python for PEST++ setup is that you can easily add your own code to the processing mix in order to tailor your decision-support modelling approach to the demands of your particular site.

To help you reduce the steepness of the learning curve, GMDSI has developed a suite of Jupyter notebooks. You can download these from their GitHub repository.

A comprehensive set of GMDSI videos complements these notebooks. These are recordings of webinars that were delivered in the second half of 2023. Download them from the GMDSI YouTube channel.

Many thanks to Jeremy White, Brioch Henning, Rui Hugman, Mike Fienen, Katie Markovich and Cecile Coulon for delivering these webinars.