Contribute Media
A thank you to everyone who makes this possible: Read More

Developing Scientific Simulation Systems

Description

Scientific simulation models often need to incorporate many aspects into a working software system. This talk gives an overview of PITLAKQ, a complex, coupled model for water quality predictions of pit lakes. The focus is on experiences gained that might be useful for similar projects from other domains. This includes wrapping of legacy code, creating flexible but simple to use model input approaches and simplifying automation by users. Description

The model PITLAKQ helps to predict the water quality of pit lakes. These lakes can form in voids created by mining and may pose considerable environmental challenges. PITLAKQ couples several existing numerical models and adds lots of new functionality to account for the complex interactions of hydrodynamics, geochemical processes and anthropogenic influences.

Obviously, there is lots of scientific details. The talk only gives a very short overview of the functionality of the software. The main focus is on general principles that can be applied to other domains of scientific modeling. This includes:

  • wrapping legacy Fortran code
  • coupling models from different domains
  • creating a simple, yet flexible input system for model users
  • simplifying automation of simulation runs

I think the lessons I learned over the last 15 years I've been working on this project can be useful for other scientists who need to solve simulation and modeling problems in other fields.

Details

Improve this page