• Home
  • News
  • Evaluating and communicating models and model performance

Evaluating and communicating models and model performance

The first of the QWMN Capacity Building Small Grant projects will take the form of 3 interactive online workshops run over October and November that will open for registrations shortly. These workshops will equip participants with an improved ability to evaluate and communicate model performance and output, including quantifying uncertainty and an appreciation on how to incorporate these insights in new modelling projects.

The ‘Building capacity in evaluating and communicating models and model performance’ project is being run by Kate O’Brien (UQ) and Matthew Adams (QUT). Registrations for the 3 workshops will open shortly via Eventbrite. For now please note the dates in your diaries:

28th October – Workshop 1: Evaluating model performance: an introduction and overview for water modellers

How good is your model? This question is one of the most difficult but important questions for water modellers to answer. This LiveStream event introduces a systematic framework for assessing model performance, illustrated with examples from the water modelling sector. The LiveStream event will stimulate conversation and discuss strategies to tackle the major challenges for assessing model performance in professional water modelling projects, and resources will be supplied to participants, to support future model performance evaluation efforts after the event.

Key presenters at this event include Kate O’Brien (Associate Professor in Chemical and Environmental Engineering, UQ), Dr. Barbara Robson (Principal Research Scientist at AIMS) and James Weidmann (Senior Water Engineer at Water Technology Pty Ltd.).

11th November – Workshop 2: Uncertainty is the only certainty: Strategies to tackle this dual technical and communication challenge

After evaluating and critiquing model performance in the first workshop, water sector professionals will be introduced in this second workshop to a variety of methods for quantifying and communicating uncertainty in model predictions. The workshop will conclude by giving participants to opportunity to consider how evaluation of model predictions under uncertainty can be integrated into the modelling process for usage in present and future water modelling projects. Resources will be supplied to participants, to support future model uncertainty evaluation efforts after the event.

Key presenters at this event include Matthew Adams (Lecturer in the School of Mathematical Sciences, QUT), Maria Vilas (Senior Scientist in the Department of Natural Resources, Mine and Energy, Queensland Government) and Prof. Holger Maier (University of Adelaide).

26th November – Workshop 3: Communicating model performance for decision making 

Models are developed and applied to support decision-making. The problem is, all models are wrong, and only some are useful. How do you clearly communicate what your model can and can’t do for decision-makers? In this LiveStream event, our panellists use their experience in the water sector to discuss how to reduce the risk of having a good model dismissed as useless, a bad model hailed as a success, or a useful model applied in the wrong ways.

This panel discussion will follow a similar format to ABC’s “Q&A”, with panellists including Kate O’Brien (Associate Professor in Chemical and Environmental Engineering UQ), Tony Weber (Alluvium Consulting), David Hamilton (Deputy Director of the Australian Rivers Institute), Mark Baird (CSIRO Coastal Biogeochemical Modelling team lead), Angela Dean (QUT), Jo Burton (DES) and Paul Maxwell (Alluvium Consulting).

Latest news

Meet the Team – the QWMN Capability, Collaboration and Capacity Program

Read more

Review and update of the QWMN Research, Development and Innovation (RD&I) Strategy

Read more

Online Event Report: Modelling and communicating the impacts of climate change

Read more

Sign up for our newsletter

Please enter your first name

Please enter your last name

Please enter a valid email address